Crossroadsxrds2017winter DL

XRDS
Crossroads The ACM Magazine for Students W IN T ER 2017 V OL .24 • NO.2 XRDS.ACM.ORG
Human
to Human
Digital and Physical Barriers
to Changing Identities
Identifying Hate Speech
in Social Media
Gender and the Art
of Community Relations
INSPIRING MINDS
FOR 200 YEARS
Ada’s Legacy illustrates the depth
and diversity of writers, things, and
makers who have been inspired
by Ada Lovelace, the English
mathematician and writer.
The volume commemorates the
bicentennial of Ada’s birth in
December 1815, celebrating her
many achievements as well as
the impact of her work which
reverberated widely since the late
19th century. This is a unique
contribution to a resurgence in
Lovelace scholarship, thanks to the
expanding influence of women in
science, technology, engineering and
mathematics.
ACM Books is a new series of high quality books for the computer science community, published by
the Association for Computing Machinery with Morgan & Claypool Publishers.
Inviting Young
Scientists
Meet Great Minds in Computer
Science and Mathematics
As one of the founding organizations of the Heidelberg Laureate Forum
http://www.heidelberg-laureate-forum.org/, ACM invites young computer
science and mathematics researchers to meet some of the preeminent scientists
in their field. These may be the very pioneering researchers who sparked your
passion for research in computer science and/or mathematics.
These laureates include recipients of the ACM A.M. Turing Award, the Abel Prize,
the Fields Medal, and the Nevanlinna Prize.
The Heidelberg Laureate Forum is September 23–28, 2018 in Heidelberg, Germany.
This week-long event features presentations, workshops, panel discussions, and
social events focusing on scientific inspiration and exchange among laureates
and young scientists.
Who can participate?

New and recent Ph.Ds, doctoral candidates, other graduate students
pursuing research, and undergraduate students with solid research
experience and a commitment to computing research
How to apply:
Online: https://application.heidelberg-laureate-forum.org/
Materials to complete applications are listed on the site.
What is the schedule?
The application process is open between November 6, 2017
and February 9, 2018.
We reserve the right to close the application website
early depending on the volume
Successful applicants will be notified by mid April 2018.
More information available on Heidelberg social media
PHOTOS: ©HLFF / B. Kreutzer (top);

©HLFF / C. Flemming (bottom)
Crossroads
The ACM Magazine for Students
W IN T E R 2 017 V OL . 2 4 • NO . 2
14
begin
5 LETTER FROM THE EDITORS
Multidisciplinary Systems
Engineering
By Jennifer Jacobs
7 INIT
Human to Human
By Alex Ahmed and Teresa Almeida
10 ADVICE
The Excessive Power of
Ctrl+C and Ctrl+V in CS Research
and Career Development
By Nisha Panwar and Shantanu Sharma
12 UPDATES
ADA ACM’s Role in
Diversifying Chapter Activities
By Anshuman Majumdar
14 CAREERS
Software is for Humans
By Jenn Schiffer
15 MILESTONES
Technology and Dating
By Alok Pandey
Image by Redpixel PL / Shutterstock.com
16 ACCOLADES
ElectroEuro
By Talia Kohen
18 BLOGS
Interfacing an FPGA with an
External Circuit and Applications
By Alexander DeForge
2 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

HUMAN TO HUMAN
34 48 62
features end
22 FEATURE 38 FEATURE 54 LABZ
Can We Build the Cyborg Future Gender and the Art Social and Psychological Questions
We All Deserve? of Community Relations about Humans and Technology
By Ari Schlesinger By Lesley Mitchell By David M. Markowitz
26 FEATURE 44 FEATURE 55 BACK

Digital and Physical Barriers Reflecting on Robots, Equality through
to Changing Identities Love, and Poetry Digital Technology
By Oliver L. Haimson By Margaret Rhee By Vasileios Kalantzis
30 FEATURE 48 FEATURE 56 HELLO WORLD

Disability-Disclosure “We Had Tough Times, Identifying Hate Speech
Preferences and Practices in But We’ve Sort Of Sewn Our Way in Social Media
Online Dating Communities Through It:” The Partnership Quilt By Alexandra Schofield
By Cynthia L. Bennett By Angelika Strohmayer and Thomas Davidson
and Janis Meissner
34 FEATURE 60 POINTERS
Leveraging Personal Experience for 52 PROFILE
Academic Research and Outreach Sara Mauskopf 60 JARGON
By Joslenne Peña Innovative Thinking in
the Service of Parents 62 EVENTS
Left image by Xi Xin Xing; Right image by Nattee Chalermtiragool
By Adrian Scoică
64 BEMUSEMENT
On the Cover:
Image by Alicia Kubista/Andrij Borys Associates.
XRDS • W IN T ER 2 017 • V OL .24 • NO.2 3

E DI T ORI A L B O A RD Digital Content Editor A D V IS OR Y B O A RD SUB S C RIBE
Pedro Lopes
Editors-in-Chief Hasso Plattner Institut, Mark Allman, Subscriptions ($19
Jennifer Jacobs Germany International Computer per year includes XRDS
MIT, USA Science Institute, USA electronic subscription)
are available
Okke Schrijvers News Editors Bernard Chazelle, by becoming an
Stanford University, USA Ankush Patak Princeton University, USA ACM Student Member
Pimpri Chinchwad College www.acm.org/
of Engineering, India Laurie Faith Cranor,
Departments Chief Carnegie Mellon membership/student
Adrian Scoică University, USA Non-member
University of Cambridge, Bloggers
David Guerra Alan Dix, subscriptions:
UK Lancaster University, UK $80 per year
Universitat de Lleida,
Guest Editor Spain http://store.acm.org/
David Harel, acmstore
Alex Ahmed Weizmann Institute
Northeastern University, Abdelrahman Hosny
University of Connecticut, of Science, Israel ACM Member Services
USA To renew your ACM
USA Panagiotis Takis Metaxas ,
Teresa Almeida membership or XRDS
Wellesley College, USA subscription, please send
Advertise with ACM! Newcastle University, UK
Issue Feature Editor

Andrew J. Hunsucker
Indiana University, USA Noam Nisan, Hebrew
University Jerusalem,
a letter with your name,
address, member number
Vasileios Kalantzis Israel and payment to:
Judeth Oden Choi
Carnegie Mellon University of Minnesota, ACM General Post Office
Bill Stevenson ,
Reach the innovators University, USA USA
Apple, Inc., USA. P.O. Box 30777
New York, NY
Vassilios Karakoidas
Andrew Tuson, 10087-0777 USA
and thought leaders Feature Editors
Athens University of
Economics and Business, City University London, UK
Postal Information
Eliezer Bernart Greece Jeffrey D. Ullman, XRDS (ISSN# 1528-4981)
working at the Federal University of Rio
Grande do Sul, Brazil Maria Kechagia
Stanford University, USA is published quarterly in
spring, winter, summer
Athens University of Moshe Y. Vardi,
cutting edge Henri Maxime Demoulin Economics and Business,
Greece
Rice University, USA
and fall by Association for
Computing Machinery,
University of Pennsylvania, 2 Penn Plaza, Suite 701,
of computing USA Norene Kelly
Iowa State University, USA
E DI T ORI A L S TA F F New York, NY 10121.
Application to mail at
Ophir Friedler Director of Publications
and information Tel Aviv University, Israel Pedro Lopes Scott E. Delman
Periodical Postage rates
is paid at New York, NY
Hasso Plattner Institut, XRDS Managing Editor and additional mailing
technology through Gunnar E. Wolf Iszaevich
National Autonomous
Germany and Senior Editor at ACM
HQ Denise Doig
offices.
University of Mexico, Dimitris Mitropoulos
ACM’s magazines, Mexico Athens University of
Economics and Business,
Art Direction
Andrij Borys Associates,
POSTMASTER: Send
addresses change to:
Numair Khan XRDS: Crossroads,
websites Brown University, USA
Greece
Wolfgang Richter
Andrij Borys,
Margaret Gray,
Association for
Computing Machinery,
Mia Balaquiot 2 Penn Plaza, Suite 701,
and newsletters. Talia Kohen
Bar Ilan University, Israel
Carnegie Mellon
University, USA Advertising Sales New York, NY 10121.
Account Manager Offering# XRDS0171
Nidhi Rastogi Abhineet Saxena Ilia Rodriguez
Rensselaer Polytechnic Guru Gobind Singh ISSN# 1528-4972 (print)
ilia.rodriguez@acm.org ISSN# 1528-4980
Institute, USA Indraprastha University,
India Copyright Permissions (electronic)
◊◆◊◆◊ Yang Shen
Gustavo Fortes Tondello
Deborah Cotton
University of California, permissions@acm.org Copyright ©2017 by
Los Angeles, USA University of Waterloo, the Association for
Canada Editorial Assistant
Computing Machinery, Inc.
Request a media kit Department Editors
Jade Morris
Permission to make digital
or hard copies of part of
Shefali Bhatnagar ACM this work for personal or
with specifications Amity University, India
Association for classroom use is granted
without fee provided that
Abhishek Bhattacharya Computing Machinery
and pricing: Amity University, India 2 Penn Plaza,
Suite 701
copies are not made or
distributed for profit or
New York, NY commercial advantage
Andrew J. Hunsucker 10121-0701 USA and that copies bear
Ilia Rodriguez Indiana University, USA +1 212-869-7440 this notice and the full
citation on the first page
Vasileios Kalantzis C ON TA C T or initial screen of the
+1 212-626-0686 University of Minnesota,
USA
General feedback:
xrds@acm.org
document. Copyrights for
components of this work
owned by others than
acmmediasales@acm.org Bryan Knowles
Western Kentucky
For submission
guidelines, please see
ACM must be honored.
Abstracting with credit
University, USA http://xrds.acm.org/ is permitted. To copy
authorguidelines.cfm otherwise, republish, post
Anshuman Majumdar on servers, or redistribute
International Institute of requires prior specific
Information Technology, PUBLIC AT IONS BOA RD
permission and a fee.
India Co-Chairs Permissions requests:
Jack Davidson and permissions@acm.org.
Darshit Patel Joseph A. Konstan
Pimpri Chinchwad College
of Engineering, India Board Members
Phoebe Ayers;
Alok Pandey Anne Condon; Nikil Dutt;
University of Petroleum Roch Guerrin; Chris Hankin;
and Energy Studies, India Carol Hutchins;
Yannis Ioannidis; Sue Moon;
Johanna Schacht Michael L. Nelson;
Karlsruhe Institute of Sharon Oviatt;
Technology, Germany Eugene H. Spafford;
Stephen N. Spencer;
Alex Wade; Keith Webster;
Julie R. Williamson

LETTER FROM THE EDITORS
Multidisciplinary
Systems
Engineering
M
odern technological development emphasizes innovation. The success of an
algorithm, a device, or a piece of software is often directly equated with the amount
of impact it has on our world. The value our society places on innovation has led to a
great deal of research aimed at understanding specific factors involved in creating
powerful and meaningful technology. My dissertation research addressed a subdomain of
this topic by examining ways in which we might design computational tools for art and
design. My approach was to involve artists and designers directly in the development process
by creating programming models and I believe this question reflects the than men [2]. Following its release, the
tools specifically for them. Curiously, view of some engineers; the idea that memo was criticized by many in the
when this research was presented to programming is a highly specialized technology community. Unfortunate-
software developers, I was routinely domain where only a select few have ly, its messages were tacitly endorsed
asked a version of this question: “Do the skills and knowledge required to by other prominent technologists [2],
you really believe artists and designers create systems that benefit a wider and Damore’s firing from Google as a
want to program?” population. Clearly, not all program- result of his assertion was used as evi-
mers have this perspective. However, dence by some that the push for diver-
there are others in the tech commu- sity in technology had gone too far [3].
UPCOMING ISSUES
nity who believe efforts to increase It’s true individuals often exhibit
diversity in professional engineering different ways of thinking and possess
Spring 2018
lowers the bar for engineering hiring various aptitudes. It can feel like com-
[March issue] practices and hinder technological in- mon sense, therefore, to conclude tech-
Computers and Art novation [1]. A memo written by former nological innovation is best supported
Google employee James Damore, made through a merit-based technology
Summer 2018 public in August 2017, argued diversity industry with hiring practices aimed
[June issue] efforts at Google were counterproduc- at selecting people with the greatest
Online Personality, Pseudonymity, tive. Damore claimed, on average, “natural talent” for programming. Yet
women are biologically unsuited for closer examination of this argument
and Anonymity
the challenges of computer program- reveals two major problems.
Article deadline: March 1, 2018 ming because they have higher rates of First, portraying any industry as
anxiety and lower tolerance for stress a meritocracy often leads people to

incorrectly attribute success to natural result of a mismatch between the af- Some existing college-level comput-
predispositions, evolutionary trends, fordances of a particular algorithm er science programs provide opportu-
or intrinsic ability. It also ignores so- and the needs, values, and desires of nities for students to learn skills and
cial and cultural issues that provide actual people using them. Creating approaches from disciplines besides
advantages to a privileged group, while software that effectively represents engineering by taking courses in de-
significantly hindering others. The the complexity of our world and ef- sign, art, and social science. However,
Google memo is one example of this fectively serves the diverse needs of computational innovation is also de-
attitude, but this trend is also reflected its inhabitants necessitates diverse pro- pendent on improving pathways for
in the historical precedent of gender grammers with a broad range of skills both students and professionals out-
and racial discrimination in art, lit- and expertise. side of computer science to learn and
erature, science, and other prominent Throughout my doctoral research apply programming. My own research
professions. Take for example the ar- program, I’ve been fortunate to have demonstrated how, when adequately
gument by Christopher Hitchens that the opportunity to work with different supported, artists are capable of de-
women are ineffective comedians kinds of programmers, ranging from veloping new creative computational
because they lack the innate drive to professionals to hobbyists to children. tools that provide opportunities not
impress the opposite sex [4]. I also have worked with artists, scien- offered by existing software. I suspect
Second, the meritocracy argument tists, designers, and educators. This journalists, educators, designers, com-
often presents an oversimplified, inac- experience has led to my conviction munity organizers, and other profes-
curate view of what makes an industry that traditional engineering skills, sionals are similarly well-positioned
or field successful. When I say we should while important, are only one aspect to contribute innovative tools for their
seriously view artists and others who of building innovative, effective, and own fields.
lack formal computer science education interesting computational systems. It is incredibly challenging to move
seriously as programmers, the larger An understanding of the principles from practice in a domain to the work of
point is computer science itself suffers and process of interaction design en- creating systems for that domain. This
from its portrayal as a one-dimensional ables the creation of systems that are process often is made more challenging
domain. Programming frequently is not merely usable, but pleasurable, by computer science education and hir-
perceived as an anti-social domain re- engaging, and compelling. Knowledge ing practices, which can limit participa-
quiring skills in math and reasoning. of interaction design is also relevant tion to only those who follow a few high-
Damore perpetuated this attitude in to systems engineers who collaborate ly structured, traditional pathways. Not
his memo by arguing women are funda- with designers. It can help them un- everyone needs to create their own soft-
mentally less relevant to programming derstand why a designer may request ware. However, if we truly care about
than men, because they are more inter- features that seem incidental to the technological innovation, then we must
ested in people and aesthetics. Putting functionality of the system and are la- also be open to the contributions of
aside the problem with stereotyping borious to implement. people outside traditional computer
women as uninterested in formal and Teaching experience is also valu- science, and we must be innovative in
analytic fields, in reality, programming able. Engineers who have an under- the ways we support new practitioners
encompasses a plurality of skills, inter- standing of theories of learning or in systems engineering.
ests, and ways of thinking. have taught in a classroom can better
In their essay on epistemological recognize learning challenges, which —Jennifer
— Jacobs
pluralism, Sherry Turkle and Seymour in turn, helps them develop ways of
Papert described how the computer it- scaffolding learning. Teaching can
self operates “betwixt and between the provide engineers with empathy for References
world of formal systems and physical people who use their tools. It can dis- [1] Bowles, N. Push for gender equality in tech? Some
men say it’s gone too far. The New York Times.
things” by literally translating abstract pel unconstructive attitudes toward September 23, 2017; https://www.nytimes.
representations to concrete systems learners by demonstrating that new com/2017/09/23/technology/silicon-valley-men-
backlash-gender-scandals.html.
and artifacts that live in the world and creators, while inexperienced, posses [2] Conger, K. Exclusive: Here’s the full 10-page anti-
interact with people [5]. Turkle and strong design objectives, stylistic pref- diversity screed circulating internally at Google
[Updated]. Gizmodo. August 5. 2017; https://gizmodo.
Papert’s argument is borne out of the erences, and personal values. com/exclusive-heres-the-full-10-page-anti-
affect computing has on our world to- The capacity to investigate the at- diversity-screed-1797564320
day. Social-media platforms, gaming titudes and values of different com- [3] Kovach, S. Silicon Valley’s liberal bubble has burst, and
the culture war has arrived. Business Insider. 2017;
technologies, creative software, and munities and individuals is also an http://www.businessinsider.com/silicon-valleys-
countless other forms of software are important skill. Specifically, experi- liberal-bubble-has-burst-2017-8
as much defined by their algorithmic ence with methods of social science [4] Hitchens, C. Why women aren’t funny. Vanity Fair
(Jan. 2007); https://www.vanityfair.com/
qualities as they are by their interfac- research can help engineers under- culture/2007/01/hitchens200701
es, interactions, and the communities stand general values, practices, and [5] Turkle, S. and Papert, S. Epistemological pluralism
that make use of them. Similarly when attitudes of communities that use and the revaluation of the concrete. Journal of
Mathematical Behavior 11, 1 (March 1992).
these systems produce unexpected, their systems. Face-to-face interviews
undesirable, and sometimes harm- can provide invaluable insights for DOI: 10.1145/3155214
ful consequences, it is frequently the improvement of a system. Copyright held by author.

begin
INIT
Human to Human
T
here’s a myth in tech oppressions operate at all almost invariably linked gies—who are largely white,
culture: A myth that levels of our society, empow- to specific ways of organiz- male, and able-bodied—-are
technology can solve ering some individuals over ing power and authority” lauded as benefactors who
any problem, bridge others. Technology can be a [1]. As an example, Winner are making the world a bet-
any gap, and create some- vehicle to perpetuate these writes the benefits—and ter place.
thing out of nothing. And, harms, as well as a tool for re- consequences—that nuclear Such a typecasting de-
supposedly, it has no politi- sistance. It can create oppor- energy brings to humanity nies true self-determination
cal allegiance or agenda. As tunities as well as deny them. will depend not only on the of disabled people because
students, it’s difficult to re- It is undeniably political. configuration of the hard- it contributes to a narrative
sist this allure—that we, as Designing systems “for ware itself, but also on how that denies personhood (in
future programmers and good”—such as those that the technology is situated favor of “problemhood”).
technologists, are going to support and mediate en- in social institutions. Nearly This prevents a more nu-
“make the world a better gagement between hu- 40 years after Winner made anced exploration of the
place.” HBO’s series “Silicon mans—supports positive that observation, our tech- ways our constructed social
Valley” perfectly satirized1 change, but only when such nologies have grown more environment can itself be
this false promise, as ficti- systems account for cultural complex and sprawling, yet disabling, oppressive, and
tious TechCrunch Disrupt values. Taking a humanistic simultaneously more per- exclusionary. It also prevents
presenters proudly claimed approach within technology- sonal and intimate. us from discussing how
they were “making the world led interventions is challeng- In her book Giving Voice, people’s access to assistive
a better place through scal- ing, but essential. As devel- Meryl Alper engages with technologies, or the training
able, fault tolerant distrib- opers, we cannot ignore how similar ideas in the context and resources to use them ef-
uted databases.” The show our designs and products of iPad apps for children who fectively, are distributed dif-
revealed not just how ridicu- impact human lives. cannot produce oral speech ferently based on race, class,
lous and hilarious tech com- In 1980, Langdon Win- (or have significant difficul- nationality, and other fac-
panies’ claims of social good ner wrote, “if we examine ty doing so) [2]. These apps, tors. If only privileged fami-
can be, but also how we can social patterns that com- which present a grid of picto- lies can use technology to
be complicit in them. prise the environments of grams that can be selected in guarantee better outcomes
The ways we relate to technical systems, we find order to create and “speak” for their children, is that
each other and to ourselves certain devices and systems sentences, can function as truly “social good”? Or is it
are constrained by our en- a child’s primary mode of oppression, in which power
vironments—our peers, our communication. Despite remains entrenched in the
culture, and artifacts in the appearing to be wholly ben- dominant strata of our soci-
world. Because of this, it’s eficial to users, such technol- ety? (In this issue, Joslenne
indisputable that technol- Being pioneers ogies (and the culture sur- Pena explores this idea in the
ogy impacts intimacy, iden-
tity, and relationships. This in our field rounding them) can act to
reproduce social inequalities
context of technology use in
K-12 schools.)
issue explores that deeply, requires us and power dynamics that op- Broadly, our issue ex-
with the understanding that
these impacts are deter- to ask difficult press disabled people. Alper
argues that news media of-
plores questions like: Of the
opportunities digital tech-
mined by patterns of power questions about ten portrays disabled people nology presents for support-
and dominance. In other
words, systemic racism, sex-
who benefits in terms of what they are per-
ceived to lack: as problems
ing positive expression, what
groups of people are being
ism, homophobia, trans- from our work waiting for a technological afforded them? Who is be-
phobia, ableism, and other
and who is solution. Simultaneously,
the engineers and design-
ing ignored, and who is being
negatively impacted? Whose
1 https://youtu.be/IXuFrtmOYKg ignored. ers behind the technolo- self-determination is con-

begin
73% of Stanford faculty were men, and almost

90% of staff declared either white or Asian
to be their ethnicity as of September 1, 2016.
INIT
sidered taboo, undesirable,

If only unnatural, or even criminal?
These problems exist, and
privileged our research projects and our
families can academic institutions aren’t
independent of them. As au-
use technology thors, we have personally en-
to guarantee countered our own challeng-
es and barriers in conducting
better our work and writing about
outcomes for our research. As guest editors
for this issue, we struggled
their children, with selecting topics and
is that truly authors that address these
issues in a meaningful way,
“social good”? while adhering to the culture
of the ACM. This manifested
itself through restrictions on
language and subject mat-
ter considered offensive to
a general computer science Alex Ahmed
audience and the barring
of human-centered topics, work blurs the lines between rather than supporting them
which speak to how pervasive the two. throughout this process.
stigma and discrimination Ari Schlesinger discusses He lays out how such plat-
truly are. how the field of human- forms can adapt to us (and
We, as computer science computer interaction (HCI), our changing identities and
students, are the future of a which examines the impact lives) rather than the other
field that thinks and dreams of technology on individuals way around.
about the future. And yet, we and communities, has fallen Cynthia Bennett presents
are often presented a limited short of providing a nuanced an overview of her research,
vision of what that future analysis of human identity. drawing from the field of
consists of. Our curricula, Given this history, where do critical disability studies,
as well as the fields of com- we go from here? on the disclosure of disabil-
puting research to which we Oliver Haimson’s article ity on online dating profiles.
are exposed, are often sani- focuses on identity transi- Disclosure, she found, allows
tized—scrubbed “clean” of tions on social media. He daters to screen out potential
the diversity of thought and argues more work is needed matches who are ableist (that
being that characterizes hu- to design and implement is, who harbor oppressive
manity, our connections to social media platforms that views of disabled people).
others, and our connections support marginalized popu- She concludes by outlining
to ourselves. We need to be lations. Transgender people design opportunities for
aware of this erasure, and its who are navigating and dis- dating sites that could coun-
roots in systemic social prob- closing their transitions on- ter inaccurate and harmful
Liz Linder Photography
lems. This issue, “Human to line might have to deal with views about the sexualities of
Human,” highlights people potentially fraught relation- disabled people.
who are doing work to solve ships with family, friends, Joslenne Pena shares her
these problems: researchers, and employers; however, the graduate research in build-
activists, and those whose system can end up impeding ing technologies for K-12 in-

Women were not allowed to graduate from
Cambridge University before 1948.
It wasn’t until 1969, when Princeton
University admitted women.
the forefront of innovation

need to account for the com-
plexity of being human, and Technology
consequently the subjectivi- can create
ties that arise from the fact.
Computing systems that aim opportunities
to improve human condi- as well
tions have the potential to
transform communities and
as deny them.
affect policy. Take the afore- It is undeniably
mentioned example of us-
ing technology in support of
political.
those is employed in the sex
industry, one of the most dis-
advantaged groups of people
in the world, and the discrete
ways in which computer-
mediated systems can have
an impact on their lived ex-
perience. We hope that this
Teresa Almeida work, in addition to the other
inspiring research in this
structors, drawing from bar- en in tech, specifically in the issue, broadens the range
riers she experienced in her Debian developers commu- of perspectives that moti-
own life. Access to technol- nity. She found microaggres- vate our future endeavors
ogy in education, she writes, sions—small yet frequent as builders and creators of
is one way that privilege and discriminatory interactions technology. Being pioneers
disadvantage are distributed that “[wear] one down as a in our field—and designing
unequally in our society. trickle of water wears down a systems for “social good”—
Margaret Rhee reflects on rock”—were common among requires us to ask difficult
the impact of artificial intel- female developers. questions about who benefits
ligence in her poetry by high- Angelika Strohmayer and from our work and who is ig-
lighting her experiences during Janis Meissner describe the nored. Ultimately, it requires
graduate school. At the time, collaborative making of the us to welcome change, as we
writing poems and building Partnership Quilt, an interac- work together to ensure that
robots become apparently in- tive living archive that makes the ever-changing landscape
terconnected, thus motivat- tangible the voice of a commu- of technology works for ev-
ing her to this day to continue nity that is often marginalized. eryone.
exploring technologies of dif- Their work with expert crafters —Alex Ahmed and
ference through science fic- and a charity providing sup- Teresa Almeida,
tion poetry. port services to society’s most Guest Editors
Lesley Mitchell leverages vulnerable—in this case, the
her past experience as a sys- sexually exploited—produced References
Lalya Gaye/Attaya Projects
admin in her work as a gradu- a colloboration of traditional [1] Winner, L. Do artifacts have politics?
Daedalus, 121-136, (1980)
ate student of sociology. In quilting techniques and novel
[2] Alper, M. Giving voice: Mobile
this issue, she discusses a digital tools that led to posi- communication, disability, and
qualitative study that used in- tive expression. inequality. MIT Press. 2017.
depth interviews to explore Technologies that im- DOI: 10.1145/3155130

the lived experiences of wom- pact human lives and are at Copyright held by author.

begin
Transgender people are four times as

likely to have a household income
under $10,000, and twice as likely to be
unemployed as the average American.
ADVICE
The Excessive Power of Ctrl+C and Ctrl+V

in CS Research and Career Development
V
arious types of plagiarism The height of verbatim plagiarism 1. Faulty tools. Many online plagia-
are common in academia. may be illustrated by the following rism detectors can only find verbatim
Namely, (i) word-by-word or real story. Jure Leskovec, Anand Ra- plagiarism, not paraphrasing plagia-
verbatim plagiarism: turning jaraman, and Jeffrey D. Ullman wrote rism. For example, Google Scholar does
contents into a carbon copy, (ii) para- a book entitled Mining of Massive not count a citation if a paper is not
phrasing plagiarism: copying the un- Datasets (2011). Since the book is properly cited in a new paper.
derlying meaning of the content in a available online, someone by the 2. Fake reviewers. In some plots, au-
“smart” manner by slightly tweaking name of Seyed Hossein Ahmadpa- thors first write a fake paper, and then
the tone of sentences, (iii) idea pla- nah simply copied the entire thing. submit fake reviewers’ information
giarism: capturing early/unpublished Seyed only changed two elements of that turns out to assign the paper for re-
ideas belonging to someone else, and the book: the title and the author’s view to the authors or to their friends.2
(iv) authorship plagiarism: masquer- name.1 Seyed used CreateSpace, Am- 3. A lot of papers and few reviewers.
ading as an original author of a differ- azon’s book-production facility, to Some top-notch journals/conferences
ent author’s work. Moreover, occasion- produce the hard copies, and made that promise to adhere to the standards
ally, reviewers may demand their own the fake book available on Amazon usually face a crisis of available review-
papers to be cited (no matter whether for $25. ers. Most authors wish to submit their
the references are relevant or not), Recently, we were assigned a pa- papers to good journals; however, the
leading to a new type of plagiarism per to review for a well-regarded lack of expert reviewers significantly
that we call “citation plagiarism.” In journal. The writing and contents impacts the review process.
this article, we mention three real pla- of the paper were extremely poor, 4. Time and effort. Some students
giarism stories, and show ways to re- which was not an unusual problem wish to complete their degrees quickly
duce incidents of plagiarism. It is not until we noticed that the authors’ without devoting significant effort.
our intention to blame or shame any affiliation was listed as MIT. It was An identical notion is also applicable
one journal, editor, reviewer, univer- unexpected to see MIT authors sub- to some faculty (usually at the begin-
sity, student, professor, or author. In- mitting a paper of that caliber, so ning of their career), who wish to pub-
stead we will give an account of three we investigated further. We discov- lish more to quickly earn a promotion.
plagiarism stories to demonstrate ered they were not actually enrolled These are the driving factors for plagia-
how insidious the problem is. at MIT, or any other U.S. university. rism in terms of manipulating data, al-
Recently, we were the victims of Since we were curious to know what gorithms, or system settings.
plagiarism. We found clear evidence went behind this blunder, we did not 5. Trust versus truth. Journals build
of paraphrasing plagiarism in a pub- reject the paper outright, but instead up the trust of researchers through im-
lished paper. It was a random online asked for a major revision. The only pact factors, which are, occasionally,
search that caught our eye. A familiar question that the authors responded false due to citation plagiarism and self-
figure was identical to one published in to was regarding their affiliation, citation (the number of citations from a
a paper from a couple of years ago. Cu- they claimed they mistakenly wrote journal article to articles published in
riosity lead us to discover 90 percent of in the wrong affiliation. the same journal). Moreover, numerous
the content of that randomly searched After reading these stories, a natural papers are submitted to journals for re-
paper was paraphrased from not only question arises in our mind: Why is pla- view. Hence, it is hard to accomplish a
our paper, but others (available online) giarism detection getting harder? There uniform review process in any journal
too. We approached the editor-in-chief might be various possible reasons: with a good impact factor.
of the journal of the newly published
paper, who found the same issue, lead- 1 See Prof. Jeffrey D. Ullman’s May 2017 2 http://retractionwatch.com/2017/04/20/
ing to a retraction of the paper, after Google+ post; https://plus.google. new-record-major-publisher-retracting-
some efforts. com/u/0/110300059908499284579 100-studies-cancer-journal-fake-peer-reviews/

In the U.S. women earned 79.6 cents for
every dollar men made in 2015. The top three
U.S. states with the lowest gender salary gap
are New York, California, and Florida.
to verifying all the past papers of the

plagiarist author in that journal. The
chances are high that a plagiarist au-
thor is a repeat offender. For example,
the author of the fake book in our sec-
ond story conducted verbatim plagia-
rism with many books, which are still
available online. The journal may take
strict actions against the accused one.
Although journals should share infor-
mation, this may open publishers up
to liability.
5. How much time? The limited time
frame is a major hurdle in achieving
an honest review. A possible solution
would be to review a paper within a
hierarchy of students of the reviewing
professor. Of course, during the whole
process, reviewers should remain polite
and complete a timely review.
Acknowledgments
So what now? Is this the end of good conferences, like SIGMOD and AIIDE, We are thankful to Prof. Jeffrey D. Ullman for providing
worthwhile feedback on this article and the full details
papers or an end of the existing ethical believe in the double-blind rule to sup- of his experience with plagiarism. We would also like to
system for the research or review com- port a fair review system. acknowledge Andy J. Hunsucker and Adrian Scoică for
their valuable feedback.
munity? Is there any solution to prevent 3. Why not publish reviews and ac-
plagiarism? In order to answer these knowledge reviewers? This could be an Biographies
questions, we need to first answer the influential step toward an honest re- Nisha Panwar is currently a post-doctoral scholar at
following questions: view process. After accepting a paper, University of California, Irvine. She did her Ph.D. within the
Department of Computer Science, Ben-Gurion University
1. What must an editor do? To pre- the journal would publish the reviews of the Negev, Israel. She received her M.Tech. in computer
vent authorship plagiarism, the edi- with the paper. In addition, rewarding engineering from the National Institute of Technology,
Kurukshetra, India in 2011. Her research interests include
tor must first verify authors and re- the best reviewers of the year/quarter/ security and privacy issues in IoT such as smart homes,
viewers (as suggested by the authors). volume would also help in improving buildings, vehicular networks, computer network and
communication, and distributed algorithms.
The editor may also investigate the review skills. Springer’s Distributed
reviews to some extent, and avoid Computing announced the first best re- Shantanu Sharma received his Ph.D. in computer
science in 2016 from Ben-Gurion University, Israel,
citation plagiarism by reviewers or viewer award in 2012.
Image by Sielan / Shutterstock.com
and M.Tech. in computer science from the National

him/herself. 4. What should be done after identify- Institute of Technology, Kurukshetra, India, in 2011.
He was awarded a gold medal for the first position in
2. Single- or double-blind review? The ing a plagiarist? A combustible ques- his M.Tech. degree. Currently, he is pursuing his post
double-blind review process can only tion is who is responsible for figuring doc at the University of California, Irvine assisted by
Prof. Sharad Mehrotra. His research interests include
help if the paper is not assigned to a out plagiarism: editors, reviewers, the designing models for MapReduce computations, data
fake identity. However, the double-blind journal team, or the authors whose security, distributed algorithms, mobile computing,
and wireless communication.
review has its own merits and faults, paper is plagiarized? In addition,
and is a whole topic worthy of dis- once a journal uncovers an author as DOI: 10.1145/3155210
cussion. Nevertheless, some leading a plagiarist, the journal may commit Copyright held by author.

begin
The E.U. member state with the lowest

gender salary gap is Slovenia, where
women earn on average 96.8% of what
their male colleagues earn.
UPDATES
ADA ACM’s Role in

Diversifying Chapter Activities
W
ith the vast amount of ap-
plications of computer
science, and its plethora
of sub-disciplines, the
discussion of diversity and student
chapters is inevitable. When it comes
to diversification of chapter activities,
the ADA ACM student chapter of ADA
University in Azerbaijan is making ex-
cellent progress. Their activities span
various sub-disciplines and applica-
tions of computer science.
Established in April 2016, the ADA
ACM team currently has 64 members
with a keen passion for computer
science and information technology.
The chapter seeks to unite students
from all over Azerbaijan studying
computer science and related fields,
and offer them workshops, courses,
excursions, and contests coordinated Prof. Jennifer Widom of Stanford University (center) with participants of the Ada’s
by experienced professionals in their Legacy ACM Celebration of Women in Computing at ADA University.
respective sectors.
Over the summer, the group or- tendees were given a presentation on at a local level in Baku, the capital city
ganized an excursion to Azerbaijan’s Azerbaijan’s satellites, AzerSpace-1 of Azerbaijan, most of their events are
main satellite ground control station, and AzerSky, followed by a site orienta- open to the public.
Azercosmos, which is the only satel- tion session to the main antennas and “Although we are called the ADA
lite control station in the Caucasus the radio signal processing stations. University ACM Student Chapter,” ex-
region. Since the satellite control sys- Although the participants in this ses- plaied Shahnur Isgandarli, chapter
tems are computer-based, the excursion were university students, most of chair, “we are not limited to the stu-
sion was especially beneficial for the ADA ACM’s events are not restricted dents of our university.” In April of
computer systems engineers. The at- to the student community. Operating this year, they held a day devoted to
programming in Python, one of the
most popular high-level, interpreted
ADA ACM at a Glance programming languages. Aptly named
School: ADA University PyDay, it was a one-day seminar that
Photo ©2017 by Shahin Abasaliyev
Chapter Name: ADA University ACM Student Chapter covered basic and intermediate level
Location: Baku, Azerbaijan Python, along with its web applica-
Facebook: https://fb.com/ada.acm.chapter tions. It was conducted by Jahangir
Date Established: April 4, 2016 Shabiyev, a software developer from
Officers: Shahnur Isgandarli (chair); Ziyaddin Sadigov (vice-chair); Sevinj Jafarli (trea- Sumaks LLC. Python being a general-
surer); Farid Ahmadov (membership chair); Dr. Araz Yusubov (faculty sponsor) purpose programming language that
Current Total Membership: 64 has modules and libraries for almost
any sub-discipline of computer sci-

XRDSSeeks Student
Volunteers
Are you a student who
enjoys staying up to
date with the latest
tech innovations, and is
ence, this session was beneficial for ogy. The presentation was followed by
looking to make an impact
anyone who wanted a head start in Py- a three-day workshop series in data on the ACM community?
thon programming. There were about analysis and big data. Jointly funded
30 attendees, including students and by the Azerbaijan ACM chapter, the Editorial positions
industry professionals. workshop, named “Big Data Week- are now open.
As a part of their community initia- end,” was attended by more than 80
tives, ADA ACM is also big on promot- eager computer science students and For more information
ing the involvement of women in com- professionals. The participants were and to apply visit:
puting. They held a conference called, taught about techniques for data col- http://xrds.acm.org/
“Ada’s Legacy: Women in Informatics lection and analysis, and were given
2017” on April 15th, which is an annual hands-on experience implementing volunteer.cfm
celebration of women in computing. data analysis tools using Python and
This was their second year hosting this related packages. Association for
Computing Machinery
event, and they partnered with other In an effort to promote competitive
ACM and ACM-W chapters in Azerbai- programming, ADA ACM, along with
S
jan. The conference brought together the Azerbaijan ACM chapter, helped XRD
S. AC
M.O
RG
D
more than a 100 students, researchers, organize a Google Hash Code team. 0 16
VO
L .2
2•
NO
.3
G2
and professionals from industry as well. This is a programming competition
R
R IN
SP
ents
Having a surprising coincidence with organized by Google for students and tud
X
S
for
zine
aga
C MM
the University’s initials, the conference industry professionals across Europe, Cro
ssro
a ds T
he A
is actually named after Augusta Ada the Middle East, and Africa. Partici- AC M.O
RG
D S.
King-Noel, Countess of Lovelace. She pants were tasked to solve a real-life

XR
N O .1
2•
L .2
VO
was popularly known as Ada Lovelace, engineering challenge created by F A LL

20
15
an English mathematician who is now Google. Before this, in November 2016, zine
for
Stu
den
ts
aga
MM
renowned for her work with Charles the group had also organized an event
ssro
ads Th
e AC
Cro
Babbage on the mechanical general- known as “Hour of Code in Azerbaijan:

purpose computer. This year’s guest KidCode.” They gathered a group of 30
was Stanford University’s Dr. Jennifer students from an Azerbaijani school,
Widom, who is one of the leading re- and gave them hands-on experience
searchers in database systems and big with beginner-level coding.
ital ion
data. The main event was followed by a It is quite evident from their activi- Digbricat
panel discussion in which Azerbaijani ties they are passionate about organiz- Fa
S
for n
women leaders in IT, along with Dr. Wi- ing a diverse set of activities in com- est atio abric
anif bric
A M ital Fa F XRD
S. AC
M
D
with Han
d
Dig ting
.4
in sa 2•
NO
t Pr end
dom, shared their experiences with the puting for students, as well as the local S of
br icat
ion L
MM
ER
20
1 6 V
O L .2
R
Fa SU
younger generation. community. “We hope that there will Stu

den
ts
for
X
zine
On the same day, they also held a be more student chapters in Azerbai- The
AC MM
aga
ads
presentation by Dr. Widom titled “Big jan in the nearest future,” says Isgan- Cro
ssro
S
Data: Promises and Pitfalls.” It was darli, who is looking forward to more XRD
S. AC
M.O
RG
D
an instructional odyssey in which she collaborations and advancing comput-
O .2
•N
L .2 2
VO
15
20
TER
XR
W IN
traveled around the globe, offering er science throughout Azerbaijan. den

ts
Stu
free short-courses, workshops, and —Anshuman
— Majumdar AC MM
aga
zine
for
The
ads
roundtables in the areas of big data, Cro
ssro
design thinking, collaborative prob- DOI: 10.1145/3155222

lem-solving, and women in technol- Copyright held by author.
The
begin
2013 Women in India only earned 75.19%

of what their male colleagues earned.
CAREERS
Software is for Humans

Holding our industry accountable.
W
e are witnessing more
people outside of the
tech industry using
code to solve problems.
I have met teachers who automated
grade calculations using macros and
spreadsheets, office managers who
built their own conference room
booking web apps, and even a magi-
cian who learned PHP so he could
implement a text messaging API to
enhance his act.
These people may not consider
themselves developers, but they’re us-
ing our tools, much like we use tools
to solve non-technical problems. With
this comes more responsibility when
we build our tools, both in regards to
user experience and the communities
we build around those experiences.
Yet, the bar is very low these days
for ethical tech that keeps humans in
mind, and creating a space where us-
ers feel comfortable asking for help
is a small, but impactful, step that
many companies can and should im-
plement. My experience in working
with the Glitch community is that be- diversity statistics (or the lack there- sometimes seems like consequences
ing nice is good for business. We are of), the disparity in pay wages, the are not enforced in industry, even
working on an in-browser editor that constant controversy loops around when laws are being broken. The hu-
allows anyone to tinker with code sexual harassment in the workplace, bris that comes with this lack of ac-
to create the apps of their dreams, and the many companies and apps countability is, I believe, one of the
but more importantly we are ac- we see come and go as they fail to biggest issues facing the tech indus-
tively building a friendly community meet the needs of society. Many of try today.
around the product. This gives people us are creating software for humans, You are entering an industry with-
Image by Redpixel PL / Shutterstock.com
of many backgrounds the ability to but this ends up being software that out a defined set of ethics and a lack
collaborate, ask for help, and learn inhibits us, either financially or cre- of accountability compared to other
from each other. atively, because it is software that fields, so it will become your respon-
And still, it is easy to take for grant- only benefits a certain class. sibility to create and maintain your
ed the importance of humans in our It was not until I left academia and own accountability. This personal
work. The tech industry has become became a web developer when I no- ethic is how I’ve survived as a wom-
complacent about its ability to not ticed a lack of accountability in rec- an in tech, despite the intensity and
think sociologically when designing ognizing the importance of helping amount of work it takes.
software. We witness this in company society when building technology. It As you begin sending out your

14.4%
of board seats in the Fortune 500 largest companies were held by members of ethnic minorities in 2015.
Meanwhile, almost 43% of the U.S. population belongs to an ethnic minority.
MILESTONES
Technology and
Dating
resume and applying for tech posi- Since its inception, the culture of dating has undergone quite
tions, remember that only in the last a change. From personal ads in a British newspaper in the late
five years are we slowly starting to 1600s to online platforms for virtually every city, religion, race,
see companies being held account- and sexual orientation in 2017, the idea of dating has evolved
able for breaches of ethics. More me- vastly because of technical advancements over the years.
dia outlets are covering crimes like
sexual harassment in tech, and social
media has provided marginalized
tech employees a space to be heard.
1959 The first attempt toward developing a
matchmaking service was made by Jim Harvey
and Phil Fialer. As a part of a class project for the Happy
Good tech can solve problems, like Families Planning Services, 49 men and women were matched
my grade calculation and conference using a punch card questionnaire and an IBM 650 mainframe
room booking examples, but it can computer.
also just be a communication channel
where it is safer to voice concerns and
ideas. Where our voices are a little bit
louder and more likely to be heard by
1965 The first computer-based dating service,
Operation Match, was started in the U.S. by a
few undergrads from Harvard. The users were asked to fill out
those who can create change. a paper-based questionnaire, which was transferred to punch
My final advice to students graduat- cards and then processed on an IBM 7090. Thereafter, users
ing into the tech industry is to not for- received a printout listing details of their matches.
get who you are solving problems for.
We don’t learn alone, regardless of
what industry we are in, so the tools
we build should reflect that. Use your
2000 Neil Clark Warren and Greg Forgatch founded
eHarmony, which is an online dating website
designed for single men and women looking for long-term
privilege to vocalize your ethics and relationships. The platform has members from more than 150
push those around, and even above, countries. It exclusively runs a relationship research facility to
you to do the same. Harness the pow- optimize its matching process.
er of social media to not only lift your-
self up, but others around you as well.
You will probably find, like I have, that
the most rewarding part of working in
2009 Compatible Partners, an online dating service
that serves the gay community, was launched
by eHarmony Inc. in response to a settlement with the State
tech is knowing the work you do helps of New Jersey. The lawsuit claimed eHarmony was in violation
everyone, which is what software is of the law prohibiting discrimination based on sexual
supposed to be doing after all. orientation.
2012 The location-based social search app Tinder

Biography
Jenn Schiffer works at Fog Creek Software as community was launched. The app uses Facebook and
engineer for Glitch.com. Previously, she worked as
an engineer at Bocoup and the National Basketball Spotify profiles for matching users, and is available in about
Association. She received both her bachelor’s and master’s 30 languages. By the end of the year 2014, Tinder was used by
in computer science at Montclair State University, where
she was an adjunct and then department administrator for more than 50 million people every month with an average of
the computer science department before she entered the 12 million matches per day.
tech industry.
—Alok
— Pandey
DOI: 10.1145/3155132
Copyright held by author. DOI: 10.1145/3155216 Copyright held by author.

begin
Steve Jobs was the son of Abdulfattah Jandali,

an Arab Muslim immigrant from Syria,
and Joanne Carole Schieble, a Wisconsinite
whose ancestors immigrated from Europe.
ACCOLADES
ElectroEuro
A virtual coin that can decarbonize.
I
n 2017, General Electric (GE)—one rael, promoted the challenge on the Haim Bender, like Ran and Idan,
of the largest American compa- Facebook page of FemTech, a com- came from Israel. He studied math
nies specializing in oil and gas, munity in Israel for women in technol- and computer science at Tel Aviv Uni-
healthcare, aviation and software ogy with more than 1,000 members. I versity. Isaack Rasmussen, the only
development—and Eurelectric—a col- was the founder and CEO of FemTech, non-Israeli on the team, is a software
lective representing the electric industry and when I saw the announcement I developer with more than a decade of
across Europe—partnered to create the immediately thought of Sheryl Sand- experience. He is originally from Africa
Ecomagination Challenge Hackathon. berg’s words: “What would you do if and now resides in Denmark.
The challenge was to build digital solu- you were not afraid.” I decided to par- I was privileged to have been men-
tions to help decarbonize energy and ticpate. This would be my first competi- tored by another master’s student at Bar
transportation in Europe. The two-day tive hackathon. Ilan, Micah Shlain, who taught me the
hackathon was held in Berlin, Germany The formation of our team was a bit principles of software development. It
this past June, where more than 100 par- unorthodox. Before I formally selected was through this process that I was able
ticipants from around the world came the other members of the team, Ran to guide the team from idea to design to
together to solve two challenges: electri- Koretzki, a graduate student at the implementation.
fication and advanced manufacturing. Technion in Computer Science in Isra-
For the electrification challenge, the el, asked to join. We were the original A WINNING COMBINATION
goal was to develop solutions where core of the team, a true Cornell-Tech- We designed a decentralized virtual
things are powered by electricity. For nion alliance. Ran, whose technical ex- currency known as the “ElectroEuro”
example, solutions such as: renewable perience includes summer internships for trading energies through an en-
energy resource siting, electric heating with both Google and Facebook, served ergy bank in Europe, driving a low car-
and conversion to heat pumps analy- as the team’s CTO and developed most bon economy. The use of green energy
sis, electric vehicle charger siting, and of the solution’s technical architecture. would promote decarbonization, and
renewable energy integration. For the My role was the CEO of our team. monetization would make it accessible
advanced manufacturing challenge, I then added Idan Nesher, a user ex- and practical. The concept was to unite
the goal was to optimize existing manu- perience (UX) designer, to the team, Europe through electricity like the euro.
facturing processes. Sample solutions knowing the key to winning would be Our proposed currency is similar to Bit-
include: forecasting manufacturing de- a compelling presentation. As much as Coin in that it is universal and there is
lays based on parts complexity, detect- architecting an energy bank and virtual a finite quantity of it. The transaction
ing delay drifts, and optimizing critical currency system would win the hearts of of energy is carried out through it, and
production rescheduling. Both chal- the judges, the presentation would need it can be bought through goods that do
lenges sought solutions that drive the to be at the same professional level. not promote carbonization. The energy
decarbonization of Europe. The official Idan studied product design at the Avni bank consists of eight sources of energy
platform for both challenges was Predix, Institute in Tel Aviv, and subsequently ranked by the green-factor and stabil-
GE’s industrial Internet of Things (IoT) moved to Berlin and worked as a free- ity, respectively. The price of the energy
platform. It is described as a “software lance designer. He dove fast into the UX is based on two metrics: 1) the distance
platform for the collection and analysis design world, believing UX will be the to transport the energy (a fixed price),
of data from industrial machines.” future of all products since it is centered and 2) the quantity. A market is gener-
Our team competed in electrifica- on innovation. ated based on the surplus of energy per
tion, which was considered to be the Since seeing is believing, I added country and per energy source.
greater of the two challenges. two other developers to the team who Machine learning is used to pre-
would be able to do front-end devel- dict consumption, production, and
GETTING IN FORMATION opment in order to have a live demo cost, through a set of sensors that de-
Ira Blekhman, from GE Digital in Is- ready to show at the hackathon. tect features for each type of energy.

2017
Stanford professor Maryam Mirzakhani, who was the first
woman and the first Iranian to win the prestigious Fields Medal
for Mathematics, passed away due to cancer.
For example, a set of sensors detects

weather which is then used to predict
the availability of solar energy. The
other features include: location, cost
of operation, availability, difficulty of
harnessing, volume, waste, risk, fail-
ures, pollution, and cost of production.
There is then an intricate bidding proc-
ess performed on an interval where
green and stable energies are promot-
ed. The machine learning, which is the
core logic of the system, is implement-
ed on Predix, and the rest is still in the
design phase.
This solution offers significant
technological benefits: hybridization, flexible trading rules, and allowing for We further earned up to 10,000 eu-
mobility, decentralization, big data op- delayed payments. For the supplier this ros for travel expenses to Portugal for
timality, and efficiency. Hybridization entails: forgiving debts, issuing small the Eurelectric conference to present
is achieved through combining differ- loans, and imposing penalties for trad- our winning solution. This project was
ent energy sources. Mobility is realized ing polluting energies. A free market is slated to be a one-off, but GE is inter-
through creating a mechanism that al- generated because there is less reliance ested in continuing to support it to see
lows one to obtain energy from a near- on OPEC. Additionally, unlike OPEC, it validated and developed.
est neighboring EU country rather than where production is heavily driven by
through OPEC. Decentralization is in politics rather than revenue, this con- Biography
place because of the virtual currency, figuration is less politically driven and Shortly after graduating from Cornell ECE in 2006,
Talia Kohen went to work for Raytheon with a position
and it leads to the creation of a free more idealistic. Lastly, countries are focused on tracking and discrimination. She is currently
market. Big data optimality is materi- more autonomous than under OPEC. completing a master’s degree in computer science on
the evolution of words in the Google Books Corpus. She
alized through a large sensor network, OPEC regulates countries very strictly. has implemented three algorithms, two known, and
one she designed and later found in a textbook, and
which generates ample data for analy- It penalizes countries that under- with those has been able to analyze one billion lines
sis. Lastly, efficiency is achieved thanks produce by limiting their negotiation of text in two-and-a-half hours. Kohen independently
discovered Euler’s F-Vector by visualization. She has
to the collected big data that allows for power, and fines countries that overpro- received numerous awards and honors including: Anita
replacing or moving resources through- duce. Every country is individually in- Borg Birthday Celebration Director in Israel; a Google
OutStander; Google Anita Borg Scholarship for Women
out a network. clined to cheat by discounting its prices in Europe and the Middle East Finalist; Google Campus
Additionally, ElectroEuro offers po- and exceeding quotas. Ambassador; Microsoft Israel Women of Excellence
Program; Microsoft Excellence Summer Camp; ACM XRDS
litical and economic benefits: It makes We presented our pitch on a boat feature issue editor for the IoT Edition; IEEE International
green energy cheap, generates a free docked on the Westhafen Canal to a Radar Conference Poster Session Co-Chair and Steering
Committee; Raytheon Individual Performer Achievement
market, promotes production driven panel of judges. Several hours later, Award - Ionospheric study Raytheon Spot Award;
by revenue, and promotes autonomy near the conference venue for Minds + Raytheon Women’s Network MDC Site Representative; and
was a Delegate to Grace Hopper Conference for Women
for individual countries. Green energy Machines Europe, we found out we won In Computing. She is the CEO of FemTech, a community
is more affordable, because there is a first place in electrification, which was for women in STEM in Israel. Upon completion of her
master’s degree, Koehn plans to earn a Ph.D. in artificial
concerted effort to make it easier for 10,000 euros for the team! Everyone was intelligence.
both the producer and the consumer. surprised and delighted. We also won
For the consumer this means: lower- 3,000 euros for second place in the cat- DOI:10.1145/3157679
ing the price of green energy, creating egory of Predix development. © 2017 ACM 1528-4972/17/12 $15.00

begin
The XRDS blog highlights a range of topics from conference coverage, to security
and privacy, to CS theory. Selected blog posts, edited for print, are featured in
every issue. Please visit xrds.acm.org/blog to read each post in its entirety. If you
are interested in joining as a student blogger, please contact us.
BLOGS
Figure 1. A workbench for this FPGA project; on the left is

the circuit, above is the FPGA, and in the background is a
resistor color band chart.
WHAT IS A FIELD PROGRAMMABLE GATE ARRAY?

Interfacing an FPGA According to Xilinx (http://www.xilinx.com/), the inventors
of the FPGA: “Field Programmable Gate Arrays (FPGAs) are
with an External Circuit semiconductor devices that are based around a matrix of
and Applications configurable logic blocks (CLBs) connected via program-
mable interconnects.
By Alexander DeForge FPGAs can be repro-
grammed to desired Figure 2. Astable multivibrator
application or func- circuit diagram with LEDs.
Let us begin with the question; “When is breadboarding tionality requirements
computer science?” The answer is, when you have a field after manufacturing.”
programmable gate array (FPGA). Building an astable For engineers or
multivibrator and interfacing it with an FPGA was the scientists who want to
focus of this project. FPGA chips have low-level design ap- make design decisions
plications. Interfacing an electronic circuit with an FPGA at the hardware level,
allows for a simple and lightweight user interface during an FPGA is a useful
the design process. development tool.

90% of trans people in the U.S. have reported
experiencing harassment, mistreatment,
or discrimination on the job.
CIRCUIT BUILDING FUNDAMENTALS

AND THE ASTABLE MULTIVIBRATOR Figure 3. A completed astable multivibrator used with Mojo
The circuit constructed for this project, an astable v3 FPGA.
multivibrator, requires resistor-capacitor (RC) circuits
and transistors.
The basic RC circuit contains a single resistor and a
single capacitor. (This was the type used for this project.) In
an RC circuit, after a calculable amount of time, no current
will pass through the capacitor. Current can flow through
a capacitor while it charges, but not once it is charged.
The value, in Farads, of the capacitor times the value, in
Ohms, of the resistor equates to the number of seconds it
takes until the capacitor is fully charged, creating an open
circuit. This time-based functionality is what determines
the period of our clock signal.
Ohm’s law states: V = IR. V is the voltage, I is the current,
and R is the resistance. With regard to circuit construction,
this equation is useful for knowing the voltages at different
points in a circuit. This is important for equipment safety.
In simple direct current circuits (DC circuits), the current
through the components is determined by the power
source (in this case a 9V battery, or about 500 milliamps).
Resistances are determined by other design decisions;
for this project, resistances were determined by what was
needed in the RC circuits.
It is important to use Ohm’s law to calculate and place
extra resistors (or higher resistance resistors) at certain
junctions in the circuit. The FPGA input, for example,
functions at 3.3V. Inputting even tenths of a volt higher
than 3.3V could damage the FPGA.
The final piece of the astable multivibrator is the
transistor. For this project, the transistors acted as switches.
Such transistors have three leads: collector, emitter, and
base. The collector is the input, the emitter is the output, and Figure 4. Implementation schematic for
the base is what is used to control whether the collector and astable multivibrator.
emitter are a closed circuit or open circuit. When a voltage or
current is applied to the base, the collector and emitter are
a closed circuit. When no voltage or current is applied to the
base, the collector and emitter are an open circuit.
The astable multivibrator is composed of two of an RC
circuit and a transistor, in parallel. LED’s have been added
to the diagram (See Figure 2).
The portion of the circuit with an LED is where the clock
signal/square wave is captured. There are actually two
antiphase signals produced by this circuit. Only one needs
to be captured, and it does not matter which. The signal
behaves as the LED’s behave, while one LED is illuminated,
the other is not, then the other illuminates while the other
is off, etc. The theoretical diagram and what was actually

begin
100,000 Between 1933 and 1945, authorities in Nazi Germany arrested an estimated
100,000 men for homosexuality, of which 50,000 were sentenced and
as many as 15,000 were sent to extermination camps. Even after this ordeal,
they were still considered criminals by the liberating Allied forces.
BLOGS
built differ, but not fundamentally. slice controller for this basic round robin scheduler.
The completed circuit, which was interfaced with the The Mojo IDE comes with code for simulating a CPU.
FPGA, has a variable clock rate by way of the potentiometer. There are some assembly-like instructions, which are
Each clock tick can both be seen with an LED, and heard provided, but any assembly-like instructions can be
from a simple speaker, and depressing a button makes the defined for a given application. Paired with a round robin
circuit live. Tapping voltage from the same spot as an LED scheduler and external clock, this is a test environment for
is how you extract the clock signal from this circuit. All haptic two-job logical concurrency.
that is required is an extra wire rooted at that spot with the There are additional files needed, which are also in the
other lead placed into one of the FPGA pinholes. XRDS github repository: modified_cpu.luc, modified_
instRom.luc, modifiedcpu_mojo_top.luc, and sliceclock.ucf.
A PROOF OF CONCEPT For this project, the majority of instRom.luc and
The Mojo v3 was the development board used for this cpu.luc were left intact. In modified_cpu.luc and modified_
project. The following is a disclaimer. These devices can be instRom.luc, the major changes were to allow a CPU to
expensive, and having the ability to download the neces- emulate logically concurrent behavior; it can be active or
sary software to develop on these devices is not straightfor- inactive and raises a “complete” flag. All the scheduling
ward. For the Mojo v3, there is no open source development was added to modifiedcpu_mojo_top.luc. Two CPUs are
alternative, and obtaining licensure with Xilinx to be able created and then managed with respect to an external
to build projects with Embedded Micro’s IDE will be an signal from PIN1. The constraints file sliceclock.ucf is
obstacle for some due to legal reasons. When deciding on nearly identical to pinIn.ucf.
a development board, it is recommended to follow through Once loaded, the FPGA will illuminate the first LED if
on all steps to be able to develop for the chip, up until actu- job1 completes first, or it will illuminate the second LED
ally loading a ROM onto the FPGA, before you buy. if job2 completes first. But when both jobs are complete,
The Mojo v3 contains the Spartan-6 FPGA chip. The all eight LED’s will illuminate. If no external circuit is
hardware definition language (HDL) Lucid was used in the interfaced, the first LED will light and nothing else will
embedded micro Mojo IDE version B1.3.6. Programming happen because job2 can never be active. When PIN1 has
at this level involves using signals, whose state does not voltage, job2 executes, and if PIN1 doesn’t have voltage,
persist between clock cycles, and flip-flop gates, whose job1 executes. If a circuit is interfaced, then all eight LED’s
state does persist between clock cycles. Both are accessed will illuminate instantly. This is because the instRom
as bit values or arrays of bit values. programs are short compared to the speed of the FPGA. A
The applicable files for this tutorial are modified_ further step would be to run more substantial programs
blinker.luc, modifiedblinker_mojo_top.luc, and pinIn.ucf, with this test environment.
which are located in the XRDS github repository at:
https://github.com/acmxrds/winter-2017/tree/master/blog. WRAPPING UP
Once loaded into the FPGA, the built-in LED’s will now When engineering or doing computer science at the hard-
blink in unison with the clock signal from the astable ware level, there is an option to gain intuition about your
multivibrator. The main changes were to the blinker.luc system haptically, and there is an option to develop con-
and pinIn.ucf files. In pinIn.ucf, the FPGA is instructed trols for your system externally. To do so requires develop-
to access PIN1 on the development board. In modified_ ing an interdisciplinary knowledge of electronics. All code
blinker.luc, the FPGA is instructed to treat PIN1 as an input presented is available in the XRDS github. Happy hacking!
signal. That input signal controls a case statement with two To replicate what was done here (and specifically for
cases, one for voltage and one for no voltage. In this setup, the Mojo v3), all that is required is opening a new project
each case sets the built-in LED’s to on or off. (“From Example,” “Blinker Demo,” or “Basic CPU”) and
making the modifications.
APPLICATION
What we now have is an external variable-interval clock and Biography
software that supports alternatively executing two sets of Alexander DeForge is a student at the University of Maryland University College, studying
computer science. He is primarily interested in systems, including operating systems,
code—the two cases. User-controlled logical concurrency computer networking, and other low level applications.
can be emulated in this FPGA for exactly two jobs. The ex-
ternal variable-interval clock can be used as a haptic time DOI: 10.1145/3155224 Copyright held by author.

Introducing ACM Transactions
on Human-Robot Interaction
Now accepting submissions to ACM THRI
In January 2018, the Journal of Human-Robot Interaction (JHRI) will become an ACM
publication and be rebranded as the ACM Transactions on Human-Robot Interaction (THRI).
Founded in 2012, the Journal of HRI has been serving as the

premier peer-reviewed interdisciplinary journal in the field.
Since that time, the human-robot interaction field has

experienced substantial growth. Research findings at the
intersection of robotics, human-computer interaction,
artificial intelligence, haptics, and natural language
processing have been responsible for important discoveries
and breakthrough technologies across many industries.
THRI now joins the ACM portfolio of highly respected

journals. It will continue to be open access, fostering the
widest possible readership of HRI research and information.
All issues will be available on the ACM Digital Library.
Editors-in-Chief Odest Chadwicke Jenkins of the University of Michigan and Selma

Šabanović of Indiana University plan to expand the scope of the publication, adding a new
section on mechanical HRI to the existing sections on computational, social/behavioral,
and design-related scholarship in HRI.
The inaugural issue of the rebranded ACM Transactions on Human-Robot Interaction is

planned for March 2018.
To submit, go to https://mc.manuscriptcentral.com/thri
feature
Can We Build
the Cyborg Future
We All Deserve?
Knowing who we represent in HCI helps us understand
what is at stake. Intersectionality can help us do better.
By Ari Schlesinger
DOI: 10.1145/3155116
W
e are all cyborgs. No, not the “Terminator” Arnold Schwarzenegger kind;
we are a different type of cyborg. We are living in a world where the technology
we create is shaping who we are. When you think about it, the boundaries
between you and your technologies are fuzzy at best.
Most of us rely heavily on our tech to get through the day. Smartphones wake us up in the
morning with ringtones labeled oxygen, drip, and krypton. They let us swoon over perfectly
plated food pictures posted by someone we will never meet, halfway around the world. They
connect us to our friends through video chat, even if we’re in the same room. We see the
whole world through the high-megapixel lens of our smartphone. Whether your day is
shaped by a potential tweet, an Insta- tech is designed by and whom our experiences, user needs, user errors,
gram post, a Facebook message, or a tech is designed for. user practices, and user preferences.
Snap, you’re experiencing life as a cyborg. This means we need to talk about But who is this illusive user? Is it you,
The cyborg future is a networked human-computer interaction, or HCI for me, Laverne Cox, Katya Zamolod-
existence where you and your tech- short. HCI is a field dedicated to interfac- chikova, your weird relative, a tech
nolog y shape the world. Letting go es between people—often referred to bro, the Terminator?
of physical boundaries, where do as “users”—and computing technolo- When we’re talking about the user
you end and where does your smart- gies. This is a pretty broad field. HCI we’re talking about complex people
phone begin? The closing distance experts and practitioners have back- with differences from one another. I’m
between you and your phone is only grounds in design, computer science, willing to go out on a limb and say you
the tip of a giant, cyborg iceberg psychology, media studies, and more. and I have at least as many differences
(iceborg?). Think about it, if tech Likewise, HCI research covers a mas- as we have similarities. Some of these
is literally shaping who we are and sive variety of research styles, topics, differences relate to socially-situated
what the world is, it is an incredibly technologies, and users. Research cov- identity factors, like race, gender, sexu-
powerful tool that is being used to ers topics like user experiences with dif- ality, and class. Other differences arise
build our future. ferent application interfaces, anonymity from things we do, love, care about,
If we’re all cyborgs, living in a on social media sites, people’s password- work on, and know. All these differenc-
networked world co-constructed by sharing practices [1], and interactions es are instructive. They make the world
our tech, we need technology that is with artificially intelligent chatbots [2]. interesting. Can you imagine how bor-
designed by and for a wide variety of HCI is infatuated with the user. ing the world would be if everyone and
people. We need to know whom our We are constantly talking about user everything were exactly the same?

Exposure to things that are dif-
ferent than we are allows us to grow,
learn, and engage with the world in
life-expanding ways. Given how im-
portant difference is, we need to know
if we’re taking these differences into
account in our understanding of the
user. When we are designing technol-
ogy for a wide variety of users, we will
be well on our way to building our bet-
ter cyborg future.
So, whom are we designing for? In
research published in 2017, Rebecca
Grinter, W. Keith Edwards, and I inves-
tigated this question [3]. Unfortunately,
what we found wasn’t great. The bad
news is, we don’t do a very good job at
thinking about a wide variety of identity
factors like gender, race, and class when
thinking about the user in HCI.
Although we may be cyborgs, gen-
der, race, and class have immense in-
fluence on how we experience and act
in this world. These important identity
factors are a critical part of who the
user is. That’s why this research ex-
amined how these identity categories
were talked about in HCI’s premiere
publication venue, the CHI confer-
ence. By gathering publications that
talked about these categories, we can
understand who we’ve been designing
for and develop plans to better design
technology for all the differences of
human-cyborg users.
In order to fully understand who
we’ve been designing for, we need to
know what research says about users
in all their complexity. Gender, race,
and class are very broad umbrella
categories, and don’t tell us about
specific terms we use to identify our- up with 140 publications that men- percent of all SIGCHI publications. Ei-
selves and express our differences. So tioned at least one of these identity ther way, that’s a truly small number of
we developed a detailed set of terms terms. Here, publications refer to full publications. If it tells us anything, it’s
we use to describe our identity. This papers, short-papers (notes), and ex- that research isn’t talking about these
set of more than 50 identity keywords perimental papers (alt.chi). Ultimately, socially situated identity attributes
inclued transgender, boys, women, not a lot of research talked about users when talking about the user.
black, white, Native American, Asian, with these common and important Things get a bit more encouraging
homeless, middle class, poverty, and terms. To get a feel for the scale, at the when we look at how these identity
Image by Studiostoks / Shutterstock.com
Chicanx (including its lexical variants time of this research, there were nearly terms are talked about over time. His-
across the gender spectrum: Chicano, 14,000 ACM SIGCHI publications. This torically, the presence of these key-
Chicana, Chicanx, etc.). With these means that CHI papers focused on words is heavily skewed toward the
words, we were able to find previous these identity keywords represent only recent past; most of the publications
research in the ACM Digital Library 1 percent of ACM SIGCHI publications, are from the 2000s, with two during
(ACM DL) that would help us under- Although the 14,000 includes more the 1980s (1.43 percent), two during
stand who we’ve been designing for. than just CHI papers, the numbers are the 1990s (1.43 percent), 33 during the
At this point, you might be wonder- only slightly better if you include all the 2000s (23.57 percent), and 103 dur-
ing, what did the research find? Well, publications our searches gathered; ing 2010–2016 (73.57 percent). Even if
after querying the ACM DL, we ended there were a total of 309 hits, or 2.21 there isn’t as much research that is ex-

feature
plicit about who the user is, there is a 1964, the year of the Civil Rights Act. argued if we want to build a better fu-
lot more now than there was even only Now, dismissing all black women em- ture for everyone, we must focus our
10 years ago. ployed at one company within a single attention on those who are marginal-
Although this is only a small per- year would appear to be discriminatory ized in intersectional ways. If we can
centage of CHI publications, by inves- right off the bat, but the court didn’t address the needs and struggles of the
tigating what is included in them and see it that way. Here’s what happened, individuals, we will have made strides
what is missing, we can learn how the women are a protected group under forward for everyone. This is absolutely
user is described and figure out how to employment law and black people are true when working toward building the
do better for building the cyborg future a protected group under employment cyborg future we all deserve.
we deserve. On further analysis of the law; however, these laws made no stip- Building a better future for HCI re-
final 140 publications, we found that ulation for people who are both black quires us to do more when it comes to
the focus on the umbrella categories— and women at the same time. As a re- the identity of the user. By centering
gender, race, and class—was not evenly sult, the court explained that because people with intersectional, margin-
distributed. Additionally, prior research GM still had (white) women employees, alized identities, we can break away
looked at some identities more than it wasn’t violating anti-discrimination from discriminatory practices that
others. Though there is some overlap, statues protecting women. The court render large portions of the popula-
the numbers reveal that 70 percent of also dismissed their claims for race tion invisible.
publications focused on gender, 12.12 discrimination, suggesting the wom- Intersectionality is an essential
percent on race, and 30 percent on class. en join a different lawsuit against GM framework for any work that interfaces
The focus on identity was unbalanced. based on race discrimination. But, with human identity. For this research,
It’s disheartening that certain iden- the plaintiffs, the black women suing, it’s important to ask how the publica-
tity attributes were the focus of so few countered that their case was specifi- tions we collected stack up. How many
publications; however, it is also encour- cally a problem of gender and race- focus on more than one identity cat-
aging that these percentages don’t add based discrimination, which would be egory within a single publication? In
up to 100 percent. Identity is not an ex- misrepresented in joining a race-based total, 24 papers focused on more than
clusive category. When thinking about case. Nevertheless, the court doubled- one umbrella identity category. While
users, HCI can and should also con- down and completely dismissed the encouraging, at the end of the day, most
sider gender, race, and class at the same case. The court concluded the govern- of the corpus didn’t fare this well. In
time. This is critical because identity is ment was not intending for a special order to build better technology, better
complicated; people do not experience category of black women to be pro- human-cyborg computer interactions,
just one identity attribute at a time. Put tected under the law since there was we all need to think about how intersec-
another way, your race, gender, and no employment law addressing these tional identities affect our work.
class co-exist at all times and these in- intersections of identity. The experi- Another identity-related shortcom-
tersections impact your experiences of ence of black women, who experience ing we uncovered was how papers about
the world. A person is not black on Mon- racism and sexism in combination, was gender often collapsed identity differ-
days and a woman on Fridays, nor are erased. ences in ways that make the experienc-
they white on Mondays and a man on There are clear, concrete, discrimi- es of some users invisible. It is great that
Fridays. This brings us to an important natory consequences when we fail to research considered gender, but gen-
topic, intersectionality. consider identity overlaps. Crenshaw der was often a shorthand for women,
Intersectionality is a flexible frame- which is only one of many gender iden-
work for analyzing identity attributes tities. Further, with intersectionality as
like gender, race, and class. It allows us a guiding framework, it becomes clear
to investigate how identity attributes that women as a category cannot ad-
impact one another, a person’s sur- equately address the experience of the
roundings, and larger social systems.
The term intersectionality was
When we understand many differences that exist within this
identity-construct. HCI should thus
coined in 1989 by legal scholar Kimber- how identity-based consider the wide spectrum of gender
lé Crenshaw [4]. But so far, this descrip-
tion is pretty abstract. What does it
discrimination identities that exist, not just narrow rep-
resentations of women.
mean in a more concrete way? In Cren- is designed into Taking intersectionality seriously
shaw’s article, she wrote about a 1976
lawsuit against the company General
our technologies, gives us an opportunity to improve
development at the intersection of hu-
Motors (GM), DeGraffenreid v. General we are able mans and computing. To build the
Motors, where five black women sued
GM for discriminatory practices. GM
to address cyborg future we need, we need to be
clearer about the identities of the user.
fired all its black women employees our shortcomings This includes working with people
hired after 1970 in a seniority-based lay-
off scheme; it’s important to note they
and develop whose identities have not been given
enough space in HCI. Our research
didn’t even hire black women before new paths. uncovered multiple identity areas that

produced no matches in the SIGCHI
database at all. People whose identity
Identity is not for a world with ever-blurring bound-
aries between people and technology.
and experiences are informed by being an exclusive category. When we understand how identity-
genderqueer, gender fluid, queer, Asian
American, Chicanx and Latinx (and all
When thinking based discrimination is designed into
our technologies, we are able to address
lexical variations), First Nations, Alas- about users, our shortcomings and develop new
kan Native, Pacific Islander, and lower
class are users, but these identity key-
HCI can and should paths. We are able to design differently.
To create the better cyborg future
words resulted in no matches in the also consider gender, Donna Haraway formulates in her “Cy-
SIGCHI database. Moreover, intersec-
tional HCI requires us to acknowledge
race, and class at borg Manifesto,” we all need to play an
active role in developing the technol-
who is given space in our research and the same time. ogy that is shaping our world [6]. It’s
who has been left out. Knowing where not about one totalizing user. It’s about
we have fallen short as a field helps us embracing difference and diversity. It’s
pinpoint specific areas to build on for about embracing a multiplicity of users.
improving HCI. Further, it helps us tar- We need to step up and take ownership
get focus areas for making a better cy- of our responsibility for the entangled
borg future a reality. tions of users; this is just a first step. relationships between our social worlds
Through this research, we were able We need qualitative and quantitative and our technical ones. With intersec-
to uncover critical insights into iden- work to be invested in intersectional- tionality as a framework, we can build
tity work at CHI, but it also uncovered ity. We also need to be critical about the better computer interactions of the
critical questions for the community at the categories we are already using. cyborg future we all deserve.
large: How is it possible that so few pa- As we covered earlier, many papers
pers matched our search criteria? Given that talked about gender did so by us- References
the nearly 14,000 publications we are ing binary, cisgender categories (cis- [1] Singh, S., Cabraal, A., Demosthenous, C.,
Astbrink, G., and Furlong, M. Password Sharing:
pulling from, and HCI’s foundational gender meaning people whose gen- Implications for Security Design Based on Social
reliance on the user, it is truly remark- der identity is aligned with the gender Practice. In Proceedings of the SIGCHI Conference
on Human Factors in Computing Systems,
able that so few publications feature identity they were assigned at birth). ACM, New York, 2007, 895–904. http://doi.
keywords from common demographic Only a few papers mentioned trans org/10.1145/1240624.1240759
identity categories. If not with these identities. These publications were [2] Luger, E. and Sellen, A. Like Having a Really Bad PA:
The Gulf between User Expectation and Experience
common terms, how is the identity of focused exclusively on trans commu- of Conversational Agents. In Proceedings of the 2016
CHI Conference on Human Factors in Computing
the user represented? nities [5]. In other papers, however, Systems (CHI ‘16). ACM, New York, 2016, 5286-5297.
While there is a lot more work to do a side effect of this simplified repre- DOI: https://doi.org/10.1145/2858036.2858288
in building a cyborg future that priori- sentation of gender is the erasure of [3] Schlesinger, A., Edwards, W. K., and Grinter, R.
E. Intersectional HCI: Engaging Identity through
tizes the identity diversity of users, it trans identities. This erasure contrib- Gender, Race, and Class. In Proceedings of the 2017
is clear HCI is striving toward this fu- utes to widespread discrimination CHI Conference on Human Factors in Computing
Systems (CHI ‘17). ACM, New York, 2017, 5412-5427.
ture. Each year, more work is centered against transgender and gender non- DOI: https://doi.org/10.1145/3025453.3025766
on computing technology created by conforming folk, which puts their [4] Crenshaw, K. Demarginalizing the Intersection
and with people from many different, lives and wellbeing at continued risk. of Race and Sex: A black feminist critique of
antidiscrimination doctrine, feminist theory and
intersectional identity backgrounds. Intersectional HCI can help us do bet- antiracist Politics. University of Chicago Legal
While only a beginning, the 24 papers ter in building the human-computer Forum, 1989, 139–167.
with an intersectional focus covered in interactions we all deserve. [5] Haimson, O. L., Brubaker, J. R., Dombrowski, L., and
Hayes, G. R. Digital footprints and changing networks
our research are a good start. Many of The words we use to talk about during online identity transitions. In Proceedings
these publications were in-depth, quali- our users matter. Words are the main of the 2016 CHI Conference on Human Factors in
Computing Systems, ACM, New York, 2016, 2895–
tative studies into a particular intersec- medium we use to communicate our 2907. http://doi.org/10.1145/2858036.2858136
tional identity, like young black men or work. The words we choose can con- [6] Haraway, D. A Cyborg Manifesto: Science,
technology, and socialist-feminism in the late
homeless mothers. However, there were tinue to support structural, identity- twentieth century. In Simians, Cyborgs, and Women:
also a few papers that investigated mul- based discrimination, or they can The Reinvention of Nature. Routledge, New York,
1991, 149–181.
tiple identity categories quantitatively, draw attention to the intricacies, expe-
though in these publications identity riences, and needs of users as complex
Biography
categories were still investigated sepa- human-cyborgs.
Ari Schlesinger is a full-time cyborg and a Ph.D. student in
rately and then explained sequential- When it comes to the cyborg fu- human-centered computing at Georgia Tech. Schlesinger’s
ly—data on gendered presented first, ture we all deserve, we are clearly on research is focused on identity and infrastructure, looking
into how we can develop technology that builds just values
then data on race, then data on class. the right path. The CHI community is into multiple layers of a computational artifact. Find out
With this research at the roots, we can more at www.AriSchlesinger.com
thinking more about the user’s identity
continue to build and strengthen how than ever before. And, researchers are
identity is handled in HCI. increasingly communicating that iden-
© 2017 Copyright held by owners/authors.
There is more to intersectional HCI tity is something we need to address if Publication rights licensed to
than considering identity intersec- we are going to understand and design ACM 1528-4972/17/12 $15.00

feature
Digital and
Physical Barriers
to Changing
Identities
Social media sites often erect barriers to changing identities online,
which can be similar to physical world barriers faced by marginalized
groups. How can social media be designed to enable rather than
constrain life changes?
By Oliver L. Haimson
DOI: 10.1145/3155118
P
eople with changing and marginalized identities often have a difficult time moving
around in the world. While my focus here is on transgender people, a marginalized
group that has been especially visible recently, many of these ideas can be
generalized to other marginalized populations. Transgender (or trans) is a term to
describe “people who move away from the gender they were assigned at birth, people who
cross over (trans-) the boundaries constructed by their culture to define and contain that
gender” [1]). As an everyday and tangible example of the sorts of physical barriers trans
people face, several states in the U.S. recently passed or considered legislation requiring
people to use public restrooms that correspond to the gender listed on their birth certificate.
Of course, public spaces do not literally and remove trans people from public you were allowed to use was at your
have “bathroom police” designating restrooms. home. This would mean you could not
who can and cannot enter; and even if Setting up legal and social barriers be on campus for longer than an hour
they did, such gatekeepers would find it against using public restrooms does not or two; you could not go to a museum
difficult to accurately determine some- sort people into categories in a simple, or a concert; you could not get a job.
one’s birth sex. However, such laws biological way. Instead, it creates a world The recent biographical film “Hidden
promote dangerous behaviors, giving where trans people cannot exist in pub- Figures,” set in the 1960s, includes sev-
legal justification for people to harass lic spaces. Imagine if the only bathroom eral scenes where the main character,

an African-American mathematician, ample, Sherry Turkle discussed the communicate and interact with oth-
is forced to run half a mile back and internet as a social laboratory, where ers as a new gender.
forth from her workplace to the nearest people took refuge to experiment However, much as physical spaces,
bathroom she was allowed to use. While with their identities in ways that were like bathrooms, hinder people from
many people viewed this scene as an much more difficult in the physical living their lives, online spaces often
outdated historical artifact, it is all too world [2]. On some online platforms, erect similar types of barriers. As an
familiar to trans and non-binary people experimenting with one’s name, gen- example, think of the last time you
who have identified and frequented the der, and appearance is relatively easy. filled out an online survey or form.
unisex restroom nearest to their work- This is one reason that large com- Though many survey authors are get-
place or campus building, which, in munities of trans people use sites ting better at using inclusive gender
“Rest area” © David Fulme (daveynin; Flickr)
most cases, is neither nearby nor con- like Tumblr and Reddit—where you options, it is often the case that a
venient. Restroom discrimination is a can be anonymous, pseudonymous, survey asks a person to choose either
clear example of a physical barrier that or private by obscurity; that is, it is male or female, without any other op-
makes it difficult for a marginalized unlikely one’s parents would know tions. Maybe it’s a market research
group to exist in the physical world. Tumblr exists, think to search for survey for a car insurance company,
you, and be able to find your Tumblr in which case people for whom those
FROM PHYSICAL WORLD account. Online spaces are impor- categories don’t fit likely accept it as
TO DIGITAL WORLD tant means for people to find support a microaggression and move on. Yet
The online world is often posited as and information, ask questions and in other cases, limited gender op-
an antidote to such barriers. For ex- get answers, and see what it is like to tions feel harmful for some because

feature
it signals that other people do not first name remains an outlier condi-
COMMUNICATIONSAPPS acknowledge their identity in the way
they prefer.
Inadequate gender options on
tion with no technological solution in
place. Financial identity management
is a case in which technology that fails
forms may seem like a small inconve- to handle identity change can have
nience, but considering the frequency devastating real-world consequences.
with which one is expected to fill out
demographic questionnaires, it can CHANGING IDENTITIES
be a substantial source of stress, par- ON SOCIAL MEDIA
ticularly in professional or healthcare Social media sites similarly often have
contexts. The examples I describe in- difficulty managing identity changes,
volve surveys that do not give enough in ways that affect people socially
options. However in other cases, sur- rather than financially. In contrast
veys or forms ask for too much infor- to the identity exploration and infor-
mation and require uncomfortable mation seeking that happens on sites
levels of disclosure, even when gender like Tumblr and Reddit for people
information is irrelevant to the con- exploring marginalized identities,
text. For an excellent article exam- Facebook stands in contrast. It is a
ining the nuances and complexities site inhabited by people’s “real life”
Access the of trans-inclusive survey design, see connections, where they are required
Bauer et al. [3]. to use their physical world name and
latest issue, As another barrier to technological identity,and where multiple accounts
inclusion for trans people and others are prohibited. This can cause sub-
past issues, embarking on life transitions, many stantial difficulty for people during
technological systems are not set up identity changes.
BLOG@CACM, to handle identity change well. Lars The difficulty of identity changes is
MacKenzie’s recent article uncovered compounded by the network of people
News, and the ways credit agencies’ data systems who remain on one’s Facebook net-
more. (mis)handle the financial records of
people changing gender [4]. Such sys-
work over time, even after those people
are no longer meaningful in one’s life.
tems, which use technology to impose The algorithmically curated news feed
financial decisions with real conse- seems to slowly fade people away over
quences (such as being denied hous- time if one does not interact with them
ing, loans, and credit), have difficulty frequently, but these old friends and
handling a person who used to have acquaintances remain on one’s friends
one name and now has another. In list. When disclosing life events and
many cases, these systems and com- changes, one’s status update may pop
Available for iPad, panies “out” people as trans to poten- up in old friends’ feeds if certain words
iPhone, and Android tial employers and landlords, make are used (“I’m excited to announce…”)
people’s credit histories inaccessible or if enough people respond saying
to them, and even flag people as fraud- “Congratulations!” [5].
ulent when they attempt to access Disclosing a new identity, par-
their own finances. Such systems have ticularly one that is stigmatized and
clearly adapted to adequately handle places one in a marginalized group,
someone who changes their surname can be especially difficult given the
after marriage; yet changing one’s wide range of connections from the
past who remain on one’s friends list:
Available for iOS, people from high school, former co-
Android, and Windows workers, and people whom one bare-
ly even remembers. Going through
http://cacm.acm.org/
about-communications/
As people move one’s friends list and unfriending
based on anticipated reactions to,
mobile-apps from one life for instance, a gender transition, is
stage to another, a common practice but emotionally
exhausting. Many people appreciate
technological the ability to make mass disclosures,
systems must and disclose a life event to everyone
at once rather than through a series
move with them. of one-on-one conversations. How-

ever, the presence of a sticky social
network is often draining, and may
Financial identity systems when some users’ goals may
conflict with a company’s goals.
hinder change rather than support management is To understand how future social
the ever-changing nature of life.
Additionally, Facebook’s “real name”
a case in which media sites should be designed, it is
important to study how people’s be-
policy has disproportionately target- technology that fails haviors’ and networks change across
ed trans people and other marginal-
ized users, shutting down accounts
to handle identity the lifespan, through multiple and
intersecting life changes. Each person
of people whose names do not match change can have faces many transitions as they grow,
their legal documents [6]. Though
Facebook has made improvements
devastating real- age, and change. As people move from
one life stage to another, technologi-
to this policy, and seems to enforce world consequences. cal systems must move with them.
it somewhat less than in previous Life transitions rarely occur in isola-
years, in many online trans commu- tion; instead, people often face mul-
nities people still frequently discuss tiple life changes simultaneously. For
frustrating experiences with Face- instance, gender transitions often in-
book’s “real name” policy. By enforc- volve relationship breakups, and rela-
ing this policy, Facebook imposes a Facebook uses may be antiquated. tionship breakups often involve mov-
literal barrier between trans people Designing for life transitions ing. In each of these circumstances,
being able to exist within their online means designing for life. How can unique and complex challenges occur
social networks. social media keep up? We can look at around disclosure and presentation
Technological systems that im- the social media sites that are popu- on social media. By studying technol-
pose barriers on people’s social me- lar among young people to under- ogy use around overlapping life tran-
dia activity are ultimately not entirely stand how people may want to use sitions over time, and how people’s
different from the bathroom legisla- social media in the future. Snapchat, social media behaviors and networks
tion some states attempt to impose for instance, does an excellent job of change across the lifespan, we can
on trans people. Both attempt to po- letting contacts drift away over time. identify how social media can best
lice who does, and does not, have ac- Those people whom one has not sent support users over time as they transi-
cess to public spaces—one physical, a Snap to recently no longer appear tion through different identities, net-
one digital. The difference is inten- in the list of people to send one’s next works, and profiles.
tion. Facebook does intend to treat Snap to. The default is to send a snap
trans people fairly, and has a track re- to no one—a person must deliberate- References
cord of supporting lesbian, gay, bisex- ly think of whom they want to com- [1] Stryker, S. Transgender History. Seal Press, 2009.
ual, transgender, and queer (LGBTQ) municate with. As people fade out of [2] Turkle, S. Life on the Screen. Simon and Schuster, 1995.
rights. Yet intention is not enough; one’s life, they are no longer thought [3] Bauer, G. R., Braimoh, J., Scheim, A. I., and Dharma, C.
Transgender-inclusive measures of sex/gender for
technology must be built to support of and thus not included in one’s au- population surveys: Mixed-methods evaluation and
recommendations. PLOS ONE 12, 5 (2017).
changing identities, and identities in dience for personal snaps. Yet Snap-
[4] Mackenzie, L. Z. The afterlife of data identity,
which the present and the past may chat also allows people to share con- surveillance, and capitalism in trans credit
be in conflict. This is not easy. tent using “Stories,” a way to share reporting. TSQ: Transgender Studies Quarterly 4, 1
(2017), 45–60.
content with people in one’s broader
[5] Tufekci, Z. Why Twitter should not algorithmically
RETHINKING BARRIERS network. Snapchat is one successful curate the timeline. The Message. Sept. 4, 2014;
AND BOUNDARIES example of a social medium that en- https://medium.com/message/the-algorithm-
giveth-but-it-also-taketh-b7efad92bc1f
Everyone faces transitions throughout ables different levels of sharing with-
[6] Haimson, O. L., and Hoffmann, A. L. Constructing and
their lives, from teenager to adult, sin- out mandating that one’s network enforcing “authentic” identity online: Facebook, real
gle to married, college student to pro- stick around over time. names, and non-normative identities. First Monday
21, 6 (2016).
fessional, or employed to retired. Many Design decisions that enable and
may also experience other intersect- sometimes encourage content to be
ing exciting, tragic, or confusing life shared with fewer people do not align Biography
events and identity changes. As people with many social media sites’ goals, Oliver L. Haimson is a Ph.D. candidate in the Informatics
cross through boundaries between life which tend to involve users sharing Department at University of California, Irvine’s Bren School
of Information and Computer Sciences. He conducts social
stages, technological systems must more content with more people. Us- computing research focused on how people present and
move barriers out of the way. Barriers ers have diverse goals when using disclose changing identities on social media during life
transitions. Through his research, he hopes to impact
include things like one-account-per- social media: Many of them wish to technological inclusion of marginalized users.
person rules, or difficulty dealing with share more content with more people,
past content (for example, having to while others prefer to share content
un-tag every old photo manually). It with limited audiences. Often the
may be one profile cannot represent same person has each of these goals
a person throughout their lifespan; in different contexts and with differ- Publication rights licensed to ACM
the one-profile-per-person model that ent content. It is not easy to design 1528-4972/17/12 $15.00

feature
Disability-Disclosure
Preferences and
Practices in Online
Dating Communities
Based on a cooperative research project, this article explores the experience of
dating online with a disability, contextualized with an overview of the historical
connection between disability and asexuality. It concludes with ideas for
decoupling this inaccurate association through online dating platforms.
By Cynthia L. Bennett
DOI: 10.1145/3155120
M
y first attempt at online dating started with me swiping through photos on my
phone in search of the best ones. I decided on an array from my travels; they
would make me look like I’ve had a variety of life experiences, right? Next was
the text profile. I had heard few matches bothered reading this part, but I had to
write something. The alternative, leaving the profile empty, might have been disastrous
for my mental health. “I’m blind,” I began. I had read numerous articles online and talked
to friends with disabilities about the mean comments and ignorant questions matches
often asked about their disabilities, leading me to believe leaving the profile blank when
my photos show my disability would result in similar treatment.
Back to composing my text pro- social science literature. I was trying disabled woman’s experience telling
file, I wrote seemingly hundreds of to self-present, or show a version of her Tinder matches about her disabil-
sentences, all beginning with that my most attractive self—a person who ity, and the horrifying ableist messages
phrase. I settled on, “I’m blind. I’m has visited six different countries. But (i.e., disability-based discrimination
happy to answer questions, but I am I was also negotiating how to self-dis- or prejudice) she received from her
not your inspiration, charity, or edu- close, or share a truth. I know from matches [1]. Typical messages includ-
cation and it will not be our only topic countless personal experiences of ed inappropriate questions about her
of conversation.” A bit harsh, but I ignorance or ill treatment, that blind- intimate relationships and criticisms
couldn’t chance getting any messages ness would be very new to or problem- such as she should only date other peo-
like the ones my friends received. atic for some potential dates. ple who use wheelchairs. After reading
Looking back, I was negotiating In early 2016, I posted an article to the stories themselves, my colleagues
some concepts well documented in my online social network about one and paper [2] co-authors John, Kiley,

and Sarah reached out to me with their ies to frame our work. As human- disabilities. Though disabilities may
own personal experiences with online computer interaction researchers, seem ever present to nondisabled
dating. Together we began reading we wanted to bring in theory from a people, they become backgrounded
even more of these accounts posted field of study that is rarely cited in re- by people with disabilities until they
online. However in light of the chal- search on accessible computing, but are revealed to us through inaccessi-
lenges that disabled people seemed that is ironically well positioned to ble encounters. These encounters can
to be having when dating online, we theorize disability and accessibility. be concrete, such as the experience of
also found almost no academic work The field arose from disability rights a wheelchair user encountering stairs
that investigated their experiences, activism, and unlike any other schol- but no elevator, or more abstract,
despite the fact that nearly one in five arly field, many contributions are like when a disabled person does
Americans are disabled, according to written by disabled scholars. Specifi- not think about her disability until a
the U.S. Census Bureau [3]. Drawing cally, we borrowed two concepts for passer-by inquires about it.
from past research in this area, we this project. The first is that of visible Background and methods. We con-
began to study the online dating ex- and invisible disabilities. Visible dis- ducted a survey with two groups of
periences of people with disabilities, abilities refer to disabilities that are online daters. First, we recruited peo-
Image by TarikVision / Shutterstock.com
including how they choose to self- visually apparent in-person, while ple identifying as having a disability,
present and self-disclose and how invisible disabilities are not. We used impairment, mental health condi-
online daters, both disabled and non- these terms to understand peoples’ tion, or as being deaf. The second
disabled, prefer to learn about their expectations for disability-related group consisted of people who did
potential partners’ disabilities. disclosure while online dating, where not identify with any of those catego-
the intent is to eventually meet in- ries—we referred to them as “nondis-
OUR STUDY person. The second concept is called abled” in line with other disability
Throughout this project, we have “dys-appearance” [4], and it explains studies scholars. While we use the
consulted the field of disability stud- how we move in and out of noticing term “disability” in this article to re-

feature
fer to anyone who did identify within Our second strongest theme was choose what matches to answer rath-
those categories, we opened our re- using disclosure as a method for fil- er than disclosing to everyone who
cruitment description to make sure tering out ableist matches. In this views their profile.
labels did not interfere with us learn- way, our disabled participants typi-
ing about people’s experiences. In cally considered disclosure helpful POSITIONING DISABILITY
fact, we used this approach for most both for the recipient of such informa- POSITIVELY
of our demographic questions, such tion, as well as for themselves, helping Our paper also invited more research
as disability description, gender, them to avoid matches who would not at the intersection of disability and
race, and sexual orientation. By be accepting of their disability. For online dating. Looking back at our
adopting the approach of leaving de- example, a 24-year-old woman with a findings, another theme emerges:
mographic information open to allow chronic illness shared her reason for Positive encounters by our disabled
participants to describe themselves disclosing: “I’d much rather be open participants meant disability was
freely, we were able to learn richer de- about my health and have them freak not mentioned or that it was treated
tails about our participants’ identi- out online, than have used up some of neutrally. For example, one 36-year-
ties too. (We recommend other re- the little energy I do have going to meet old female heterosexual with physi-
searchers take a similar flexible someone who I’m not even sure would cal disabilities described her most
approach to help their participants concider [sic] a relationship with me positive encounter online dating: “He
feel better represented.) because of my circumstances.” We asked me about my life and interests
We asked our participants with called this “strategic dys-appearance” instead of my disability.” A 39-year-
disabilities to share their most re- since it is intentional dys-appearance old blind gay male described his
cent, most positive, and most nega- for our disabled participants’ benefit. most positive experience as, “Meet-
tive experiences with online dating Though they found a way to view dis- ing someone that didn’t have issue
and to explain how they disclose their closure positively, it still meant they with my blindness at all.” These ex-
disability to potential dates. Both dis- might not be disclosing disability in periences suggest disability might
abled and nondisabled participants the way they preferred. be more complex than a negative
were asked to share their experiences With these findings, we offered trait best handled by strategically
encountering people with disabilities some design suggestions to improve dys-appearing it through self-dis-
while online dating and to write how disclosure experiences for online dat- closure. But they also suggest posi-
they preferred a disabled match dis- ers with disabilities. Allowing people tive experiences were characterized
close their disability. to filter based on matches’ answers by neutral, not necessarily positive,
Findings and design suggestions. to questions around disability, for treatment of a disability. This invites
Two strong themes emerged about example, may allow disabled online further questions. Was it unimport-
disability-related disclosure. First, daters to filter out potential ableist ant because the disabled partner dis-
most participants, irrespective of matches without explicitly disclosing closed in the way the match preferred?
disability status, felt they should a disability in their profile. Second, Another finding worth further un-
know about a disability right away in line with previous research on us- packing is our disabled online daters
when online dating rather than dis- ing ambiguity in design to encour- found disclosure of physical disabili-
covering it when meeting in person. age open discussion on potentially ties helpful for a practical purpose—
A 27-year-old heterosexual female difficult subjects, like HIV status [5], for planning accommodations. This
shared, “I would prefer physical dis- building ambiguity into design could combined with the neutral experi-
abilities be disclosed in the profile, help disabled daters to disclose more ences marking our disabled partici-
mental disabilities disclosed after a on their terms. For example, if their pants’ most positive encounters begs
few dates.” One difference between answer to disability-related questions the question of whether this neutral
our disabled and nondisabled partic- is “ask me,” disabled daters could or practical preference for disclosure
ipants was the reason for this prefer- tells the whole story. For example,
ence. Most nondisabled participants future work could ask how disabled
considered disclosure of visible dis- people appreciate their disabilities.
abilities a reasonable expectation
they did not need to justify further.
Historic and These questions can help us under-
stand how to design experiences for
In contrast, some disabled partici- present-day online daters that allow them to share
pants wanted people with visible dis-
abilities to disclose so they could pre-
examples what they like about being disabled,
and how to represent disability in
pare accommodations. For example, contribute online dating promotional materials
a 25-year-old bisexual woman with
anxiety and depression mentioned,
to continued that highlight these positive aspects.
“So someone who is deaf, I think it disbelief that DISENTANGLING ASEXUAL

would be important for me to know
that so we can plan accordingly for a
disabled people AND DISABLED
To further expand on our paper’s
date (I don’t know sign language).” can be sexual. findings, there is another potential

thread of future work: how asexual-
ity and disability are entangled. Dis-
“I would prefer represented minorities in their proj-
ects, by releasing accessible products,
ability studies scholar Tom Shake- physical disabilities by thoughtfully designing for unique
speare has historicized disability
and asexuality [6]. For example, in
be disclosed user needs like that of our disabled
online daters to disclose more on
ancient Greek mythology, Hephaes- in the profile, their terms, by designing beyond the
tus, who was disabled, was married
to the goddess Aphrodite. Aphrodite
mental disabilities user interface, and by publishing di-
verse promotional materials.
perceived Hephaestus’ disability as disclosed after
ACKNOWLEDGMENTS
a weakness that took away his “man-
hood,” or sexuality; she cheated on
a few dates.” This article is based on our Computer
him with able-bodied men. Another Supported Cooperative Work and So-
cultural trope is disability is inflict- cial Computing (CSCW) paper [2]. I am
ed as punishment for sin, sin which grateful for my co-authors John Porter
makes the disabled person unfit for (http://www.jrp3.com/), Kiley Sobel
sexual partnership. Today, disabled W3C, Apple, and Android for accessi- (https://students.washington.edu/ksobel/),
people are still infantilized. We are bility guidance), in the case of disabil- Sarah Fox (https://www.sarahfox.info/),
addressed with terms of endearment ity and online dating, ignoring basic and Julie Kientz (https://faculty.wash-
such as, “sweetie,” and group homes accessibility in product development ington.edu/jkientz/) for leading this
and institutions housing disabled feeds into the historic entanglement project and offering feedback on
adults place restrictions on their of disabled and asexual by not includ- drafts of this article.
whereabouts and activities. These ing some disabled people as possible
historic and present-day examples users of online dating services in the References
contribute to continued disbelief first place. [1] Parisi, K. I Told My Tinder Matches about My Disability
– And Got the Worst Reaction. Refinery 29. Feb. 11,
that disabled people can be sexual. Similar to opening up disclosure 2016; http://www.refinery29.com/disabled-tinder
Recall the woman I referenced earlier methods, online dating platforms can [2] Porter, J. R., Sobel, K., Fox, S. E., Bennett, C. L.,
who is regularly asked on Tinder how broaden what types of people we con- and J. A. Kientz. 2017. Filtered out: Disability
disclosure practices in online dating communities.
she has intimate relationships. sider datable. Initial improvements In Proceedings of the ACM on Human-Computer
Online dating platforms can help could come in the form of adhering to Interaction 1, 87. 2 (November 2017). DOI: https://
doi.org/10.1145/3134722
decouple this inaccurate association established accessibility guidelines, [3] United States Census Bureau. Nearly 1 in 5 people
between disability and asexuality. but further exposure could occur by have a disability in the U.S., Census Bureau reports.
Press release. July 25, 2012; https://www.census.
The first step is access. Our respon- increasing representation of disabled gov/newsroom/releases/archives/miscellaneous/
dents and media narratives provided daters in ads and popular media. cb12-134.html
experiences of online dating. What is [4] Paterson, K. and B. Hughes. Disability studies
and phenomenology: The carnal politics
missing are people who do not have CONCLUSION of everyday life. Disability & Society 14, 5
these experiences, possibly because Our project at the intersection of (September 1999), 597-610. DOI: http://dx.doi.
org/10.1080/09687599925966
they have tried and cannot access online dating and disability has
[5] Handel, M. J. and I. Shklovski. Disclosure, ambiguity
online dating platforms, or because emerged with exciting possibili- and risk reduction in real-time dating sites. In
personal experiences and these very ties for designing online dating ex- Proceedings of the 17th ACM International Conference
on Supporting Group Work (GROUP ‘12). ACM,
narratives shy them away from at- periences to be more inclusive. We New York, 2012, 175-178. DOI: http://dx.doi.
tempting to date online. For example, showed how social norms around org/10.1145/2389176.2389203
my attempts to date online are quite disclosing disabilities, based on [6] Quarmby, K. Sex, lives, and disability –What can
disabled bodies teach us about sex, and why
few, but not for the reason most would whether they are visible in-person, should we listen? Mosaic. March 3, 2015; https://
expect. As a blind woman, I imagined extend to online dating communi- mosaicscience.com/story/sex-disability
the most difficult part of matching ties. Second, we introduced strate-

would be understanding what poten- gic dys-appearance, or extra work
Biography
tial dates look like. However, I have done by participants who disclose
Cynthia L. Bennett is a fourth year Ph.D. student in the
never had difficulty recruiting friends their disabilities to filter out ableist Human Centered Design and Engineering department at
to describe potential dates over a matches. I suggested two future re- the University of Washington. While much accessibility
research focuses on creating accessible technology
glass of wine. Instead, I have encoun- search directions around online dat- end products, her research instead focuses on how
tered significant barriers to engaging—first, to explore positive aspects to make the design process itself more accessible for
blind and low vision people. She is a fierce proponent of
ing with text profiles and messaging of disability, and second, design op- bringing disability studies in conversation with accessible
portunities to push back on historic computing research, as it is the only scholarly field built
features of online dating apps with by disabled people. Visit her personal website at https://
a screen reader, and I have yet to go and inaccurate assumptions that all www.bennettc.com/
on a date with someone I met online. people with disabilities are asexual.
While following established accessi- We hope this project inspires re-
bility standards for making websites searchers and developers to foster
and apps accessible is sometimes positive experiences for all users by Publication rights licensed to ACM
unintentionally overlooked (Search centering disabled and other under- 1528-4972/17/12 $15.00

feature
Leveraging Personal
Experience for
Academic Research
and Outreach
Use your individuality to build your career path whether it leans
toward academia, outreach, or both. The existing underlying threads
between your experiences and the pursuit of research problems
might surprise you.
By Joslenne Peña
DOI: 10.1145/3155122
I
recently finished pursing a research project called “Evaluation of a Technology
Education Pipeline Project” with my advisor and colleagues. The purpose of this
project was to develop and analyze an online community built for the iTech Academy
Summer Camps, which took place at Penn State’s College of IST campus. My broader
inspiration for this project stemmed from my personal background. I am a first-generation
Hispanic college student, from a low-income family. I wanted to give younger students the
opportunities I never had, ranging from available software and technology to concepts.
The iTech Academy offers four different week-long camps for high-school and middle-school
children. The camps aim to broaden interest and participation in computer technology
careers. Our research goal was to un- I taught web development and robot- scholarships, grants, and other fund-
derstand the effectiveness of the iTech ics courses to local high-school and ing opportunities. I always viewed
Academy camps and related online middle-school students in the area. my demographical background and
community, and to concentrate the Their devotion and ambition to learn, my general identity as a detriment.
camps’ efforts in exposing students especially that of younger female stu- For me, surpassing the norm was
to STEM disciplines earlier in their dents, truly amazed me. This research abnormal. Pursuing a four-year de-
education careers. We performed an brought me back to the beginning of gree was unfamiliar. Now imagine,
online community analysis, and as- my academic career. an advanced degree? Certainly, out
sessed pre-and-post information that of the ordinary. It is was no surprise
we acquired through surveys. (See Ar- NEVER UNDERESTIMATE that when I told my mother I wanted
Image by Xi Xin Xing
itajati et al. for an early paper on our YOUR PERSONAL EXPERIENCE to continue schooling she was con-
work [1].) While I was mostly involved The process of checking boxes un- cerned. This was more uncharted
as a researcher, I also had the oppor- der certain categories is something I territory. Was it not enough that I un-
tunity to participate as an instructor. am accustomed to when applying for derwent four years of undergraduate

Photo Credit TK

feature
topics of interest. While I enjoyed my discrimination perpetuated by tech-

computer science studies very much, nology access. It was distressing and
this made high school harder for me intimidating to say the least. My high-
over time. school experience left me with the im-
Our school resided in an urban en- pression that education provides a very
vironment where other uncontrollable limited world of possibilities, whereas
factors contributed to disinterest in college exposed me to a broad range of
school, such as troublesome behavior, possibilities and illuminated the world
underpaid instructors, deteriorating of education in a substantially more
building conditions, lack of tutors, and positive light. Having said that, it has
more. With all these challenges, it was been crucial for me to still keep a criti-
difficult to stay engaged in a less than cal lens as I navigate through my world.
rigorous curriculum and classroom With my life experience, I have gained
environment. Thinking back, I still do the intellectual maturity to identify
not know whether to consider this un- and comprehend societal difficulties
intentional discrimination. Being dis- within educational institutions.
advantaged because of your location During the summers of 2011 and
or background is quite disheartening. 2012, I would go back home to New
Having an unbalanced educational York City to teach at a technology camp
education unscathed? For me, it was system damages our society and reduc- called iD Tech Camps, which was held
not. My appetite for learning and ana- es ones chances of acceptance to aca- at Columbia University. I taught sev-
lytical thinking had only just begun. demic institutions. I knew attending eral computing topics to children
I remember my first exposure to college was my next step, but I felt inse- between the ages of seven and 13. I
technology. I was seven years old, and cure about my abilities. I felt the lack of taught Scratch programming, game
my mother took me along on a “bring access to state of the art technology in- design, web design, and more. I held
your kid to work” day. At the time, she troduced barriers to my journey. I was review sessions, designed lessons, and
worked at a small financial company. I already at a disadvantage without even built a fun learning environment for
remember walking around her office knowing it. the children. I figured out quickly that
in utter amazement. I remember the When I started college, I finally I enjoy working with students and I
big and bulky IBM machines that en- gained access to modern tools and an have fun teaching. While challenging,
compassed most offices in the 1990s. improved curriculum. I could study it was also rewarding to see their prog-
From then on, I knew I was going to under experienced instructors and ress and watch them present their final
work in a field involved with technolo- practitioners, who offered thoughtful projects to their parents. This experi-
gy. Little did I know, regardless of how guidance. However, when I walked into ence taught me that this was just one
trivial, this moment would shape my my introductory level courses I noticed of many ways of giving back to my com-
future endeavors. women and underrepresented minori- munity. I became thoroughly inter-
My parents did their best to keep up ties were missing in STEM fields. I also ested in pursuing more opportunities
with technological advancements like noticed diversity was a controversial to give students a chance to obtain the
gaming consoles and new desktops. I issue, one I was blissfully unaware of knowledge they deserved at a younger
remember playing around with Win- growing up in the big melting pot of age.
dows 95 and AOL, but it was not until New York City. I strongly believe the Graduate school was no different,
high school that I started program- apparent advantage other classmates in the sense that underrepresented
ming and dabbling in computer sci- had in their knowledge and skills minorities were just as scarce, if not
ence. I attended a New York City public stemmed from the disparity in educa- completely absent. As a graduate stu-
high school where you had to choose tional systems in the U.S., especially dent, I have been fortunate enough to
a technical or business major; I chose merge two fields of interest together:
computer science. As a teen girl, hav- my main research area, human-com-
ing an interest in science, technology, puter interaction, and my community
engineering, and mathematics (STEM) outreach efforts. Moving through a
was somewhat unusual. Coupled with In fact, the most difficult public-school education sys-
the fact that my high school lacked
resources and did not instill the disci-
inconsequential tem and volunteering led me to realize
resources are still not readily available
plinary behaviors needed for academic exposure to to many students who are underrep-
Image by Thiranun Kunatum
excellence, it was easy to draw your

influence from those not focusing on
something resented minorities. I was captivated
with conducting research partly de-
their studies and to develop less than can become voted to empowering people from a
average work ethics. For example, we
did not have the latest computers,
monumental disadvantaged background, and allow-
ing them to choose their path and not
software, or textbooks to keep up with later in life. be hindered by resources (technical or

not). I wanted to build technologies
that improve engagement with STEM-
Having an for the intended teaching material, es-
pecially with technologies that can au-
based disciplines, focusing on K–12. unbalanced tomate simple tasks to free them up for
Students at a later age often say, “I
wish I had this experience when I was
educational system lectures, activities, and more. It is also
important to recognize the potential
younger!” and this was the exact feel- damages our in those who will eventually become
ing I had in college. I took these experi-
ences and ran with them. I merged the
society and reduces the leaders of our society—the stu-
dents. I would love to see a classroom
idea of summer camp research with ones chances that nurtures failures and mistakes,
my passion for broadening participa-
tion for underrepresented minorities
of acceptance while having a hands-on approach
to learning with technology. If these
in STEM, and this led me to the iTech to academic wishes turned into reality, the impact
Academy research.
Another side of this issue arose
institutions. would be incredible. It would mean our
educational system is finally reaching
from a different audience. Education equality in accessibility, regardless of
research gives a lot of attention to the location or funding. This would give
student population, and rightfully students from underprivileged back-
so. However, I was also interested in their usage implications. grounds a shot to obtain the same op-
exploring the instructor population, All of this is to say that one experi- portunities while receiving a better
because understanding their perspec- ence can lead to many. One can never education. I do want to note simply
tive, needs, and goals is imperative to be sure which experience may lead to using technology does not guarantee a
student success. Instructors should be a certain choice or path. In fact, the better education. It takes collaborative
invested in early exposure to STEM dis- most inconsequential exposure to work with the students and teachers to
ciplines, and nurture a flexible skill set something can become monumental achieve proper, personalized learning.
that adapts to changes in technology. later in life. My drive and passion for However, at a high level, integrating
There is no doubt the education para- research, as well as helping those from technology has the incredible ability
digm is changing fast. Students and in- my background, began to integrate to improve the educational experience
structors must keep up with new tech- into my work. I am very happy that I for both teachers and students.
nologies and teaching approaches. can better understand the issues of If there is a single takeaway from
Thus, it is crucial instructors endorse such a complex topic, and I can make this article, it is to urge people to ac-
student engagement, participation, a difference if it comes to that. One cept their individuality and leverage it
and collaboration while learning new should embrace and learn from prior for their future efforts in local commu-
things. With this in mind, I started in- experiences, because the journey to nities, academia, and beyond.
vestigating how to build technologies come is unpredictable at best.
for instructors, to make them end-user References
developers and help them retain stu- CONCLUSION [1] Aritajati, C., Rosson, M. B., Pena, J., Cinque, D., and
Segura, A. A Socio-cognitive analysis of summer
dents in STEM disciplines. Research I hope to see future educational tech- camp outcomes and experiences. In Proceedings
shows a loss in interest is due to dis- nologists, researchers, and designers of the 46th ACM Technical Symposium on Computer
Science Education. ACM, New York, 2015, 581–586.
engagement or a lack of comprehen- practice creative freedom and incor- [2] Master, A., Cheryan, S., and Meltzoff, A. N. Computing
sion, which is typically the instructors’ porate personal experience into their whether she belongs: Stereotypes undermine girls’
interest and sense of belonging in computer science.
responsibility [2, 3], This endeavor fol- work. I challenge them to reflect on Journal of Educational Psychology 108, 3 (2016),
lows a more indirect approach than di- past moments and perhaps utilize 424–437.
rectly observing students. I discovered them as motivation, or even use them [3] Wang, J., Hong, H., Ravitz, J., and Ivory, M. Gender
differences in factors influencing pursuit of
some interesting findings regarding to materialize a research project. Fur- computer science and related fields. In Proceedings
communities of practice within the thermore, I would love to see students of the 2015 ACM Conference on Innovation and
Technology in Computer Science Education. ACM,
instructor ecosystem [4]. For instance, and teachers with more access to mod- New York, 2015, 117–122.
many instructors are inclined to learn- ern educational tools for learning, es- [4] Peña, J., Shih, P. C., and Rosson, M. B. Scenario-
ing new teaching methods and tech- pecially those in urban or disadvan- based design of technology to support teaching in
inverted classes. IConference 2016 Proceedings.
nologies. They realize this is impera- taged communities. Serious progress iSchools, Grandville, MI, 2016.
tive not only for their skill set, but also has been made, but there is much
to engage and retain students. In the more to be done. Biography
first phase of the research, I explored It important to acknowledge the in- Joslenne Peña is a Ph.D. candidate in the College of
Information Sciences and Technology at Penn State-
their views on teaching practices and dividuals who overcome uncontrolla- University Park. She is also under assignment as a
the role of technology. The next steps ble obstacles to enable learning in the research associate in the Human Centered Systems
group at Honeywell Aerospace. Her research interests are
involved observing interactions in the classroom; they should receive the nec- human-computer interaction, human factors psychology,
classroom with students in STEM dis- essary tools required to enhance their and computer science education. She holds diversity and
outreach near and dear to her heart.
ciplines, and understanding the re- methods of teaching. Teachers should
lationships in the classroom, as well have the freedom to select the technol-
as the technologies that are used and ogies they believe are most appropriate © 2017 ACM 1528-4972/17/12 $15.00

feature
Gender and
the Art of
Community
Relations
Aspects of one’s personal identity can change the way you experience
being part of a community, especially if you are in a minority group.
The author reports on her experiences of conducting research with
women who participate in the Debian Linux project.
By Lesley Mitchell
DOI: 10.1145/3155124
A
fter nearly two decades working as a professional systems administrator, I took a
career break to complete my dissertation in philosophy. As it turned out, I wasn’t
done with academia when I handed it in. My dissertation had taken a dry and
theoretical look at the clash between traditional theories of personhood and
personal identity in political philosophy, which are deeply rooted in the ideas of men writing
in the 17th and 18th centuries, with those stemming from the work of modern feminist
philosophers from the 1960s on- What caught my imagination was My research, and subsequent disser-
wards. I realized the practical upshot the different experiences people have tation, explored the impact of gender
of these theoretical tensions was play- within groups based on their visible on the experiences of women in geek
ing out across almost every area of my and invisible identity traits. That is to communities. I chose to focus on the is-
life, but most particularly within geek say, the things other people use to de- sue of gender because it is the area most
communities. This prompted me to fine, or pigeonhole, you. These include familiar to me from my own personal
change my focus. My studies in sociol- things like skin color, gender markers experiences, and where I had the most
ogy allowed me to research the effects like body shape or facial hair, or the relevant previous academic training.
of identity, particularly gender iden- use of a wheelchair or hearing aids. However, there exist many other iden-
Images by Ollyy / Shutterstock.com
tity, within a community that I know Also relevant are other more subtle tities that can lead to marginalization
and love—the Debian Linux project. cues, such as accent and dialect, which within these communities. Further-
The process not only gave me the op- may be used to label the speaker in more, while Debian is a global project
portunity to gain insights into the ex- terms of class, race, or native tongue. and has participants from around the
periences of the women I interviewed, While mode of dress and the amount of world, it functions primarily as an An-
but also allowed and encouraged me gestures used when speaking may be glophone community and participa-
to reflect on my own relationship with seen as indicative of particular genders tion is somewhat skewed toward North
gender and geek culture. or sexualities. America and Europe. Thus the social

Photo Credit TK

feature
norms of the group tend to be those of sonal identity, rather than imposing the the EU] women are under-represented
modern Western society as found in term as an external label on people who relative to men, making up less than
those regions. are unwilling to claim it. One problem 45 percent of scientists and engineers”
with this approach comes from clashes [1]. It’s harder to get reliable figures for
GENDER AND GEEK CULTURE between the stereotypical image of a F/LOSS communities and the fan bases
My interpretation of the term “geek” is geek in modern Western culture, and of geek media. One area in which the
a positive and inclusive one. That is, any the negative connotations that it still demographics are noticeably differ-
person who is passionately interested retains in some social groups, with oth- ent from this male-dominated trend is
and committed to learning a signifi- er aspects of one’s identity. This leads video gaming, where the annual studies
cant amount about a particular topic. some people to reject the identity or produced by the Entertainment Soft-
For me, people can be geeks about any label of geek, while otherwise partak- ware Association show the gender split
topic, be it music, sports, or knitting. ing in the interests and actions that the of gamers is now approximately 50:50.
However, in order to more closely align identity implies. Women over 18 are a significantly larger
with the common usage of the term, for An online image search for the term proportion of the market than teenage
my dissertation and this article, I limit geek produces pictures and cartoons of boys, who tend to be expected to be the
the term to two broad groups: firstly, bespectacled, white, young-ish males main consumers [2].
those involved with computers, com- using pieces of technology with unfash- I knew growing up that, while none
puter science and information technol- ionable, and even outlandish, hairstyles of my friends shared my interests, there
ogy, and science and engineering; and, and clothing. Much further down the were other geeks in the world, even if
secondly, the creators and fans of “geek results, one may find the occasional im- none of them seemed to look like me.
media,” such as comic books, graphic age of a man of color or a white woman. I even had some understanding that
novels, science fiction, fantasy, horror Contemporary media provides further groups of geeks came together to form
and related genres in all formats. The examples: Sheldon Cooper from “The communities, even before this was eas-
first group might include, for example, Big Bang Theory,” Eliot Alderson from ily facilitated by the rise of the inter-
people working in STEM roles, makers “Mr. Robot,” or Harold Finch from “Per- net and the web. I knew too, because
and hackers, and Free/Libre and Open son of Interest.” It should, however, be all the portrayals of geeks that I saw
Source Software (F/LOSS) developers noted, unlike earlier examples of this always included references to a larger
and users. The second might include genre, men of color and white women community. I was fortunate not to be
authors and other creatives involved in can be found playing supporting geek tormented by my peers for my outland-
the production of geek media, as well roles on these television shows. ish interests, though I was certainly
as cosplayers and other convention at- Where demographic data is avail- considered odd in the all-girls high
tendees, fan fiction writers, and so on. able this stereotype is reinforced. For school that I attended. It was at univer-
I have been many types of geek in example, several major U.S.-based sity where I first found myself in a male-
my life. I got my first computer for my technology companies—including Mi- dominated environment, one of only a
eighth birthday, and, almost imme- crosoft, Apple, Google, Twitter, and handful of women in a course of more
diately, started to teach myself how to Facebook—have made this data pub- than a hundred. However, it was also at
code. My reading list has leaned toward licly available. In all cases the technical university that I first connected with a
science fiction and fantasy novels ever workforce is predominantly male, with geek community. I not only discovered
since. I grew up watching “Battlestar four or more men for every woman. mailing lists dedicated to TV shows,
Galactica” (the original!), “Star Trek: While the 2015 “She Figures” study, and chat rooms and message boards
The Next Generation,” “Twin Peaks,” funded by the European Union, found filled with like-minded people. I also
“The X Files” and “Stargate SG-1.” I’ve “in more than half of the countries [in found a group of people who used and
been to and run conventions for “Star- programmed the same type of comput-
gate,” written fanfic, dressed up in cos- er that I did and would regularly meet
tume, met the cast, and been on the up in person to talk about “all things
set, too. My undergraduate degree is Acorn.” (Acorn was a British designer
in chemistry. I have worked in a techni- It is easier to hide and producer of a series of computers
cal role for general businesses and at
highly specialized IT firms. I have al-
indicators of one’s in the ‘80s and ‘90s, which were par-
ticularly dominant in the educational
ways claimed, and to some extent taken gender online market, but also as home computers.
pride, in my geek identity. However, not
every person who fits into the categories
behind masculine These were some of the first devices to
use the now ubiquitous ARM RISC pro-
I have mentioned necessarily consid- or ungendered cessors and, combined with the RISC
ers themselves to be a geek, and many
who do embrace the geek identity, like
nicknames and OS operating system. Acorn spawned a
dedicated following of users and devel-
me, are participants in many different avatars, than it is opers. Some of this community contin-
geek activities. I prefer to consider only
those people who choose to incorporate
when meeting face- ues to exist even now, nearly 20 years on
from the cancellation of the final Acorn
the idea of being a geek into their per- to-face. branded computer.)

FREE AND OPEN FOR WHOM?
One reason I chose to use the Debian
They all agreed rape from women working and playing
in geek communities, it took me by sur-
Linux project as the population for my there was a steady prise. So many women had experienced
study was that I had been a peripheral
part of the community, for both social
stream of smaller geek communities in much less positive
ways than I had. While I had suspected
and professional reasons, for many actions on a daily that I had been treated differently than
years. This provided me with connec-
tions who would be prepared to vouch
basis that was simply my male friends by some employers,
when I looked for work, and by male co-
for me to potential participants. A sec- tiring to deal with, workers when I was in work, it had never
ond reason was previous research had
been performed with the community.
wearing one down really been at a level that was sufficient
for me to fully recognize it as a struc-
Both qualitative anthropological stud- as a trickle of water tural difference in attitudes toward me
ies (such as that conducted by Gabriella
Coleman and documented in her book,
wears down a rock. and my skills as a female technologist.
This was partially because I was entirely
Coding Freedom: The Ethics and Aesthet- oblivious. It had never occurred to me
ics of Hacking [3]) and quantitative de- to judge people’s technical abilities
mographic investigations (for example, based on their gender and, thus, it nev-
the research published in “Women in er really crossed my mind that others
Debian 2013” [4]), provided me with built via these methods proceeded in would do this to me. I suspect several
useful background data, including an a different way given that it is easier things helped to keep these issues to a
estimate of the gender distribution of to hide indicators of one’s gender on- minimum for me, even though I have
the project at large. At the time of my line behind masculine or ungendered never made any deliberate effort to hide
study, less than 20 of more than 1,000 nicknames and avatars, than it is when or obfuscate my gender. I have, for ex-
registered Debian developers identi- meeting face-to-face. Previous research ample, always quickly pointed out when
fied themselves as female, while a very has suggested the greater anonym- I have been inadvertently misgendered.
slightly larger number were estimated ity possible when interacting using My first name, which has been used by
to be active in the wider community. De- computer-mediated methods can have several generations of my ancestors on
spite this, there has been an active sub- both positive and negative side effects both sides of my family, is one that is not
project of Debian Women (https://www. on the way communication progresses immediately obviously gendered, espe-
debian.org/women/) whose stated aims between participants. Secondly, I also cially when spoken. The name I chose
have included trying to increase the hoped to explore whether, and in what to use to define my online personality in
number of female developers and active way, the introduction of an in-person the mid-‘90s and still use today, is only
community members. Furthermore, component—by virtue of an in-person obviously gendered if you recognize
the community voted to endorse not encounter in the actual world—to a re- the pop-culture reference from which
only a code of conduct 1, but also a diver- lationship initially built virtually pro- it is derived. Much later, when pictures
sity statement.2 My hope, therefore, was duced any noticeable changes to the became all the rage, I picked a non-hu-
that I would gain an understanding of relationship dynamics. Finally, I was in- man avatar that, nevertheless, quickly
not only any negative aspects of being terested to discover if gender presenta- became associated with me. Further-
a minority member of a significantly tion, for example for women who dress more, it has always been easy for me to
gender skewed population, but also the in a particularly “masculine” or “femi- be considered “one of the boys” since I
positive. nine” style, produced different experi- have never been “feminine” in my cloth-
There were three secondary areas I ences of the community. Sadly, I didn’t ing, speech or general persona, nor am
hoped to explore. Firstly, Debian is pri- really gain any insight on any of these I heterosexual. Nonetheless, although I
marily a virtual community. That is to points, as the experiences shared by my have been fortunate to avoid the experi-
say members mostly communicate and participants didn’t lend themselves to ence of sexual harassment and assaults
maintain the community via computer- analysis of these ideas. that some of my female peers have had
mediated methods, such as mailing It is important to note, I was not look- to deal with at work, at conferences, and
lists, forums, and real-time text based ing to produce generalizable results at conventions, I have certainly experi-
chat. In-person events in the actual from this research, nor did I expect to enced being called “bossy” or “aggres-
world do occur, including, but not lim- be able to generate clear answers to any sive” when similar behavior by my male
ited to, the annual DebConf conference of the big questions about the recruit- colleagues has been styled as “asser-
held in a different location around the ment and retention of women in this tive” and “go-getting.” I have also seen
world each year to improve accessibil- sort of community. My aim was solely to my career stall where less well qualified
ity for all members of the community. capture and report the individual expe- or experienced men prospered.
I was, therefore, interested to see if I riences of each of the women to whom In the summer of 2016, five women
could learn whether the relationships I spoke. from the global Debian community, in-
When I first started to come across cluding one transgender woman, made
1 https://www.debian.org/code_of_conduct first-hand accounts of discrimination, themselves available to me to be inter-
2 https://www.debian.org/intro/diversity sexual harassment, abuse, and even viewed by video call. These women were

feature
recruited either having answered my

call for participation, which was issued
It had never occurred and in some cases decades. I was also
able to cement new relationships, in-
by email on the Debian Women mailing to me to judge cluding with some of my participants,
list, were known to me personally in ad-
vance, or were recruited at the 2016 Deb-
people’s technical and make new connections within
the community. Also as a result of this
Conf in South Africa by another partici- abilities based on trip, I have agreed to become involved
pant. More women and one non-binary
person expressed an interest in partici-
their gender and, in the new diversity and inclusion ini-
tiative, which aims to work alongside
pating, however, they were unable to fol- thus, it never really groups, such as Debian Women. The
low through on this for various reasons
including scheduling conflicts, my use
crossed my mind goal is to provide support, includ-
ing advice on inclusive language, to
of proprietary video conferencing soft- that others would do other teams, such as those set up to
ware, and lack of confidence in being
able to express themselves in English.
this to me. welcome to new members and main-
tain the website. I was able to attend
Sadly the resource limitations, both in the conference as a recipient of one of
time and money, of being a master’s travel bursaries provided by the Debi-
student meant I was unable to travel to an project, with the stated intention
interview people face-to-face. of increasing the accessibility of the
I conducted semi-structured quali- conference to those from under-rep-
tative interviews with each woman, dent, they all agreed there was a steady resented groups including minority
each of which was recorded. Using this stream of smaller actions on a daily ba- genders and sexualities, older people,
method, the interview proceeds like a sis that was simply tiring to deal with, and those from countries outside of
formalized conversation, in which the wearing one down as a trickle of water Western Europe and North America.
interviewer has a small number of ques- wears down a rock. As a sociologist, I I believe there is still more work to
tions to guide the participant’s initial describe these actions as “microagres- be done understanding the experienc-
thoughts, but beyond that says little and sions,” a term coined by Harvard pro- es of minority groups in both the STEM
mostly allows the participant to control fessor Chester M. Pierce to describe the workforce and in F/LOSS communities.
what they do or do not choose to reveal. small insults and dismissals that he Only by genuinely listening to the peo-
Each interview was then transcribed to saw daily inflicted by white Americans ple affected, will we be able to generate
text as closely as possible to how it had toward people of color. It also often strategies to improve current commu-
been said, including hesitations and seemed that the structures of Debian nities and build better, more inclusive,
pauses. This was an exceptionally time were designed without consideration of ones in the future. Perhaps, in fixing
consuming part of the research proc- the different roles and responsibilities the issues we identify in our communi-
ess, as it was my first time producing women tended to have outside of the ties, we can also show the way to society
transcriptions. Not only were several project, which often made it harder to at large.
participants non-native English speak- achieve one’s goals and easier to incur
ers, sometimes the clarity of the audio further microagressions. In contrast to References
was poor due to the variable nature of this, and in line with Foucault’s propos- [1] European Commission, She Figures 2015.
Publications Office of the European Union,
internet connections. For example, one al of the connectedness of power rela- Luxembourg, 2016.
participant’s home bandwidth was not tions and resistance, these women built [2] Essential Facts About the Computer and Video Game
sufficient for video calls, so she con- their own support and mentoring net- Industry. Entertainment Software Association.
2017; http://www.theesa.com/article/2017-
nected from a local coffee shop instead, works. This was within Debian in the essential-facts-computer-video-game-industry/
with all the background noise that im- form of the Debian Women sub-project, [3] Coleman, G.Coding Freedom: The Ethics and
Aesthetics of Hacking. Princeton University Press,
plies. When complete, the transcripts but also outward to other women in New Jersey, 2013.
were analyzed. I tried to see if any com- different technology and open source [4] Arjona, L., Langasek, P., and Ruiz, M. Women
mon themes appeared and whether communities experiencing similar in Debian 2013. DebConf13. August 2013;
http://penta.debconf.org/dc13_schedule/
similar experiences were apparent, problems. attachments/243_women_in_debian_2013.pdf
either related to the areas I described
previously or ones the participants had LOOKING FORWARD Biography
raised themselves. As I had promised the women who Despite dreams of being a marine biologist or an
All my participants were positive participated in my research, I was in astronaut, or possibly both, during her childhood,
Lesley Mitchell studied chemistry and then became a
about their overall experiences with the attendance at DebConf 17 in Mon- career systems administrator. Later, while looking for
a way to stretch her mind further, she stumbled into
Debian project, frequently bringing up treal. I gave a talk explaining my re- the study of philosophy and then sociology, focusing
the good friendships and other rela- search, and had a number of fruitful on the relationships between individuals, identity, and
communities. In her spare time, she has been known
tionships they had found and built in discussions with community mem- to help run science fiction conventions, cook elaborate
the community. There was, however, a bers around these topics. The trip also meals, take copious photographs of inanimate object,
and wave a sword about. She is inseparable from both her
darker side. While only one participant allowed me to meet, hug, and eat with small, black cat and her hat.
suggested she had been on the receiv- friends with whom I speak with on a
ing end of a serious gender-based inci- daily basis, but hadn’t seen for years © 2017 ACM 1528-4972/17/12 $15.00

Attention: Undergraduate and Graduate
Computing Students
There’s an ACM Student Research Competition (SRC)
at a SIG Conference of interest to you!
It’s hard to put the ACM Student Research Competition experience into words, but we’ll try…
“The ACM SRC gave me an amazing opportunity to present “It is a wonderful experience to attend the ACM SRC, where you
my research at the first conference I ever attended. I can collect insightful feedback, shape your presentation skills
met with talented students, discovered their work, and and make outstanding peer friends. I will definitely appreciate
discussed interesting ideas with them. The feedback I this learning experience throughout my professional career.”
received from expert researchers allowed me to place my Xiaoqing Xu
work in the broader picture and it helped me steer my University of Texas at Austin | ICCAD 2016
research accordingly. It was a great experience, and I would
recommend it to any young researcher.”
“It was a fantastic and memorable experience to participate
Lisa Nguyen Quang Do in SRC. Writing the abstract, preparing the poster, and
Fraunhofer IEM | PLDI 2016
presenting your ideas to the audience and convincing them is
challenging but rewarding and builds up your confidence and
“The experience of pitching your own research and ideas, communication skills. SRC is a really nice event where you can
regardless of how bold they might be, is an important stage of meet the experts from various fields, exchange your ideas,
each young researcher’s career. SRC was one such opportunity. and gain valuable feedback on your research. Go, SRC!”
It allowed me to present my research topic to a highly skilled Ting Su
panel of researchers and educators, and provided me with East China Normal University, China | ICSE 2016
significant feedback and discussion regarding my work.”
MHD Yamen Saraiji “The ACM SRC was a great opportunity to discuss my research
Keio University | SIGGRAPH 2016
with experts in my area. The constructive feedback I received
gave me new ideas and pushed my thesis work forward. It
“The ACM SRC was an incredible chance to start a conversation was also an invaluable experience in terms of practicing
with various experts in technology. I not only gained valuable presentation and communication skills that would serve me
feedback on my work, but I was also able to learn more about well in my future career.”
my future in research: the process of applying to grad school, Arik Hadas
the best and worst aspects of research, and more.” Open University of Israel | Modularity 2016
Jess Cherayil
Wellesley College | SPLASH 2016
Check the SRC Submission Dates:
“The SRC was a very enriching and enjoyable experience. http://src.acm.org/submissions.html
The insights and constructive feedback I received from judges
and attendees broadened my understanding in the field and
encouraged me to continue working on the topic.”
Mirko Gelsomini
Politecnico di Milano | ASSETS 2016
u Participants receive: $500 (USD) travel expenses

u All Winners receive a medal and monetary award. First place winners advance to the SRC Grand Finals
u Grand Finals Winners receive a handsome certificate and monetary award at the ACM Awards Banquet
Questions? Contact Nanette Hernandez, ACM’s SRC Coordinator: hernandez@hq.acm.org
src-2017-cacm.indd 1 7/6/17 4:04 PM

feature
Reflecting on
Robots, Love,
and Poetry
Finding the poetry in programming and
the algorithms in poems
By Margaret Rhee
DOI: 10.1145/3155126
A
s the poet William Carlos Williams writes, “A poem is a small (or large) machine
made of words” [1]. The intersection of poetry with machines is provoking on
interaction. The conversation, as ephemeral as it may be, is a human-human
interaction also conducive to consider. As we contemplate how technology informs
human-human interactions, it may be useful to reflect on human-human interaction and
engagement with literature, namely with words. The cybernectics of these kinds of relations
never stops short of being astonishing.
I write science fictional poetry about be human. The robot offers us so many larities and slippage between the two—
robots, and my first poetry book Love, questions to grapple with, in our urgen- poetry and AI—I hoped would frame
Robot was published this fall [2]. Writing cy for humanity, compassion, empathy, the collection and urge the exploration
about robots poetically sprung from a and justice. Could we think of robots as of poetry, love, and AI. Without being
conversation when I was a graduate stu- entities that evoke more compassionate engaged with robots and technology
dent. At the time, I worked with several characteristics than humans? Could we as a cultural history, my poetry would
roboticist-artists, namely Ken Goldberg think of robots as object-choices of de- have lacked the grounding and the lack
who served on my dissertation com- sire, or more compassionate lovers? of imagination that AI theory, computer
mittee. I also worked with Eric Paulos Research then inspired my poetry. science, and other fields could offer.
on tangible media. While my scholarly For example, my poetry on robot love Graduate study is a time of deep
work centers on culture and difference draws upon Alan Turing’s question, thinking, training, and development. It
as an analytical and historical project, “Can machines think?” As a poet, I is also a time for conversations. When I
their work and others in robotics and switched the “think” to “love.” In addi- reflect back on my graduate experience,
computer science opened up my artistic tion, there is the question of the son- I fondly remember the importance of
explorations. Currently, I am a visiting net and artificial intelligence (AI) that conversations that are paradisciplinary.
assistant professor in media study. As Turing poses. Could a machine write Not only across the humanities and so-
someone studying media and litera- poetry? Would it then prove humanity? cial sciences, but the sciences as well.
ture, I cannot ignore the current digital On the other hand, I was researching Not conversations that begin by talking,
age as a societal phenomenon. poetics and learned about how we can but with questions. Not the performance
Philosophically, I am interested in think of poetic forms, such as the son- of knowledge, but perhaps the opposite.
how robots teach us what it means to net, as machines. The uncanny simi- The desire to develop by engaging.

In this poem “Sleep, Robot,” I imagined the circumstance of a robot and a human in love, and the daily pattern of falling asleep
together. I wondered how a robot and human would “sleep” together when a robot has no need for sleep. What if the human
Photo Credit TK
lies awake with the robot? What does it mean for a robot to be run down enough to “sleep” or even be replaced? Many of
the poems in the first section of the book are accompanied by algorithm poems that also speak to the circumstances of the
poem through code. I hope readers can think about algorithm as language, and poetics as programming.

feature
As a graduate student, I sat next to graduate training, but beyond it, mak-
INTER ACTIONS a postdoctoral roboticist. Since I was ing your own work unique and a contri-
a poet, I had various poetry related fli- bution to our larger world.
ers, books, and posters around. My new Friedrich Kittler surveyed how the
neighbor asked me, upon seeing the gramophone, film, and typewriter
photograph I had of Langston Hughes— transformed language and writing
the celebrated African American poet during the 19th century. I want to re-
of the Harlem Renaissance—who was flexively ask, how in our current digital
in the photograph. I began telling him, age of ephemeral spaces foster the ex-
when he kindly responded, “I know who change and development of intellectual
he is. I write poetry too.” He not only and creative thought/acts? Exploring
created robots, he wrote poetry, and robot poetics for me, has led to ques-
even took graduate courses in poetry. tions around the poem and machine
ACM’s Interactions magazine I was studying robots, but I was not yet intersectionally.
explores critical relationships writing poetry about robots, until my In going back to Turing, and his test
between people and new friend offered another way of hy- on artificial intelligence. His article,
technology, showcasing bridity. Together, we wrote poetry and “Computing Machinery and Intelli-
emerging innovations and exchanged poems, remembering the gence” [3] refers to the sonnet as an in-
industry leaders from around humanities and science divides are ar- dication of the human or the machine:
the world across important bitrary, or that transgression should Q: Please write me a sonnet about the
applications of design thinking and can occur. subject of the Fourth Bridge.
and the broadening field of At the time, prompted by our con- A: Count me out on this one. I could
interaction design. versation, I began to find poetry never write poetry.
about robots was an interesting way If we are entering a time where
Our readers represent a growing to engage and question demarcations machines can write poetry, but human
community of practice that is of difference through the science beings are denied basic rights and au-
of increasing and vital global fictional. It was also a reprieve from tonomy in the service of political agen-
importance. scholarly research. In every sense, the das, then the insistence on our own
robot love relations in Love, Robot are humanity feels important. Collabora-
queer relations, but not explicitly so. tion, conversation, and creativity inter-
“Queer” is an academic and political vene in the siloing of way of thinking
term that most often refers to LGBT and training. As humans, engaging in
issues or identity. But it also refers to these interactions that foster thought
non-normativity, and an embracing feels important to design with humans
of the edges that trouble the norm. In in mind. Perhaps my overall point is
many ways, worldbuilding into science regardless of our interests in comput-
fictional worlds allows us to transgress er science, the computer and poetry
boundaries, and robots offer us queer may have much more in common than
visions of our futures. Thus, writing we could imagine. And that imagina-
the robot poems was a reprieve. It was tion can help world build in the ways
refreshing to take a different turn from we need.
my academic writing, which deals pret-
ty centrally with issues of difference References
(race, class, sexuality), and to explore [1] Williams. C.W. Introduction. Selected Essays of William
Carlos Williams. New Directions, New York, 1969.
how science fiction poetry can ask
[2] Rhee, M. Love, Robot. The Operating System, New
questions about difference through York, 2017.
To learn more about us, worldbuilding.
visit our award-winning website [3] Turing, A. Computer machinery and intelligence.
Mind 49 (1950), 433-460.
http://interactions.acm.org We have the freedom to recreate our
worlds through science and technology.
Follow us on Literal human-human interactions and
Biography
Facebook and Twitter creating across disciplines will prompt

Margaret Rhee is a poet, artist, and scholar. She is the
author of several poetry chapbooks and the full-length
To subscribe: imaginative transgressions between collection, Love, Robot (The Operating System, 2017). As
a scholar, she is at work completing her first monograph,
http://www.acm.org/subscribe humans and machines. Conversations How We Became Human: Race, Robots, and the Asian
and questions that engage with art, sci- American Body. She received her Ph.D. from UC Berkeley in
ethnic and new media studies in 2014, and held academic
ence, and technology spark new ideas appointments at the University of Oregon, and UCLA.
Currently, she is a visiting assistant professor in the
Association for to shape your own work in directions Department of Media Study, University at Buffalo SUNY,
Computing Machinery you may not have imagined. It also may and a visiting scholar at the A/P/A Institute at NYU.
give you permission to nurture inter-
ests that may not exist within your own © 2017 ACM 1528-4972/17/12 $15.00
46
IX_XRDS_ThirdVertical_V02.indd 1 12/5/17 3:25 PM XRDS • W IN T ER 2 017 • V OL .24 • NO.2
Check out the new acmqueue app
FREE TO ACM MEMBERS
acmqueue is ACM’s magazine by and for practitioners,
bridging the gap between academics and practitioners
of computer science. After more than a decade of
providing unique perspectives on how current and
emerging technologies are being applied in the field,
the new acmqueue has evolved into an interactive,
socially networked, electronic magazine.
Broaden your knowledge with technical articles

focusing on today’s problems affecting CS in
practice, video interviews, roundtables, case studies,
and lively columns.
Keep up with this fast-paced world

on the go. Download the mobile app.
Desktop digital edition also available at queue.acm.org.

Bimonthly issues free to ACM Professional Members.
Annual subscription $19.99 for nonmembers.
feature
“We Had Tough Times,

But We’ve Sort Of
Sewn Our Way
Through It:”
The Partnership Quilt
Using capacitive touch sensors and traditional quilting techniques
lead to the collision of seemingly disparate worlds and resulted in
the creation of the Partnership Quilt, a living archive of voices in
the shape of an interactive piece of craftwork.
By Angelika Strohmayer and Janis Meissner

DOI: 10.1145/3155128
Q
uilting is about making connections and establishing relationships between different
parts. The craft process involves sewing several textile layers together to create a thicker
padded fabric. However, in the case of our recent involvement in the Partnership Quilt,
the connections also emerged on a collaborative level. The collective act of sewing
simple rosettes and quilting blocks turned into an interactive piece of art by bringing
different and disparate expert knowledge together. We as human computer interaction (HCI)
researchers and interaction designers worked alongside the quilting designers from Six Penny
Memories,a as well as staff and service users from Changing Lives,b specifically Girls are Proud
Photo courtesy of Angelika Strohmayer and Janis Meissner
(GAP) and the Male Action Project (MAP). Changing Lives is a U.K. based charity focused on
those deemed “vulnerable”—the we combined elements of social-care was taught in schools in the North
homeless, people suffering from ad- practice, professional quilting skills, East of England, and many of those
diction, and the unemployed. The as well as interaction design tools to who put a stitch in the quilt had previ-
GAP/MAP projects specifically provide produce a living archive of experiences ously done this in their childhood, but
people-focused services to support from the women GAP supports, along- the type of teaching they spoke of rare-
those engaged in sex work, survival side those who support them. ly happens anymore. Many different
sex, or anyone experiencing sexual ex- techniques were used to construct the
ploitation. Through this collaboration, THE FIRST STITCH quilt, such as English paper-piecing,
The quilt is contextualized in local applique, machine-quilting, and tie-
a https://six-penny.com histories of craft and women’s work. quilting. Using these traditional tech-
b http://www.changing-lives.org.uk Quilting used to be something that niques, the women “[were] carrying on

Photo Credit TK

feature
ed by a group of women over what

ended up being roughly one year. It
consisted of various phases that were
carried out during drop-in sessions,
two days of work during the Quilt-
ing Extravaganzas, and work carried
out by individuals in their respective
homes and/or workspaces. The entire
project included: (1) the sewing of in-
dividual rosettes; (2) the appliquéing of
these rosettes onto a background fab-
ric; (3) the joining up of these squares
and machine quilting to place the
backing, wadding, and the quilt top; (4)
the creation of the quilted layer of ca-
pacitive touch sensors and tie-quilting
of the traditional quilt; (5) the attach-
ing of the Touchboard with the embed-
ded audiofiles from the reflection; and
(6) the edging of both quilts. To be able
to go through each of these six stages
of the production of the quilt, “we all
used our own little bits of knowledge,”
but ultimately as one case worker put
it, you “just need a little squad of wom-
en, and you can get so much done.”
Once the quilt was nearing its last
stages during the second Quilting
Extravaganza, we sat around a large
table, the quilt laid out in front of us,
and a cup of tea in hand, to marvel at
a tradition that’s particularly popular together and tie-quilt, solder, and at- our accomplishment and reflect on
in this area.” tach the layer of low-cost DIY capaci- what we had done. It was here we re-
We introduced the idea of turning tive touch sensors to the traditional corded a roughly 30-minute discussion
the quilt blocks into capacitive touch quilt. After which, we audio-recorded a of many of those involved in the quilt.
sensors to be able to share experiences group reflection on the creation proc- During this discussion, we reflected
of making the quilt with others who in- ess and our feelings toward the quilt with many different aspects of the
teract with it. One of our goals in our made up of the voices of many of those quilt such as the history of it, the tech-
overall work is to alleviate stigma in who partook in the creation of it. This niques used, as well as the different ex-
stigmatized groups. People engaged recording was later edited into 12 indi- periences we had in the crafting of it.
in sex work are often stigmatized. vidual audio clips that were then trans- One of the women who produced many
Through the quilt project, it’s process ferred to the quilt. of the rosettes said one of her sons had
and outcomes, we hoped to work to- Following these days, we visited the taken two rosettes that she had sewn
ward advocacy tools to reduce stigma quilting shop from where Six Penny together and placed them on his pillow
(e.g. public exhibitions of the quilt) and Memories work to witness how they at night. The group encouraged her to
Photo courtesy of Angelika Strohmayer and Janis Meissner
share personal well-being strategies integrated their traditional skills into sew them onto his pillow, to which she
focused on self care (e.g. sewing). Al- the digital layer of the quilt. Six Penny responded: “He won’t let us! He car-
most simultaneously, Six Penny Mem- Memories were able to quilt the sensors ries them around everywhere. They
ories were introduced to the project. we made from tinfoil and wires sol- go in his pocket, in his school bag, he
They are a duo of professional quilters dered to paper onto the fabric, and at- just carries them everywhere.” Anoth-
and designers who have written sev- tached a small pocket as the new home er woman stated the quilt “needs to be
eral books, appeared on quilting TV, for the BARE Conductive Touchboard used, doesn’t it?”, urging us all to con-
and have authored hundreds of quilt- (a commercial microcontroller special- tinue to interact with it.
ing patterns. ized for capacitive touch sensing). A Twelve extracts from the recordings
week later, the charity staff and women were cureated to be shared with the
PUTTING THE PIECES TOGETHER from GAP also visited the shop to ex- wider public through the touch inter-
We organized two days to come togeth- plore the audio-recordings triggered by action on the quilt: When one of the
er as part of what we called “Quilting interacting with the finished quilt. rosettes on the quilt is touched, one
Extravaganzas” to sew fabric hexagons In the end, the quilt was construct- of the audio files starts to play. Four of

these extracts contextualize the quilt a chance for the case worker and those emotions toward it, it was also seen as
historically, six discuss the making of women to continue to be engaged with an example of what can be achieved
the quilt, and two address some of the each other despite the changes; the for a drop-in on a broader level. Drop-
outcomes that we have either already case worker said, “It’s nice to be able ins are a common type of service de-
achieved or hope to achieve in the near to get back into this project […] it’s the livery promoted by sex work support
future with the quilt. first time I’ve seen these lasses in more services and other charities. They are
Since then, the quilt has been exhib- than a month because I no longer work sessions that take place regularly,
ited at the Vlieseline stall at the Festival at their drop-in.” This sentiment con- and for one of the women who took
of Quilts in Birmingham, UK. On top tinued at the second Quilting Extrava- part in the drop-in where the quilt
of exhibiting the quilt for amateur and ganza a month later: “This quilt has was developed, the drop-in is an op-
professional quilters at the Festival of almost been a little community link for portunity for “just having a laugh and
Quilts we are also working on exhibit- us that’s kept us together,” explained that” or an opportunity for some fun.
ing it in a venue closer to those who put one of the women. As one woman explained, “to get out
a stitch in it so they can go to take their Not only has the quilting project and, just for a couple of hours, yea,
friends and family to see the work they provided a space for social interac- those couple of hours do go fast don’t
have done. Furthermore, the charity tion, but it was also an opportunity for they?” Although drop-ins are usually
aims to use the quilt not only for pub- learning new skills. By taking part in not targeted toward a long-term goal,
lic education at these private, public, the entire process, everyone involved this particular drop-in started to re-
or semi-public exhibitions, but also as in the project went through the en- volve around the quilt. Each session,
a continued form of service delivery. tire process of designing and sewing the women would sit together to sew
Since we used a combination of DIY quilting blocks, as well as quilting the some rosettes for the quilt. Initially,
sensors and the out-of-the-box touch various layers of the blanket together. they were sewing simply for the pur-
board, it is quite simple to change the On top of these traditional skills, the pose of sewing, for the mindfulness
audio files that are triggered through process of creating, wiring, and at- involved in the activity: “When we
the touch of a rosette; to transform the taching DIY capacitive touch sensors first started to make the quilt I didn’t
interactive piece of art into a living ar- to fabric were also mastered by those think we’d ever get to this far. To this
chive of voices of those involved in the involved. Another caseworker, who thing,” another shared. However, as
process. As such, charity staff are keen had heard a lot about the quilt in the the project continued to progress,
to explore different ways in which the office, supported the project by mak- and when Six Penny Memories and
recording of audio and interactions ing blocks at the drop-in center she’s Open Lab became involved, the quilt
with the quilt could be used for staff responsible for. Without ever having started to become a more tangible
and volunteer training, one-on-one re- seen the quilt before, she responded goal to work toward; something to be
flection with clients, or the sharing of very emotionally when the artifact proud of.
their services with other charities and was finally in front of her. As she In the end the quilt was a chance
members of the general public. struggled to find the right words, she for people to come together to work
concluded: “I feel a bit like, wowed. Is on something collaboratively and cre-
REFLECTING ON THE QUILT that the word? Like, like what you’ve atively, while simultaneously having a
While the quilt is being used in exhi- just said. Getting a group of lasses to- space for individual, mindful practice.
bitions and will be used for training gether […] I don’t think the GAP and A case worker sums the concoction of
and reflexive purposes in the future, MAP projects ever come up with any- emotions, craft, and collaboration that
we want to sidestep to reflect a little thing like this. Like, creative, in such went into the process, outcome, and
bit on the quilt as an object, as well a long time!” continued use of the Partnership Quilt
as the process behind the making of While there are a few very tan- as something that helped: “[We] keep
it. Just like the layers of fabric and gible skills involved in the process of werselves busy, we’ve kept ourselves
wadding that are in the quilt, it has making the quilt that those involved out of trouble, and I’ve got some lovely
layers of stories and experiences me- learned, it also supported the women memories sitting with youse.”
ticulously sewn into it with each stitch emotionally. This can be seen in the
and button. following extract from our collective Biographies
The quilt was a chance for the local reflection from when we were nearing Angelika Strohmayer is a Ph.D. student in the Centre for
Doctoral Training in Digital Civics at Open Lab, Newcastle
women to sit together quietly during the end of the project: University. She is interested in the ways in which digital
the drop-in session, while carrying out Caseworker: “I bet [the sewing] technologies can be designed for support services to aid
their service delivery while simultaneously being useful for
an activity. It was, however, also an op- chilled you out a bit…” advocacy work and the sex worker rights movement.
portunity for people to stay in touch GAP client: “It did, ‘cause at the Janis Meissner is also a Ph.D. student in the Centre for
Doctoral Training in Digital Civics at Open Lab, Newcastle
after these drop-ins. The first Quilt- time I refused medication as well, so University. As maker technologies give individuals an
ing Extravaganza day took place more […] it was the only thing I focused on. opportunity to develop their own objects and tools,
Janis is interested in exploring the ways in which these
than a month after the caseworker who ‘Cause all the other stuff was negative, technologies can empower different communities.
had been a driving force in starting the and that was positive.”
project was moved to work at a differ- On top of these personal experi- © 2017 Copyright held by authors
ent drop-in center. As such, the day was ences of making the quilt and the 1529-4972/17/12

PROFILE DEPARTMENT EDITOR, ADRIAN SCOICĂ
Sara Mauskopf
Innovative Thinking in
the Service of Parents
DOI: 10.1145/3155218
Having graduated from MIT with a “I’d never considered MIT before, MIT in order to stay close to theoretical
perfect GPA, and kickstarted her career because I had previously thought it was computer science, her life would take a
with high-profile jobs at four of Silicon for people smarter than me. I didn’t yet another unexpected turn.
Valley’s more iconic brands: Google, think I was smart enough to go there, I “During my senior year, I just burned
YouTube, Twitter, and Postmates, Sara just knew I was someone who worked out. I was studying so hard, and always
Mauskopf’s story became all the more hard. But then I heard all of these kids trying to do everything perfectly, that
impressive when she decided to walk who were going to apply, and I was doing it was just exhausting. So when Google
away from salaried life and bravely the same math program as they, so I extended me a job offer, I decided to
launch her own startup, Winnie. thought that if they can do it, then I take a break from school and just go
It therefore felt like no small privilege can do it too,” she reminisced, adding for it. They seemed like a really fun
to interview her and, in the process, she visited the MIT campus after the company, with lots of young people
share with current and aspiring computer summer school was over, and she just like me, and it was on the other
science students some the good, immediately recognized it as the right side of the country in a completely new
the bad, and the difficult lessons and place for her. environment. I figured I could always go
perspectives she gained along the way. “MIT offered me a place, so I back to MIT for a Ph.D. if I wanted that,”
accepted, and I enrolled thinking at the she explained. Even though she thought
MATH GEEK DISCOVERS time that I was going to major in math,” she would eventually move back to her
COMPUTER SCIENCE she confessed, adding she was unaware native East Coast, it’s been 10 years
Although Mauskopf now resides at the at the time that she had yet to discover since she moved to California and she
heart of the global tech industry, her early her true passion. loves it so much that she’s not planning
school years were not stereotypically “With time, all the other kids that I on moving back anytime soon.
spent tweaking software, but rather talked to—and all of my friends—were “I think San Francisco was a really
in the pursuit of a more abstract form doing this new thing called computer good fit for me, and a new start. Starting
of logic. She grew up in the suburbs of science. And it looked like a lot of fun,” my first job at Google was also very nice,
Philadelphia, where she had been drawn she recalled. In the beginning she felt because Google had a sort of collegiate
to mathematics for as far as back as she slightly jealous of how they would solve feel to it, especially back in the day.
could remember. She would dedicate all hard problem sets and work on cool People would eat dinner together, and
her spare time to pursue the subject. technical projects together, while she play volleyball, so it kind of felt like an
“I was always interested in math was mostly on her own. After having extension of college in a way. It made
growing up. I didn’t know anything about attended a career night for math majors, the transition to adult life really easy,”
computer science, or that it might be and realized she wasn’t excited about she explained, pointing out she is very
related or interesting to me, but I truly any of the advertised options there, she grateful to have started her career with
loved math, and was very driven to be decided to follow her heart and enrolled a company that takes such good care of
the best at it. It was sort of my hobby,” in the introductory CS class, which she their employees.
she bemusedly reminisced. aced. “The class had a final project for However, after having spent more
Consequently, toward the end of high which they picked winners, and I was one than three years with Google, during
school she envisioned her ideal future of the winners. Then I realized that I was which she learned how to help a tech
university major in this field, and decided good at it, and I liked it, and I switched company navigate the tumult of the
to attend an elite summer program for majors,” she proudly remembered. 2008 financial crisis, Mauskopf began
math and science sponsored by the She would go on and later graduate to realize that she was ready for a new
state of Pennsylvania on the Carnegie with a perfect GPA, a rare feat at one of the challenge. She had begun using a new
Mellon University campus. In a dramatic best technical universities in the world. social platform called Twitter, and she
boost of confidence, it was at this really liked it, so she applied for a job
summer school that she met colleagues LIFE IN SILICON VALLEY there and she was offered her dream
who talked about applying to MIT for Though Mauskopf felt personally very job as a product manager. She decided
college, which inspired and emboldened happy in academia, and had originally to take a leap of faith and accept.
her to prove herself there. planned on doing a graduate degree at “Everyone thought I was crazy. I joined

Twitter in 2010, during the tail end of she could hire her own team and run the
the financial crisis, when the company company according to her own values.
had a little over 200 employees,” she “Our head of engineering is someone
pointed out jokingly, but explained her I’d worked with at Postmates, and early
philosophy in life is that one should on we also worked with a former
follow their heart, and not succumb colleague from Twitter. We are a
to peer pressure. “All my job moves woman-led company, and a woman-led
seemed crazy to other people, but you engineering team, and we strongly value
know… I just knew it was right for me.” people of all backgrounds. And so many
Twitter went on to be very people want to work with us, that we
successful, and she of course went consider it our secret weapon to be lucky
on to do one more company move (to to hire extremely talented people whom
Postmates) before she decided it was other companies just ignored. And we
finally time to try her hand at running really value having time outside of work,
her own business. whether that’s family time or time
spent pursuing education and hobbies.
BLAZING HER OWN TRAIL Unless there’s an emergency, we don’t
In 2015, Mauskopf celebrated a major want people working nights and
life event when she had her first baby, weekends,” she explained, adding that
and the experience opened her eyes to the work-life balance made them
a whole new world. It is no secret that more, not less productive, despite the for a small subset of the more affluent
being a parent is hard, and requires fact that it goes against the 60-hour- suburban parents.
not just patience, but also a wealth of week startup narrative popular with “It’s really important to us that
knowledge that is not easily accessible early-20s engineers. Winnie is a trusted, informative and
to young parents. quality community of parents, so we
“I had a child, and I almost THE IMPACT OF WINNIE do heavily moderate, employing both
immediately went back to work at Regarding her aspirations for Winnie, humans and AI tools, so that action can
Postmates, so I didn’t have time to join Mauskopf explained that her mission is be immediately taken on content that’s
any mothers groups, make connections, to help all its users be amazing parents, of questionable validity, such as anti-
and get all the inside info. I was looking regardless of their gender, age, race, or vaccine content. We don’t want parents
for very basic things like finding socioeconomic background. to get bad advice on Winnie, and we
childcare, or activities to do with your “Winnie is for all parents, it’s not don’t want them to get medical advice
child on weekends, and none of that just a mothers group. That’s a big on Winnie, so we make sure we point
information was online. I thought it was distinction that we’ve made. There’s now them to their doctor when such issues
crazy, and I realized that there was a a trend with millennial parents where come up.”
massive opportunity. It wasn’t going the fathers take a more active role, and
to require novel engineering, it was a the interesting thing about dads is that ADVICE TO HER YOUNGER SELF
data problem, and I knew already how to they’re completely locked out of the As her plans for the future look bright,
build it. So after talking to a coworker traditional mother networks, so they with Winnie growing healthily alongside
at Postmates who was also an engineer don’t have access to quality information. her career, I asked Mauskopf what
and a mother of two, we got excited While they want to take a greater role advice, if any, she would have given her
about the idea. We had the perfect in parenting, they’re handcuffed, so younger self if she had the chance.
backgrounds to attack this problem, and we’ve had a lot of feedback from dads “The biggest piece of advice I
a real desire to solve it, so we just went who are happy that finally there is a would give my younger self would
for it.” She decided she would quit her product for them as well. This is all the be to focus more on relationships
day job and found a startup to help new more important that now for the first than on being the best. In school I’d
parents navigate the world. time in American history, the majority always focused on being the best, and
“Again, everyone thought I was crazy, of households, whether single-parent or beating everyone else, and I somewhat
because startups are a lot of work, they two-parent, have all adults working, so continued that in my early years of
usually fail, and on top of that I had time is extremely precious and having working. And what I’ve realized in
just had a baby. But for me, it was the the right information at your fingertips the meantime is that you can’t be
perfect time,” she proudly explained. is crucial.” successful unless the people around
When I asked about what the She shared that while her personal you are also successful, and that those
experience of building a company had experience as a parent did inform some I met along the way are critical to the
been like for her, in contrast to the of the content and design choices of the success of my company today.”
glamorized Silicon Valley stereotype, product, she is very careful to build the
Mauskopf explained the best perk was best product for the public at large, not COPYRIGHT HELD BY AUTHOR

end
Word art depicting areas of interest for the Stanford Social Media Lab.
theories about Facebook and Twitter.

LABZ In general, people have consistent
folk theories about these sites and
Social and Psychological conceptualize them as a rational as-

sistant, an unwanted obser ver, a
Questions about Humans transparent platform, or a corporate

black box. Finally, a third area of re-
search in the Social Media Lab inves-
and Technology tigates the efficacy and ethics of us-

ing conversational agents (e.g.,
The Stanford chatbots) to communicate about

mental health issues. Our evidence
suggests automated assistants are
Social Media Lab not yet equipped to handle mental
health dilemmas effectively, and col-
T
laborations bet ween technolog y
he Social Media Lab at Stan- series of automated text analyses re- scholars and practitioners are needed
ford Universit y evaluates vealed systematic differences in how to address this new frontier.
communication and psycho- fraudulent scientists write their re- I am a researcher in the Social Me-
logical questions about hu- search reports compared to genuine dia Lab primarily focusing on how
MediaX at Stanford University
mans and technology. Our lab has scientists. The data suggest language language reveals what people are
several notable areas of research. patterns may be one way to diagnose thinking, feeling, and experiencing
One major area investigates the the scientific fraud problem that has psychologically. I combine computa-
linguistic traces of digital media to received substantial academic atten- tional social science (e.g., gathering
reveal social and psychological dy- tion. A second area of work evaluates and analyzing language data through
namics, including deception, persua- how people conceptualize social net- automated means) with media stud-
sion, and well-being. For example, a working sites, particularly their folk ies to understand how language re-

English mathematician Alan Turing is widely recognized the father
of computer science. His work helped the Allies win WW2 against Nazi Germany.
Nonetheless, he was convicted of indecency due to his homosexuality.
flects social and psychological pro- BACK

cesses, such as deception and
persuasion. My approach uses several
data sources, from social media (e.g.,
Equality through
Yelp) to professional media (e.g., sci-
ence publications), to take advantage
of the increasingly recorded nature of
Digital Technology
communication behavior that is left Dedicated to Alan Mathison Turing (1912-1954),
online for analysis. a pioneer of computer science.
One recent area of my work investi-
gates how deception occurs in mobile The mass adoption of the World Wide Web (WWW) constitutes a
dating conversations. With Professor major point in the history of civil rights and the empowerment of
Jeff Hancock, Founding Director of the oppressed. One of the main features of the WWW was online
the Social Media Lab, we recruited communication through (then-popular) chat rooms (e.g. AOL). It
subjects to provide their messages to was the first time people with a similar mindset and shared agonies
a dating partner and rate the decep- could communicate without being limited by geography, and, more
tiveness of their message content. importantly, without feeling exposed or vulnerable. For the LGBT
The frequency of lies was low (ap- community, the WWW and its post-products, like social media,
proximately 10 percent of all mes- represent a major backbone of their everyday communication. (For
sages), however, nearly 60 percent of example, a study found 80 percent of LGBT respondents utilized
the lies could be categorized into social media, compared to 58 percent of the general population1).
two key impression management AOL chat rooms were only the beginning. In the mid ‘90s, dating
goals: (1) self-presentation, or lying websites like gay.com started a socializing revolution by allowing
to appear interesting and attractive users access to the profiles and photos of other members. Soon after,
to the partner, and (2) availability other websites offering LGBT content became popular and went
management, or lying to control how mainstream in the LGBT community. The second social revolution
available or desperate a person ap- happened due to the rise of smartphones and other handheld devices,
pears to the date. This is one of the which made communication even simpler and faster with new
first studies to evaluate the frequen- location-based dating apps, such as Grindr (and years later, Tinder).
cy and content of deceptive mobile More recently, the tech industry itself has fought for the rights
dating conversations using actual of oppressed communities. A landmark example was Google’s
messages from daters. “Legalize love” campaign in 2012, the goal of which was to foster safe
We are always looking for new col- conditions for LGBT people inside and outside their workplace in
laboration opportunities across our countries with anti-gay laws. Another important example was when
programs of research at the Social Salesforce.com, a San Francisco-based cloud computing company
Media Lab. For more information on and one of the largest tech employers in Indiana, threatened to stop
our studies or research approaches, further investments unless then-Governor Mike Pence repealed a
please contact the Founding Direc- specific law that would have enabled discrimination against LGBT
tor, Jeff Hancock ( jeff.hancock@ people. Moreover, Apple’s CEO Tim Cook was the first CEO of a
stanford.edu), or the Associate Direc- Fortune 500 company to come out as gay.
tor of the lab, Sunny Liu (sunnyxliu@ One cannot help but think the remarkable impact of digital
stanford.edu). technology on today’s fortification and acceptance of the rights of
the LGBT community is destiny’s apology to one of the pioneers of
Biography computer science and a father figure of computing and artificial
David M. Markowitz (M.Sc. Cornell University; Ph.D. intelligence. Alan Turing was prosecuted in 1952 for homosexual acts,
Stanford University) is a researcher in the Department of
Communication at Stanford. He uses automated methods and died in 1954 from cyanide poisoning.
to collect and analyze language data to reveal social and
psychological dynamics (e.g., deception, persuasion, 1 https://cdt.org/blog/tech-industry-advance-equality-for-entire-lgbt-community
distress).
—Vasileios Kalantzis
DOI: 10.1145/3155163
Copyright held by author. DOI: 10.1145/3155220 Copyright held by author.

HELLO WORLD
Identifying Hate Speech in Social Media

BY ALEXANDRA SCHOFIELD AND THOMAS DAVIDSON
H
ate speech refers to state- vary from person to person, and it can be numerical labels from our data file (see
ments that attack or delegiti- tricky to find and represent the contextual Listing 1).
mize particular groups of people elements that distinguish hate speech. A classifier is a program that distin-
based on a demographic guishes different classes of objects using
category—race, gender, religion, sexual GETTING LABELED DATA labeled examples. It works by using
orientation, and so on. For social media What constitutes hate speech? There is features of the examples that help
platforms, like Facebook and Twitter, hate no clear objective definition of hate distinguish between the classes. Features
speech that specifically encourages speech, making its identification a are measurable values, such as the length
violence against a group is explicitly subjective task. A natural way to answer of a post, the number of times an
prohibited in the terms of service. this question is to show people examples offensive word appears in it, or the
However, these media platforms are large of texts that may be hate speech and have presence of a URL. In our work, we refer to
and hard to moderate, so those rules them decide what is and is not hate the set of input features as an array X, and
cannot fully prevent hateful posts, nor do speech. This enables us to construct a the correct labels as y. This formalization
they regulate other kinds of hate speech. corpus of data that are labeled into is referred to as “supervised machine
To mitigate this problem, machine different classes. In our example, we use learning,” because we infer a decision
learning can be used to identify potential an existing dataset of 24,783 tweets.1 model from labeled data and evaluate our
hate speech in larger collections of text. Each tweet was labeled by crowd workers: performance on the basis of these labels.
This problem is closely tied to spam 1,430 examples of (0) HATE speech,
filtering, a classic problem in text 19,190 examples of (1) OFFENSIVE USING LEXICONS
classification and natural language speech, and 4,163 examples of (2) An obvious starting point to identify hate
processing. NON-OFFENSIVE speech. Note the classes speech is to use offensive words and slurs
Based on past work in automatic hate are imbalanced: Though all the labeled as features. We can use a lexicon, a
speech detection [1], included is a brief tweets were potentially hateful, only 5 vocabulary list corresponding to some
tutorial on using Python, specifically percent were classified as such by the particular attribute, to choose a list of
scikit-learn [2], on a set of tweets to majority of crowd workers. We first load words that might help us distinguish hate
distinguish between hate speech and the text of these tweets and these speech from other posts. Hatebase.org
other text. We also describe some of the has an extensive list of terms used in hate
challenges in solving this task efficiently. speech.2 Our example lexicon includes a
1 A CSV of the data can be found here: https://
In the process, we demonstrate the github.com/t-davidson/hate-speech-and-of-
small sample of words commonly
complexity of the problem; definitions of fensive-language/blob/master/data/labeled_ associated with sexism, racism, and
what specifically constitutes hate speech data.csv homophobia. There are different ways to
convert a lexicon into features. In this
Listing 1. case, for each document, we are going to
use as features the number of instances
import pandas of each word as well as the total count of
data_url = 'https://raw.githubusercontent.com/t-davidson/hate-speech-
words in the lexicon. For a lexicon of 10
and-offensive-language/master/data/labeled_data.csv'
data = pandas.read_csv(data_url) words, we’ll have 11 features per tweet
tweets = data['tweet'] (see Listing 2).
y = data['class'] There are many algorithms for training
text classifiers, many of which are
implemented in scikit-learn and can
Listing 2. be swapped out for each other pretty
easily. To start, we are going to use a
import numpy as np logistic regression classifier, which is a
tokens = [re.split("[^a-zA-Z]*", tweet.lower()) for tweet in tweets]
simple model that works well with small
X_lexicon = np.zeros((len(tweets), len(hate_lexicon) + 1))
for i, tweet in enumerate(tweets):
for j, term in enumerate(hate_lexicon):
X_lexicon[i,j] = tweet.count(term) 2 A CSV based on the Hatebase lexicon can be
X_lexicon[:,-1] = X_lexicon.sum(axis=1) found here: https://github.com/t-davidson/
hate-speech-and-offensive-language/tree/
master/lexicons/hatebase_dict.csv

datasets. We want to see how well our
classifier labels data it didn’t use in Listing 3.
training. This helps us to prevent a from sklearn.linear_model import LogisticRegression
problem known as “overfitting,” where the from sklearn.metrics import accuracy_score
classifier makes predictions by learning from sklearn.model_selection import cross_val_predict
random noise in the data rather than the model = LogisticRegression(penalty="l2", C=0.01)
y_pred = cross_val_predict(model, X_lexicon, y, cv=10)
true signal of interest. In this case, we are print('Accuracy score: {:.2f}%'.format(accuracy_score(y, y_pred) *
going to use k-fold cross-validation: We 100))
split our tweets into k random groups,
then for each group, we use the other k – 1
groups to train the classifier and the
selected group to test it. In this case we Listing 4.
set k to equal 10. This strategy is already
implemented in scikit-learn (see
Listing 3).
Next, we introduce a few arguments to
make our classifier work well. One is the Listing 5.
penalty, also known as a “regularizer;” this
model = LogisticRegression(class_weight="balanced", penalty="l2",
is a way to reduce the amount a classifier C=0.01)
overfits the data it sees. An L1 penalty
encourages a model to weight only a few
coefficients as nonzero, while an L2
penalty encourages all of the weights in Listing 6.
the model to be lower. C is the inverse
from sklearn.feature_extraction.text import TfidfVectorizer
regularization strength: A small value like vectorizer = TfidfVectorizer(
0.01 implies that regularization is going to tokenizer=tokenize,
be important to the model. preprocessor=preprocess,
We can evaluate how well the classifier use_idf=False,
decode_error='replace',
works based on a few metrics. Accuracy is
min_df=5,
the total proportion of training examples max_df=0.5
whose predicted labels match their )
correct labels. In addition, for each class, X_count = vectorizer.fit_transform(tweets)
we can describe the recall, or the
proportion of examples from that class
that are correctly labeled, as well as the more than 99 percent. Perhaps most any information about word order.
precision, or the proportion of examples importantly for our application, the HATE Word counts across a large vocabulary
labeled as that class that are labeled recall increases from almost 0 to 34.6 tend to be sparse; that is, most words do
correctly (see Listing 4). percent. How can we do better? not appear in most documents. This can
This starting model obtains 77.45 be a problem, as rarely-observed features
percent accuracy on our data, which BAGS OF WORDS that distinguish classes may not be
seems pretty good at first glance. One problem that arises with the lexicon observed enough for the model to use
However, if we examine the data, we will method is a short list of offensive words them. One way we can help improve this is
see all but 25 of the tweets are being may not identify all hate speech. Slurs text preprocessing, in which we apply
classified as OFFENSIVE, with those 25 appear in text that is not considered hate some standard transformations to clean
classified as HATE and no tweets classified speech. For example, these terms have the text. This will ensure the remaining
as NON-OFFENSIVE. Due to the class been observed as used in jest in people’s words are those we care about. We use the
imbalance, the classifier is almost always everyday interactions, and hate speech following preprocessing steps:
selecting the most frequent class. We can arise without slurs [1]. Conversely, 1. replacing URLs and Twitter
want to reward our classifier for equally using a large lexicon of offensive words mentions with fixed strings (@sampleuser
valuing recall on all three classes of tweet. leads to a problem of false positives, as and @otheruser become MENTIONHERE),
We can do this by instructing the classifier many innocuous texts become classified 2. filtering words to only be
to reweight its optimization function to as hate speech. In these cases, we may contiguous sequences of alphabetical
balance the contribution of each class of need information about other words or characters (strip away all punctuation,
data to the training (see Listing 5). phrases to train our classifier. We can digits, and apostrophes),
The accuracy here decreases signifi- train a “bag-of-words” model instead, in 3. lowercasing all words ( “Obama”
cantly, to 60.03 percent, but the recall on which the features include frequency becomes “obama”),
the NON-OFFENSIVE class increases to information for a wide vocabulary without 4. stemming words, or removing

suffixes of words so that words sharing a
root are treated the same (“created” and
“creates” would both be shortened to
“creat”), and
5. removing words that appear in
fewer than five or more than half of all
documents, as well as stopwords,
standard common words like “a” or “the.”
After defining code to do this in
preprocess and tokenize functions
(available in the code sample), we can
ACM Transactions on Spatial import a large number of word features
from all of our tweets. Here, we use a
Algorithms and Systems TfidfVectorizer object to automati-
cally load our features into a term
frequency (TF) matrix using the previous
ACM Transactions on Spatial Algorithms and Systems preprocessing treatments (see Listing 6).
The resulting feature vector contains
(TSAS) is a new scholarly journal that publishes high- 3,615 features for each tweet. For this kind
quality papers on all aspects of spatial algorithms and of data—sparse, discrete counts of
systems and closely related disciplines. It has a multi- observations—a multinomial naïve Bayes
classifier can be efficient and accurate, as
disciplinary perspective spanning a large number of it is specifically built to model counts of
areas where spatial data is manipulated or visualized observations instead of arbitrary continu-
(regardless of how it is specified - i.e., geometrically or ous variables. We run this classifier using
the same cross-validation technique as
textually), such as: geography, geographic information
previously discussed. This time, instead of
systems (GIS), geospatial and spatiotemporal “balancing” the classes in the logistic
databases, spatial and metric indexing, location-based regression classifier, we force the classifier
to treat all classes as equiprobable by
services, web-based spatial applications, geographic
disabling fit_prior (see Listing 7).
information retrieval (GIR), spatial reasoning and This new model obtains 85.98 percent
mining, securing and privacy, as well as the related accuracy—a significant improvement.
More importantly, we are up to 49.51
visual computing areas of
percent recall for classifying hate speech
computer graphics, computer as hate speech.
vision, solid modeling, and Each of the models we have used so far
has stored a list of coefficients. These
visualization where the
coefficients determine how much each
spatial, geospatial, and feature contributes in deciding the class
spatiotemporal data is central. of a tweet. We can then find the top 10
features with the highest coefficients for
each class to understand more.
The journal is committed Some of these features match our
to the timely dissemination previous lexicon, while others better
of research results in the capture other distinctions between the
classes. Terms like “yellow” and “bird” may
area of spatial algorithms appear because they can be identified as
and systems. slurs in the Hatebase lexicon for having
potentially offensive connotations, but
they are often used with other meanings
and so tweets containing these terms
were classified as NON-OFFENSIVE.
For further information or to submit your manuscript,
visit tsas.acm.org ADDING DISTRIBUTIONAL
SEMANTICS
Subscribe at www.acm.org/subscribe The features have no way to distinguish

lexicon used before. Using only features
Listing 7. among the top 20 highest-weighted of
from sklearn.naive_bayes import MultinomialNB each class in the previous classifier (a
model = MultinomialNB(fit_prior=False) combined set of 40 features), one may
y_pred = cross_val_predict(model, X_count, y, cv=10) still obtain 86.41 percent accuracy using
logistic regression. However, the lack of
accuracy on the hate speech itself is still
Listing 8. problematic; these vector representations
do not convey enough about the nuances
from sklearn.decomposition import TruncatedSVD behind the distinction between hate
tfvectorizer = TfidfVectorizer(
tokenizer=tokenize,
speech and other offensive speech. This
preprocessor=preprocess, illustrates the difficulty of identifying hate
use_idf=True, speech, particularly differentiating it from
decode_error='replace', other forms of offensive language that are
min_df=5,
max_df=0.5 prevalent on the internet.
)
X_tfidf = tfvectorizer.fit_transform(tweets) CONCLUSION
tsvd = TruncatedSVD(n_components=20)
X_lsa = tsvd.fit_transform(X_tfidf)
We demonstrated three standard
methods for producing features for text
classification, targeting specifically the
between the different senses in which a real-numbered vector, with the idea that problem of automatic hate speech
word like “bird” might be used. Knowing vectors close together in the Euclidean identification. We have shown how to
which words appear with “bird,” however, space correspond to documents that are succinctly and efficiently program this
might provide more context. Distributional close together in the semantic, or kind of classifier in Python, as well as
semantic models try to get at this idea by meaning, space. We hope this might pointing to the types of features that
summarizing information about word capture information about the context might improve accuracy: lexicons,
co-occurrence, or which words appear words of a tweet, such that a tweet bag-of-words features, and distributional
together in the same context. In this case, containing a slur being used as a slur will semantic models. We also demonstrated
the context we care about is a tweet, so look different than one where the word is these features are not sufficient on their
we simply use statistics describing how being used in a more innocent context. own for accurate classification of hate
often pairs of words show up together in The process of factoring a document- speech, as the presence or absence of a
the same tweets. word matrix into skinny matrices usually slur is not sufficient to distinguish hate
One simple way to do this is using works better if we reweight each word speech from offensive speech. Web
latent semantic analysis (LSA) to produce frequency based on document frequency companies are currently using these
vector representations of documents and (DF), or the log of the proportion of approaches as part of their tools to
features. LSA, also called latent semantic documents in which the word appears at identify and prevent the abuse of these
indexing (LSI) or truncated singular value least once. The TfidfVectorizer platforms by hate speakers.
decomposition (Truncated SVD), is a type allows us to enable multiplying by the
of approximate matrix factorization. In inverse document frequency (IDF) to References
the first step, it takes a large matrix and downweight terms in the feature matrix if [1] Davidson, T., and Warmsley, D., and Macy, M. and I.
Weber. Automated hate speech detection and the
finds three matrices with specific they occur in a large number of docu- problem of offensive language. In Proceedings of
properties whose product equals the ments (see Listing 8). ICWSM ‘17, Montreal, Canada. 2017, 512–515.
original large matrix. In the second step, it Using 20-dimensional LSA representa- [2] Pedregosa, F., Varoquaux, G., Gramfort, A., Michel,
V., Thirion, B., Grisel, O., and M. Blondel et al.
removes all but the top few dimensions of tions of a weighted word-document Scikit-learn: Machine learning in Python.
Journal of Machine Learning Research 12
the matrix that contribute the most to matrix along with the original count (Oct. 2011), 2825-2830.
matching the original large matrix. If we features, we obtain the best accuracy yet,
choose dimension 20, LSA will return for a of 88.25 percent.3 Upon inspection, we Biographies
document-word matrix a 20-column notice that while we only have 26.9 Alexandra Schofield is a Ph.D. candidate in computer
matrix U with one row per document, a percent recall of the hate speech class science at Cornell University. Her research focuses on
practical methods for experts outside computer
20-by-20 matrix Σ with only entries on (0), we can obtain 93.64 percent recall on science to use distributional semantic models for text
the diagonal, and a 20-column matrix V offensive tweets (1) and 84.5 percent mining.
with one row per word, such that UΣV T is recall of other tweets (2) without the Thomas Davidson is a Ph.D. student in the Department
of Sociology at Cornell University. He is currently
as close a reconstruction possible of the working on a dissertation using social network analysis
original matrix. and natural language processing to study political
3 Because the TruncatedSVD algorithm is debates on social media.
If we treat this model as a black box, probabilistic, this accuracy can fluctuate, but
the important part is that for each over many trials it averages between 88.2 DOI: 10.1145/3155212 © 2017
document and word, LSA outputs a and 88.4 percent ACM 1528-4972/17/12 $15.00

JARGON POINTERS
fit in with the “cool” tech savvy world

HUMAN TO HUMAN around us seems to be of utmost
LGBT (Lesbian Gay Bisexual, and importance for young people. Sadly
Transgender) is an initialism that has peer pressure or a need for validation
been extended from LGB, which was It’s an undeniable fact that technology
has a great impact on personal values, ties into feelings of self worth for some
used in 1990s to replace the less teens, who feel their identity isn’t
inclusive term “gay.”Another, even more expressions, feelings, and the way we
perceive relationships and experience recognized unless they’re appreciated
inclusive acronym is L.G.B.T.Q.Q.I.A, but in online.
common use the addition of a plus sign intimacy. As computer scientists we
must be able to explore the https://www.deseretnews.com/
at the end (LGBT+) is enough to denote article/865603981/Growing-up-digital-
that no sub-community is left out. consequences of how we recognize
power as well as the dynamic of How-the-Internet-affects-teen-identity.
dominance that is expressed in html
Mansplaining is a term used when

current society. Although our
a man interrupts or tries to speak over
differences do not have to define us, it “A Future for Intersectional Black
a woman in a condescending way.
is unfortunately too common for some Feminist Technology Studies”
He “mansplains” when he cuts her off
individuals to be given better By Safiya Umoja Noble
and explains something to her that
opportunities professionally and According to Noble, the struggle
she already understands. When a man
academically, which can create to recognize multiple, interwoven
interrupts or corrects a woman who is
imbalances in the workplace and systems of oppression has been
talking on the basis of her expertise
beyond. In this issue, we set out to ongoing for more than 40 years. The
or her experience, she is assumed to be
recognize these problems and focus of Black feminists has been
ignorant from the man’s point of view,
understand how technology accelerates the global liberation Black women,
or inhibits such social stigmas. which is crucial in the fight to end
or that she is incapable of knowing.
—Shefali Bhatnagar oppression. However ignoring the
intersection of race and gender,
SIGMOD (Special Interest Group of doesn’t give a full view of how to
Management of Data) is an ACM’s special RECOMMENDED ARTICLES tackle that same oppression in the
interest group that is concerned with digital world. Noble calls for more
the techniques, principles and work on how digital technologies are
“Growing up Digital: embedded within an imbalanced
applications of data management
How the Internet affects teen identity” power structure, which is evident
technology. SIGMOD has academics,
By Chandra Johnson in the way in which technology
researchers, software developers,
This article was part of a longer is designed, how information is
students et. al. in its member pool.
series highlighting how technology manufactured and distributed, and
negatively impacts the lives of how digital communication manifests
teenagers who, in order to find a place on the web. Recent real-life examples
SJW (Social Justice Warrior) is a
in the social media world, fall prey to of Twitter verifying a known white
pejorative term for a person who
negativity and lowered self-esteem. supremacist are the types of problems
generally enforces an overly aggressive
The reporter shares the story of young Noble seeks to address when she
and incendiary style of discussion while
Hannah Smith, who ended her life calls for a “intersectional internet
promoting socially progressive views
due to some hurtful comments on infrastructure.”
including civil rights, multiculturalism,
her ask.fm profile. Smith’s suicide http://sfonline.barnard.edu/traversing-
feminism and identity politics. This
is an example that illustrates how technologies/safiya-umoja-noble-a-
acronym was once positive, but it has
vulnerable today’s youth have future-for-intersectional-black-feminist-
devolved into a snarl word today.
become. The idea that teens look for technology-studies/
acceptance on various social media
platforms is not new, but this specific
SWIFT (Supporting Women in
incident emphasizes how significant
Information Technology) was a project
a role the Internet plays in their
that was started by Dr. Maria Klawe
lives. Which raises the question, is it
when she was the NSERC/IBM Chair.
necessary to find acceptance online?
A five-year long UBC research project,
Wanting to be liked, accepted, and
its directive was to encourage
and increase female participation in
the field of information technology.

consequences for their acquaintances,
NOTABLE BOOKS Black Lives Matter
or concern for unknown people.
A global network that is chapter-based
James discovers youth aren’t really
working for the betterment of all Black
The App Generation: How Today’s concerned about moral or ethical
people—including the often ignored
Youth Navigate Identity, Intimacy, and privacy. She tries to describe ways in
Black LGBTQ+ community—who
Imagination in a Digital World which she can bridge the gap in digital
have systematically suffered and
Howard Gardner and Katie Davis, media by including “conscientious
been “targeted for demise.” Their
Yale University Press (2014) connectivity,” that marks sensitivity
motto “Black Lives Matter” began as
It is convenient to say that the present toward online dilemmas and ethical
a rallying cry on Twitter in 2013 as
generation is deeply connected online behavior.
a direct response to the acquittal of
with the digital world. This book George Zimmerman for the murder
explores the meaning of being of Trayvon Martin. Today the group,
“dependent” on apps rather than just INTERESTING APPLICATIONS founded by three Black women, has
being “enabled.” The authors also broadened their stance to not only
explore the difference in the lives of include the issue of state violence
youth in our current times and the Ustream
but also fighting “anti-Blackness”
time before the digital revolution. If you are participate in a social
socially and politically. They use
According to Gardner and Davis, justice cause and want to protest and
social media as a tool to organize
there are three significant phases of share it on the internet, the Ustream
and disseminate information on the
adolescence: identity, intimacy, and broadcasting app is just the right
liberation of Black people.
imagination. They’ve done a lot of thing. This app allows you to easily
http://blacklivesmatter.com/
research in these areas, conducted stream video from your mobile device
interviews with young people, and and promote it on social media.
compared situations and mindsets It offers content that’s live, as well
before and after one’s life “switched several streaming sites to tune in to ON THE SMALL SCREEN
over” to digital. They also explain the ongoing demonstrations.
downside of apps: They deprive young www.ustream.tv/
“The Future of Content, Technology
people of their identity, assimilate a and Society”
superficiality in relationships, and happn Futurist Gerd Leonhard gave a
inhibit imagination and creativity. Happn is a dating app that works on keynote at PING Helsinki on content,
But, benefits also exist: In some the concept of two people crossing humanity, technology, and media, and
cases they can encourage a sense paths sometime before and are using what their future holds. He brought
of identity, which if used maturely, Happn to find a partner. The idea is up the issue of machines replacing
might evolve relationships and make that people destined for each other humans and the paradigm shift in
them deeper in certain situations and may have crossed paths, and Happn the way that humans work. How will
also increase creativity. uses this notion in their algorithm to content marketing work in the coming
make it seem more romantic. years, and what does it take to be on
Disconnected: Youth, New Media, https://www.happn.com/ the “human team?” These are some of
and the Ethics Gap the questions he attempts to answer.
Carrie James https://www.youtube.com/
watch?v=Euca8PZU0es
The MIT Press (2014) USEFUL WEBSITES
In this book, the author analyzes
the mindset and thought process of “5 Crazy Ways Social Media Is
American Association of People
young people and adults about some Changing Your Brain Right Now”
with Disabilities (AAPD)
of the online dilemmas they might This animated video by AsapSCIENCE
The AAPD works to improve lives of
face concerning ethics. Conducting gives an insight into how social
disabled people. The group acts as a
interviews extensively with people media affects our brains—from how
“gatherer, connector, and stimulant
of ages between 10-25, the author we perceive our emotions, to how
for change” in the lives of the disabled.
explains their thinking in relation to we focus our attention, as well as
Its mission is to expand the economic
their online participation, privacy, and how we make decisions differently.
and political power of disabled people
property. She describes three major Clocking in at less than four minutes,
in the U.S. and abroad.
activities that people do online, that the narrator also quickly explains
http://www.aapd.com/
hold consequences for themselves, “phantom vibration syndrome,” which
is extremely common. https://www.
youtube.com/watch?v=HffWFd_6bJ0

FEATURED EVENT EVENTS
CONFERENCES CONTESTS & EVENTS
Second International Conference on DesignCon

Human Computer Interaction Theory The 23rd edition of this global event
and Applications (HUCAPP 2018) will bring together more than 5,000
Vila Galé Santa Cruz Hotel professionals from the high-speed
Funchal, Portugal communications and semiconductor
January 27-29, 2018 communities for three jam-packed
http://www.hucapp.visigrapp.org/ days of education and activities at
the Santa Clara Convention Center.
Beginning on January 30th, there
Sixth International Conference on
will be 100-plus sessions across
TEI 2018 Green and Human Information
14 tracks offering state-of-the-art
Stockholm, Sweden Technology (ICGHIT 2018)
design methodologies, applications,
March 18–21, 2018 Holiday Inn Chiangmai
technologies, and unparalleled
Chiang Mai, Thailand
The last two decades have seen networking opportunities.
January 31-February 2, 2018
the rise of technologies, like http://www.designcon.com/
http://icghit.org/
smartphones and ubiquitous
systems, which are so integrated Game Developers Conference (GDC)
with human life they have 14th International Conference on
GDC is a prime forum featuring more
become inseparable from it. As Technology, Knowledge, and Society
than 500 lectures, panels, tutorials,
the distance between technology St. John’s University
and roundtable discussions for
and daily human life decreases, New York, NY
engineers, game developers, audio
a trend of using old-school March 1-2, 2018
professionals, artists, producers, etc.
technology is beginning to emerge. http://techandsoc.com/2018-conference/
on multiple comprehensive subjects
Consequently, the theme of TEI related to game development and
2018 is “Post-digital Design,” which 13th Annual ACM/IEEE VR technologies. From March 19th
is inspired by the increasing use International Conference on to March 23rd the Moscone Center
of so-called vintage technologies Human Robot Interaction in San Francisco, CA will be home
like vinyl records, and the usage of (HRI 2018) to the conference. GDC also features
natural materials in everyday lives. McCormick Place the Independent Games Festival,
The conference has some Chicago, IL which is one of the largest and the
suggested topics, including: March 5–8, 2018 longest running festival related to
“Novel interaction techniques that http://humanrobotinteraction.org/2018/ independent games worldwide.
bridge the physical and digital http://www.gdconf.com/
worlds, wearable technologies
beyond the smartwatch, [the] role 12th International Conference on
of physicality in embodied sense- Tangible, Embedded and Embodied alt.chi
making (e.g., human perception, Interactions (TEI 2018) Known informally as the “avant-
cognition and experience), and Stockholm, Sweden garde of CHI,” alt.chi is a forum
philosophical, ethical and social March 18–21, 2018 for controversial, unusual, thought
implications.” https://tei.acm.org/2018/ provoking research. It was created
This is an interdisciplinary by the CHI program committee to
conference that features a provide a venue for innovative work
CHI 2018
separate “arts” track, which one that is often ignored. Authors can
Montréal, Canada
doesn’t often find at technology submit their work in alternative
April 21-26, 2018
conferences. ways: dance, spoken word, music,
https://chi2018.acm.org/
TEI 2018 will be held in and other forms of expression are
Image by Nattee Chalermtiragool
Stockholm, Sweden. Visitors encouraged. Authors are invited to

to the conference can also HCI International 2018 share “bold, compelling, critical, and
indulge in nearby tourist Caesars Palace innovative works that challenge or re-
attractions like the Vasa Museum Las Vegas, NV imagine human computer interaction
and the Stockholm Palace. July 15-20, 2018 research and design.” The deadline
https://tei.acm.org/2018/ http://2018.hci.international/ for submission is January 15, 2018.
https://chi2018.acm.org/authors/alt-chi/

FEATURED EVENT
depending on the destination country.

GRANTS, SCHOLARSHIPS, AND Explanation: The Intern Group
FELLOWSHIPS leadership grants for students and
recent graduates wishing to study
The Rainbow Scholarship Fund abroad, in order to pursue six-week
for LGBTQ+ Students Wishing summer programs in Colombia,
to Study Abroad Australia, Shanghai, or Dublin. The
http://www.rainbowsig.org/rainbow- scholarship is designed to cover 50
scholarship/ percent of the course fee.
Deadline: January 11, 2018
Eligibility: LGBTQ+ students Out to Innovate Scholarship
(U.S .citizens, permanent residents, for LGBTQ+ Students in STEM
or DACA Students) currently enrolled http://www.noglstp.org/programs-
at an American university or college, CHI 2018
projects/scholarships/ Montréal, Canada
and who will spend at least four weeks Deadline: June 2, 2018
studying in the U.S. between Summer April 21–26, 2018
Eligibility: Students pursuing
2018 and Spring 2019. degrees in science, technology, From introduction of the computer
Benefits: Up to $10,000. engineering, or mathematics (STEM) mouse in the mid-20th century to
Explanation: The Rainbow programs at a U.S.-based college or the use of VR headsets now, the
Scholarship awards deserving university who are active supporters discipline of HCI has come a long
lesbian, gay, bisexual, transgender, of and participants in programs or way. Devices like laser projection
and queer (LGBTQ+) students who organizations that promote LGBTQ keyboards don’t even require
aim to participate in high-quality, inclusion and visibility. physical buttons to be pressed! As
rigorous education abroad programs. Benefits: $5,000 per student the technology advances, we are
To be considered for the Rainbow Explanation: Two scholarships are bound to see a lot of innovation
Scholarship, applicants must self- available for the 2018/2019 academic from the industry and academia.
identify on the Fund for Education year, one for an undergraduate and CHI 2018 aims to bring
Abroad (FEA) one for a graduate student, to promote together researchers, designers,
(http://fundforeducationabroad.org ) academic excellence and increased and practitioners to interact
scholarship application. visibility of talented LGBTQ+ students and encourage the sharing of
in STEM careers. ideas in this domain. There are
The Intern Group Scholarships also various demonstrations as
for 2018 well as weekend GameJams and
https://www.theinterngroup. ScienceJams, wherein attendees
com/scholarships/?utm_ will have an opportunity to design
source=Studyabroad.com&utm_ and implement their own games
medium=Scholarship%20page&utm_ and engage with science. Another
campaign=Scholarship interesting track is “alt.chi,” which
Deadline: February 2, 2018 encourages people to present
Eligibility: Current students and controversial questions and
recent graduates for the summer/fall criticize accepted practices in the
of 2018. field of CHI.
Benefits: Either $2,000 or $2,500, The conference will be held
in Montréal, where conference
goers can also visit attractions like
the Notre-Dame Basilica and the
Botanical Garden.
https://chi2018.acm.org/
Image by Alex Papp

end
BEMUSEMENT
Inner Gollum
Source: http://www.phdcomics.com/synd/c/phd091517hswy.tif
Time Travel Thesis
Source: https://imgs.xkcd.com/comics/time_travel_thesis.png
Theft Quadrants Puzzle:

Love in Distealtopeia
Source: https://imgs.xkcd.com/comics/theft_quadrants.png
Nathan and Kristen have fallen in love (via the internet)

and Nathan wishes to mail her a wristwatch. Unfortunately,
they live in the country of Distealtopeia, where anything
sent through the mail will be stolen unless it is enclosed in
a padlocked box. Nathan and Kristen each have plenty of
padlocks, but none to which the other has a key. How can
Nathan get the wristwatch safely into Kristen’s hands?
Source: http://www.ipi.edu.pl/en/it-s-interesting-to-know/
347-love-in-kleptopia
Find the solution at:http://xrds.acm.org/bemusement/2018.cfm
SUBMIT A PUZZLE
Can you do better?
Bemusements would like your puzzles and mathematical games
(but not Sudoku). Contact xrds@acm.org to submit yours!

ACM Books.
In-depth.
Innovative.
Insightful.
ACM and Morgan & Claypool Publishers present ACM Books: an all-new series
of educational , research and reference works for the computing
community. Inspired by the need for high-quality computer
science publishing at the graduate, faculty and
professional levels, ACM Books is affordable,
current, and comprehensive in scope.
ACM Books collections are available
under an ownership model with
archival rights included. We
invite you to learn more about
this exciting new program.
For more info please visit
http://books.acm.org
or contact ACM at
ACMbooks-Info@acm.org
M
Association for
Computing Machinery
&C
2 Penn Plaza, Suite 701
New York, NY 10121-0701, USA
Phone: +1-212-626-0658
Email: acmbooks-info@acm.org
M
Morgan & Claypool
Publishers
&C
1210 Fifth Avenue, Suite 250
San Rafael, CA 94901, USA
Phone: +1-415-462-0004
Email: info@morganclaypool.com
IntelligenceCareers.gov/NSA
It’s not only WHAT you do,
It’s WHY you do it.

What’s your expertise? Computer/Electrical Engineering
Is it data science or neural networks? Advanced machine Computer Science
learning or artificial intelligence? Cybersecurity
Data Science
Whatever your technical specialty, the National Security
Agency has a position that mixes your expertise with our Information Assurance
mission: Mathematics
Cryptanalysis
The safety and security of our families, our communities Signals Analysis
and our country. Security & Counterintelligence
Paid Internships, Scholarships & Co-op
Apply today and many more
IntelligenceCareers.gov/NSA
U.S. citizenship is required.

NSA is an Equal Opportunity Employer.

Crossroadsxrds2017winter DL

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Crossroadsxrds2017winter DL

Uploaded by

Copyright:

Available Formats

XRDS

Who can participate?

PHOTOS: ©HLFF / B. Kreutzer (top);

2 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

26 FEATURE 44 FEATURE 55 BACK

30 FEATURE 48 FEATURE 56 HELLO WORLD

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 3

Issue Feature Editor

4 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 5

6 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 7

73% of Stanford faculty were men, and almost

sidered taboo, undesirable,

8 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

the forefront of innovation

depth interviews to explore Technologies that im- DOI: 10.1145/3155130

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 9

Transgender people are four times as

The Excessive Power of Ctrl+C and Ctrl+V

10 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

to verifying all the past papers of the

and M.Tech. in computer science from the National

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 11

The E.U. member state with the lowest

ADA ACM’s Role in

12 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

and professionals from industry as well. This is a programming competition

King-Noel, Countess of Lovelace. She pants were tasked to solve a real-life

was popularly known as Ada Lovelace, engineering challenge created by F A LL

Babbage on the mechanical general- known as “Hour of Code in Azerbaijan:

younger generation. community. “We hope that there will Stu

traveled around the globe, offering er science throughout Azerbaijan. den

design thinking, collaborative prob- DOI: 10.1145/3155222

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 13

2013 Women in India only earned 75.19%

Software is for Humans

14 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

2012 The location-based social search app Tinder

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 15

Steve Jobs was the son of Abdulfattah Jandali,

16 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

For example, a set of sensors detects

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 17

Figure 1. A workbench for this FPGA project; on the left is

WHAT IS A FIELD PROGRAMMABLE GATE ARRAY?

18 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

CIRCUIT BUILDING FUNDAMENTALS

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 19

20 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

Now accepting submissions to ACM THRI

Founded in 2012, the Journal of HRI has been serving as the

Since that time, the human-robot interaction field has

THRI now joins the ACM portfolio of highly respected

Editors-in-Chief Odest Chadwicke Jenkins of the University of Michigan and Selma

The inaugural issue of the rebranded ACM Transactions on Human-Robot Interaction is

22 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 23

24 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 25

26 XRDS • W IN T ER 2 017 • V OL .24 • NO.2

XRDS • W IN T ER 2 017 • V OL .24 • NO.2 27

28 XRDS • W IN T ER 2 017 • V OL .24 • NO.2