Principles of Maker and DIY Fabrication: Enabling Design Prototypes at Low Cost

Principles of maker and DIY fabrication:
Enabling design prototypes at low cost

Bradley Camburn and Kristin Wood, Singapore University of Technology and
Design & Massachusetts Institute of Technology International Design
Centre, 8 Somapah Road, Singapore 487372, Singapore
Do-It-Yourself (DIY) fabrication is the practice where the end user creates a
product for personal use rather than commercial production. This paper reviews
how DIY practitioners can produce useful artefacts with limited resources.
Fabrication principles were extracted from the DIY design repository
Instructables.com. A set of candidate principles was iteratively refined and
converged to five unique principles. Case studies are presented that illustrate
approaches for implementing each. A first empirical study verifies the
repeatability of the principle classification through crowd-sourced assessment. A
hypothesis is that the principles seen in DIY fabrication can support design
prototyping. This is validated through an empirical study that shows a positive
correlation between exposure to the principles and enhanced design outcome.
Ó 2018 Elsevier Ltd. All rights reserved.
Keywords: prototypes, design practice, distributed design, case studies, social

design
P
rototyping is critical to success in the early stages of design. For
example, projects without a working prototype are rarely funded on
the incubation platform Kickstarter.com. Empirical design research
of prototyping has provided: strategic methods (Camburn, Dunlap et al.,
2015; Menold, Jablokow, & Simpson, 2017); effects of timing efforts (Elsen,
H€aggman, Honda, & Yang, 2012; Faas, Bao, & Yang, 2014; H€ aggman,
Honda, & Yang, 2013, Neeley Jr, Lim, Zhu & Yang 2013; Yang, 2005); out-
comes of process techniques such as iteration or parallel prototyping (Dow
et al., 2010); and studies of fixation (Viswanathan and Linsey 2010, 2011;
Youmans, 2011). Existing methods provide high level planning of the proto-
typing process. There is substantial opportunity to explore the fabrication of
prototypes. Knowledge of means to reduce effort and improve build quality
of prototypes would complement existing research. What then, are principle
techniques of successful prototype fabrication? One avenue for exploring
this question is to review emergent do-it-yourself (DIY) design repositories.
These repositories are unique as they provide freely accessible documentation
Corresponding author:
Bradley Camburn of the development process. This work follows a systematic review of the
bradley_camburn@ open-source database Instructables.com, to extract principle means of fabri-
sutd.edu.sg cation. The demographics, socio-economic implications, and origins of the of
www.elsevier.com/locate/destud
0142-694X Design Studies -- (2018) --e--
https://doi.org/10.1016/j.destud.2018.04.002 1
Ó 2018 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Camburn, B., & Wood, K., Principles of maker and DIY fabrication: Enabling design
prototypes at low cost, Design Studies (2018), https://doi.org/10.1016/j.destud.2018.04.002
do-it-yourself (DIY) movement have been established in other works
(Anderson, 2012; Kuznetsov & Paulos, 2010; Paulos, 2013; Triggs, 2006). A
research area of potential interest to designers are overlaps in context faced
by DIY developers and those faced by designers and engineers engaged in
prototyping for new product, service, and system development. Herein, a pro-
totype is defined as a pre-production functional representation of some aspect
of a product service or system. These overlaps may include: exploration of a
technology that is novel to the maker, limited resources, limited time, limited
equipment for fabrication, and a requirement to establish basic proof of
concept functionality rather than to deliver a commercial product as-is.
1 The Do-It-Yourself design movement

What is DIY design? DIY design is typically implemented outside of the
framework of professional design, for the purpose of practical gain
(Edwards, 2006). It may often be an individual activity, however sharing com-
munities have recently evolved (Rosner & Bean, 2009). Although research sug-
gests that DIY communities are social in origin (Triggs, 2006), there are
currently numerous platforms which are highly technology oriented. Informa-
tion moves organically in these communities. Experts may seed forums with
extensive topic knowledge (Torrey, Churchill, & McDonald, 2009). This ex-
change can directly lead to the mutual benefit of participants (Salmond,
1997). In turn, a forum of creativity is emerging where open-sharing is highly
valued (Kuznetsov & Paulos, 2010). The paradigm permits individuals a self-
reliance to modify or develop certain technologies (Atkinson, 2006). It is
possible for individuals without a technical background to fabricate a cell-
phone, or other complex tools (Mellis & Buechley, 2014). This does not elim-
inate the need for large-scale manufacture of basic components; however, it
does act to democratize innovation through information sharing (Mellis &
Buechley, 2014). DIY development can share benefits of craft such as high
quality aesthetic (Arnheim, 1994). The opportunity is much greater than this
alone, genuine technological advancements are within the scope of non-
experts (Atkinson, 2006).
What opportunities does the DIY movement present for design research?
Often in a traditional design scope, needs assessment is conducted in partial
isolation from the user (Reich, Konda, Monarch, Levy, & Subrahmanian,
1996). For DIY projects, the designer is often also the final user. Open sharing
permits iterative evolution as each participant advances a design to fit their
needs (Saakes, 2009). The development of personal fabrication also provides
a novel design arena. Open-source part databases (e.g. Thingiverse, Shape-
ways), in combination with free modelling software (Mota, 2011), and low
cost digital manufacture (3D printers or laser cutters) make design and fabri-
cation of geometrically complex parts a desktop activity (Mota, 2011). Distrib-
uted manufacturing networks (e.g. Alibaba) provide means of accessing
2 Design Studies Vol -- No. -- Month 2018
components and materials in low volume directly from suppliers (Anderson,
2012). Other platforms (e.g. Instructables, Make, Highlowtech, Reprap,
Opendesk, DIYlife) provide project guidance (Mota, 2011). There are com-
panies deploying DIY centered hardware and software (e.g. Arduino, Ada-
fruit). Labs are experimenting with DIY enabling hardware including: paper
mechatronic platforms (Saul, Xu, & Gross, 2010) or haptic interface designs
(Moussette & Banks, 2011). Open source software permits algorithmic gener-
ative design, to reduce to effort of digital modelling (Devendorf & Ryokai,
2014). Hacker-spaces, incubators, and online platforms provide outlets for so-
cial exhibition, collaboration, and inquisitive exploration. This emergence op-
erates in parallel to industry and research (Lindtner, Hertz, & Dourish, 2014).
In this framework, designer, manufacturer, supplier, and consumer act as
distributed networks. A similar notion is explored in literature on product re-
finements implemented by end-product users in extreme use cases. So called
‘Lead Users’ (Hippel, 2005) engage in similar modifications to DIY practi-
tioners. This work explores a community that emphasizes documentation
and open sharing. The reasoning is this database supports rapid evaluation
of many case studies, from diverse sources. There is an ongoing opportunity
to connect directed research with these emergent activities (Paulos, 2013).
2 Design principles identification and characterization

2.1 Identification
Koen (1988) identified heuristics, or principles, as fundamental to the engi-
neering approach. Principles provide aid in the solution of a problem but
may not directly provide a concrete solution themselves (Fu, Yang, &
Wood, 2016; Singh et al., 2009; Stone, Wood, & Crawford, 1998; Tilstra,
Backlund, Seepersad, & Wood, 2015). In industry, design principles are often
codified as standards (e.g. ISO9001). Contextual flexibility is a property of
design principles. One means of identifying principles is through categoriza-
tion and classification. These are critical as mechanisms to represent large
amounts of information in a compact form (Wood & Linsey, 2007). This
work expands preliminary efforts (Camburn, Sng et al., 2015; Perez,
Anderson, & Wood, 2015) to identify implicit principles for prototype fabrica-
tion through analysis of a DIY open source sharing platform, Instructables.-
com. It has been observed that radical innovation is human centered, and
includes novel interpretation of meaning (Norman & Verganti, 2014). There-
fore, this study adopts an ethnographic analysis of the given repository.
Prototype fabrication principles are extracted from projects on the Instructa-

bles.com database. This database was originally developed by MIT media
lab, and Squid labs. It is an online database for documentation and sharing
of open-source designs. At the time of initial assessment, January 9th, 2015,
there were a total of 152 729 entries counted on the database. Of this total,
Principles of maker and DIY fabrication 3
37 370 were listed in the Technology category. The database is continually
expanding.
The Technology article category of Instructables.com was selected for analysis

due to its analogous overlap with industrial design and engineering prototype
fabrication. In order to select articles the following criteria were applied: 1) for
diversity, an equal number of articles was taken from under each of the
‘branches’ of technology (e.g. 3D printing, assistive tech, sensors); 2) for rele-
vance, articles were sorted in descending order of views as a form of communal
vetting is indicated by frequent views; 3) finally, only articles that featured an
integrated standalone artefact were selected (e.g. not generic processes such as
‘etching’). Separate studies of processes or other database categories could be
pursued. These criteria repeatably identify artefacts resembling design proto-
types. The following information was recorded for each article in an initial
sampling (n ¼ 70): 1) parts list; 2) tools list; 3) fabrication procedure. The re-
searchers read each article, extracted the above lists and recorded quotes where
the Instructable author discussed a challenge and how it was overcome. To
identify principles, the researchers employed an inductive ethnographic
approach (Life, 1994). A set of rules defining a principle were agreed on by
two separate investigators who reviewed the articles. The principle should:
1) lead to a reduction of cost or improvement of performance as evidenced
by resolving a problem 2) it should be generalizable 3) it should be actionable.
A saturation process was used to determine when a sufficient sample size was
reached. As each new article was analysed a list of principles was started.
When a principle was identified, the reviewer would examine existing princi-
ples. If the new principle was not functionally equivalent to an existing princi-
ple already listed, then a new category is formed. If the principle was already
listed, a second entry in that category was cited. When a set quantity, in this
case n ¼ 15, of new articles was evaluated without identifying any new princi-
ples, the analysis stops. See Figure 1 for a record of the saturation analysis.
This convergence criteria is defined as ‘saturation’ (Francis, et al., 2010).
The initially identified set of principles appeared to consist of common themes,

see Appendix A. The researchers successively binned individual principles into
more abstract principle categories, see Figure 2. Principles were binned until
no further binning could be achieved without a substantial loss of information.
The process converged after 14 iterations. Inter-rater agreement criteria
(Pearson’s > 0.8, for assignment in new bins) were employed. The investiga-
tors invited outside reviewers to critique the phrasing and objectivity of the
principles against the articles and revised accordingly. Finally, a sample of ar-
ticles was then supplied to a pair of researchers outside the core team, who in-
spected them for principles. They were not exposed to the initial results. These
new principles were then successfully binned into the existing categories with
full agreement of both raters.
50
Unique Practices Identified

40
30
20
10
Figure 1 Data saturation for 0

extraction of specific 0 10 20 30 40 50 60 70
practices Number of Entries Evaluated
Figure 2 Sample of synthesis

process. Three individual
practices grouped into one
generalized principle
2.2 Design principles characterization

Three independent validation efforts were employed. In the first study, a
dramatically larger sample size was evaluated. One thousand entries were
manually reviewed. The top 30 entries, as sorted by views, from each Technol-
ogy channel were selected. Articles were selected using the same criteria as in
the initial sampling, for analogous relevance to prototyping. This increased
sample verifies that the principles are in prevalent throughout the database.
For each article, the researchers reviewed the article, and cited which, if any of
the principles was relevant. A ranking scale was also applied to assess, accord-
ing to the researcher’s judgement, whether the project could have been possible
without application of the principle (e.g. for a DIY oscilloscope which utilizes
the cathode ray tube extracted from a CRT television, repurposing was
deemed most critical; while for a laser-cut wooden lattice laptop stand, layer-
ing was deemed most critical). This assessment indicates the relative criticality,
and distribution of principles in the database.
The results of this cataloguing procedure, Table 1, show the raw percent of
cases with each principle. Ranked importance of the principles was also re-
corded. This rank is used to identify the comparatively most critical principle.
The goal of this is to determine the relative importance of the principles across
Table 1 Results from extended sample study (n [ 1000)
Principle Count of critical entries Total percent of entries with application
Repurpose commercial products 430 91%

Satisfice component quality 267 95%
Stencil fabrication patterns 216 71%
Standardize fabrication process 51 64%
Layer structural assemblies 36 72%
the database. Results were evaluated through inter-rater agreement. Pearson’s

correlation, was r ¼ 0.52 for all the categories except standardized fabrication
and satisfice component quality. Though this is a large discrepancy value, each
of these disconnects was reviewed and reconciled through discussion until
r > 0.8 was achieved. Several principles appeared more often than others.
Most articles also exhibited the application of more than one principle (the
average number of principles per article is 3.84, n ¼ 1000, s ¼ 1.2). Surpris-
ingly, each entry selected exhibited at least one principle to some degree.
A second validation consisted of deploying surveys to an online crowdsourced

databased to determine whether a large set of independent raters would iden-
tify the corresponding principles in each article. Online crowd sourced analysis
combines parallelization of computational analysis with the advantages of hu-
man reasoning (Liddament, 1999). Parallel decomposition has also been
demonstrated with tasks requiring human intelligence. One example is the Fol-
dit project. In this project, individuals solved complex protein folding prob-
lems via a game-like environment. A result of this project was successful
identification of the precursor pathway for a critical enzyme that could poten-
tially lead to renewable fuel production (Eiben, et al., 2012). Online raters have
volunteered to provide distributed ecological monitoring (Cooper, Dickinson,
Phillips, & Bonney, 2007), identification of astronomical bodies, and quantum
physics modelling (Grey, 2011). This method has potential risks, as anonymity
of the workers may influence participation styles (Cox, 1999). This can be miti-
gated via careful pilot testing.
Multiple parallel inter-rater agreement for presence of the principles was de-
ployed through online crowd-sourced assessment. Online rater assessment is
used in addition to inter-rater assessment among the researchers. This provides
several benefits, including: outside opinions, need for clarity of definitions, and
larger samples of raters. This approach provides a novel statistical robustness
beyond traditional inter-rater assessment. Amazon’s tool, Mechanical Turk,
was employed.
A series of pilot studies were conducted to evaluate design of the surveys. The
procedure was based on the formulation for crowd-sources studies outlined by
Wu, Corney and Grant (2015). In principle, it is important to support large
crowd-sourced studies through careful assessment of the results of small pilot
studies. The researchers evaluated the survey with support of a cognitive psy-
chologist to avoid confirmation bias.
A general observation of this procedure is that questions must be clear and

explicit to achieve useful results with crowdsourced platforms. This is likely
because the researchers and respondents never interact directly in any way,
nor is there any contextual hints provided to the respondent (such as the
location of the study, etc.). The final question format was the result of
twenty-three iterative survey designs. There were five Instructables posted
in each set. There were five individual survey types. These numbers were
selected based on the cost and re-imbursement expectations of the MTurk
marketplace established by the pilot trials to determine an appropriate
task loading per survey. Each type focused on evaluating one principle.
There were twenty-five distinct raters for each article-type combination.
This resulted in 625 unique responses from each set. This number was based
on calculations of a 0.95 0.05 confidence interval given the Instructables
database population size. Quality was assessed as the number of usable re-
plies. The number of relevant replies from set-A was relatively low (60%).
Many replies from set-A referred to ways of potentially improving the given
prototype, e.g. for a 3D printed robotic lamp a reply was ‘consider using
wood instead of 3D printing to make it less expensive’. This reply indicates
a misinterpretation of the question. As shown from Table 2, the question
asks the respondents to find ‘re-applicable’ features of the existing design.
The frequency of misinterpretation indicates that the questions in set-A
were misleading. For set-B, the percentage of usable replies was much higher
(91%). Note that the sample question is more explicit, and the number of
usable replies was also much higher. This drastic increase in correct interpre-
tation was achieved through improving the question clarity. It can also be
seen that the time to complete the survey increased for set-B, indicating
that more time was spent answering the question. Note that order of posting
was not a factor. The set-A questions were re-posted as a test after set-B for
a new set of articles. The percentage of usable replies remained about the
same, approximately 60%.
Table 2 Sample of quality assessment and question redesign process
Set N-Replies N-Relevant Average time (s) Sample question e structure principle
Replies (1 of the 5 types)
a 625 379 181 11 Look for a feature that adds durability to the
example prototype.
How could this be re-applied, to a different
design, to make it durable?
b 625 568 356 17 What is the cleverest structural (strength adding)
feature in this prototype?
Describe it generally- so that someone else might
be able to do the same thing for a similar design.
Across the study, there were two primary stages of assessment. In the first
portion, five entries were selected for each of the five principles. Twenty-five
raters assess each article. For these entries, control is introduced by posting
all five of the principle variant surveys for each example. Some of the examples
exhibit up to three principles but the researchers did not identify all five prin-
ciples in any of the posted samples. Therefore, there is an opportunity to match
both positive and negative identifications between the respondents and the
researcher. This initial set of twenty-five examples was evaluated using a sur-
vey where respondents provide an open ended response to determine if a prin-
ciple is present. A corresponding binary ‘yes/no’ indication of principle
presents was provided afterwards. This provided an initial test for matching
with detailed answers from the respondents. There was a high degree of match-
ing between the open ended reply and binary testing for the principle. A second
and substantially larger set was deployed with only a binary principle test. For
this set, there were 125 examples that each had been rated by the researcher to
contain one of the principles. There were 25 examples for each principle. A to-
tal of 4385 online assessments were made across both of these stages.
The technique of multiple raters is adapted from Green (Green, Seepersad, &
H€oltt€
a-Otto, 2014). A simplified Fleiss’ Kappa is employed to evaluate the
multiple-rater agreement. Due to design of the crowd-sourced platform
used, each of the raters do not necessarily complete more than one entry.
For Fleiss’ Kappa, N ¼ 1 and the form for Fleiss’ Kappa is simplified as in
Equation (1) (below).
" ! #
1 X
k
Pi ¼ n2ij ðnÞ ¼ k ð1Þ
nðn 1Þ j¼1
where Pi is the agreement for raters on a given topic (principle), N is the num-
ber of subjects (entries), the categories are indexed by j from 1, .,k; and nij
represents the number of raters who assigned the ith subject to the jth cate-
gory; Pi ¼ k only in this special case where N ¼ 1.
The Fleiss’ Kappa is used to develop a ‘Synthesized Rater’. In effect, only

those results with high agreement can be considered as statistically significant.
These significant positive identifications are compared against the researchers’
initial observations for principle presents. In this study, a positive identifica-
tion of a principle in an example is considered only in cases where Fleiss’
Kappa agreement was above k ¼ 0.8 between raters that the principle was
present.
Agreement results are shown in Table 3. Table 3 provides a summary of agree-

ment and disagreement between the researcher and the Synthesized rater. For
each principle, 30 articles in which the researcher found the principle were also
Table 3 Principle identification matches between researcher and Synthesized rater
Principle (n ¼ 30 for each) Synthesized rater found Synthesized rater did not find
same principle as human same principle as human
Repurpose commercial products 29 1

Satisfice component quality 29 1
Stencil fabrication patterns 28 2
Standardize fabrication process 29 1
Layer structural assemblies 29 1
assessed by the online raters (as detailed above). For example, in repurpose
commercial products, both the researcher and the Synthesized rater agreed
that a principle was present in 29 cases. There were also 18 negative cases
(without principle) explored. These were not included in the table for clarity.
Pearson’s correlation between the researcher and the Synthesized rater is
r ¼ 0.85 for across the full set of samples. Overall, this study provides confirma-
tion that the researcher and crowd raters agree on presence of the principles.
3 Maker fabrication principles case studies

3.1 Principle 1: Repurpose
Repurpose commercial products to reduce the effort and cost required to
achieve functionality. Repurposing, sometimes referred to as ‘hacking’, is a
means to reduce effort and cost required to achieve subsystem functionality.
Typically, one or more key components of an existing commercial product
is modified and re-deployed as a subsystem or component of the prototype.
This principle was initially verified by systematically listing the source of
each component in the sample entries and how it was employed. This is critical
because it is very difficult to achieve high precision in fabrication in a prototyp-
ing context. Custom orders for a precision component or subsystem are often
expensive and preclude a long lead time. This is due to the design and tooling
efforts required. However, there are many precise functional subsystems in
readily available commercial products. Commercial products are often priced
so competitively, due to their mass-produced nature, that it is far easier to
cannibalize an existing machine for its precision components than to acquire
these components directly or according to custom order. While compromise
on details of the design may be required, it is offset by cost reduction.
3.1.1 Repurpose case study 1: Atomic force microscope

This low cost atomic force microscope (AFM), see Figure 3, demonstrates
component repurposing. A demanding requirement in AFM design is the mo-
tion stage nominal step size and allowable deviation. The other key compo-
nents, piezo actuators, structural frame, probe, and piezo drive circuit are
all available from online stores in the tens of dollars. Precision stages, when
procured as such, typically cost thousands of dollars. The innovative approach
Figure 3 (left) Extracted DVD read head; (center) assembled low cost AFM; (right) sample scanning image (whoand)
of this solution was to cannibalize the motion stage of a DVD read head. A
DVD read head can be bought, used, on eBay at less than USD 10$. The min-
imum step size on a DVD read head is around 1 um. This project demon-
strated a functioning AFM with micrometer resolution costing around USD
100$. A comparable AFM, would be on the order of magnitude cost USD
50 000e100 000$.
3.1.2 Repurpose case study 2: Eye tracking interface

For an artist, motility is nearly equitable with survival. Graffiti artist Tony
Quan ‘TEMPT1’ was challenged to continue his work after the degenerative
nerve disease ALS left him paralyzed from the neck down and only able to
move his head. A team of designers repurposed a general motion tracking de-
vice to track the pupil, see Figure 4, so that Tony could continue producing art
even after his tragic accident. An eye tracking system typically costs on the or-
der of USD 15 000$. In this system the critical component is a real-time point-
to-point motion tracking camera. To develop just this embedded processing
module from scratch would take several engineers, and a few years of develop-
ment. The design team realized that a very similar function module was avail-
able for hand-gesture recognition in several popular gaming consoles. The
Sony Visual Communications Camera was selected for repurposing. It includes
an IR source (for stable and consistent reference illumination), a high-
resolution camera, and most importantly, embedded real-time firmware to
enable point-to-point tracking. The focal length of the original module was
incorrect, however relatively easily modified. The camera was then mounted
Figure 4 (left) The Eyewriter; (center) using the Eyewriter to create urban art; (right) art installation created with Eyewriter and digital light
projection (Q-Branch)
to a pair of low cost sunglasses. Again, the sunglasses were also repurposed to
reduce cost. The design team was able to produce a reliable eye-tracking device
for Tony which costs around USD 300$. The artist was able to continue socially
responsible urban projection graffiti art with the Eyewriter design.
3.2 Principle 2: Satisfice

Satisfice component quality to reduce expense on materials and components by
adopting the minimum viability approach. The idea is to start with a minimum
cost part. This may even include using extremely low-cost material stock like
cardboard and aluminum foil. These materials can be tested easily and up-
grade only when needed. In this way, satisficing on material and components
enables more iteration. Those materials or components that fail can be re-
placed with higher quality components. The need for quality is determined
by the solution rather than an assumption of starting with high-grade mate-
rials. This goes hand in hand with attaining a deeper understanding of the crit-
ical phenomena, to again, improve only what is needed.
3.2.1 Satisfice case study 1: Capacitive 3D mouse

In this article, a set of tri-axial capacitive field sensors was constructed to deter-
mine the position of a human hand within a small volume, see Figure 5. In this
case, relatively complex field interpolation software (open source) was em-
ployed. However, the controller does not need to perform any functions
beyond reporting a change in voltage potential, therefore a simple Arduino
processor was sufficient. The capacitance sensor itself requires a thin conduc-
tive film with a large surface area. Aluminum foil was a perfect candidate. It is
readily available and low cost. This article exemplifies achievement of a com-
plex task with what may be considered low cost materials.
3.2.2 Satisfice case study 2: Laser stereolithography

Laser stereolithography is a type of additive manufacture, or 3D printing. The
designers of this article were able to produce a functional stereolithography
system for home use, see Figure 6. The motion control must be precise, there-
fore linear bearings and stepper motors were employed. However, the control
algorithm is relatively simple, so a low-cost Arduino processor was used. The
external shell can be sanded to refinement, so plywood was used. The resin
bath is a glass beaker. The resin bath is at once an example of satisficing
and repurposing.
3.3 Principle 3: Stencil

Stencil fabrication patterns to manage complexity and increase accuracy of low
cost fabrication. This principle implies the use of any representation of the
design to manage fabrication complexity. In the selected entries, sketches,
stencils, CAD models, and schematics are frequently used to either plan or
directly aid in fabrication.
Figure 5 (left) Software interface for 3D position sensing; (center) circuit schematic; (right) low cost physical implementation
(SarahandDillon)
Figure 6 (left) Control board pinout for stereolithography printer; (center) assembled machine; (right) sample print (RobHopeless)
3.3.1 Stencil case study 1: Word clock

The world clock is a wall mounted clock that displays time using illuminated
test strings e.g. ‘it is twenty minutes to eight’, see Figure 7. A single layer elec-
tronic printed circuit board (PCB) can be produced using only an inkjet
printer, and press-n-peel PCB transfer film. The word clock designer utilized
an inkjet printed stencil to produce an at-home PCB which manages control
of the numerous LED compartments required to illuminate the clock face,
and interface with the timing microcontroller. In addition, the inkjet press-
n-peel stencil technique was also later applied to produce a transparent projec-
tion mask for the illuminated words. A key takeaway of this principle is that a
single tool can be used for multiple purposes. In this case, press-n-peel film is
used for two very different applications. Stenciling and standardization are
both seen here.
3.3.2 Stencil case study 2: Fractal antenna

The fractal antenna is a simple receiver for high band VHF radio waves, see
Figure 8. The designer drew inspiration from the fractal receivers used in
cell phones to reduce the antenna size required to reliably attain phase lock
with a given frequency. Though fabrication of a high gain fractal antenna
would seem to require complex facilities, it does not. The antenna lens archi-
tecture must be carefully constructed, to ensure that the signal to noise ratio is
low for the desired frequency. An inkjet printer is again employed to produce a
Figure 7 (left) PCB circuit schematic for Word Clock; (center) printed circuit board; (right) printed light-mask for display (drj113)
Figure 8 (left) Printed stencil for fractal HD/VHF antenna; (center) weaving copper wire segments; (right) assembled antenna (tigers58)
low-cost direct design-to-fabrication approach that enables production of a

complex, precise component. This may also be repurposing, paper is used as
an assembly jig rather than for hosting a document.
3.4 Principle 4: Standardize

Standardize fabrication process to increase the efficiency of fabrication. Repe-
tition is a means to reduce overall costs from fabrication. The startup proced-
ure for any machine or toolset is non-zero. There is a potential to reduce
overall cost by producing multiple parts with one approach. This principle
was initially verified by listing the fabrication method of each custom compo-
nent in the entries.
3.4.1 Standardize case study 1: 3 axis CNC router

In this design project, the designer used a single method, in this case stencil
guides and a band saw to produce all the custom structural elements required
for this three-axis computer numerically controlled (CNC) milling machine,
see Figure 9. By utilizing a single approach, stencil and bandsaw, the maker
was able to produce many distinct custom components in a single process.
This is another example of co-occurrence.
3.4.2 Standardize case study 2: GLaDOS robotic lamp

The GLaDOS robotic lamp design demonstrates another case of standardized
fabrication, see Figure 10. In this case, a complex assembly was produced with
Figure 9 (left) Plans for wooden CNC router; (center) structural components, all cut using stencil and bandsaw; (right) assembled CNC router
(oomlout)
Figure 10 (left) GLaDOS robotic lamp rendering concept; (right) printed components of the robotic lamp; (right) assembled robotic lamp
(dragonator)
a desktop fused ’deposition modelling (FDM) printer. These parts are differ-
entiated by design and post processing. Structural are produced with thicker
walls. Aesthetic components require post-processing with paint and varnish
to obtain a desired appearance. This example demonstrates that components
produced with a single process may serve alternate objectives, be it structural
or aesthetic, etc.
3.5 Principle 5: Layer

Layer structural assemblies. Designers have balanced tension and compression
to form rigid structures for centuries. The term ‘tensegrity’ was coined by
Buckminster Fuller (Fuller, 1962). A tensegrity is a structure that can be
simultaneously conformal and rigid along different axes of motion. They
can also be used as a structural mechanism to improve strength to weight ra-
tio, as in aircraft wings. This is achieved through the integration of tensile and
compressive members. Structural design tactics like lattices and tensegrities
can be employed to enhance the strength of artefacts. Just as at the micro-
scopic scale, crystalline structure can enhance the rigidity of a material, so
can clever lattice deployment at the meso-scale. Lattice, tensegrity, and
origami-like structures improve the strength to weight ratio of a design
when implemented correctly. This principle works in harmony with the appli-
cation of low cost materials (e.g. paper, cardboard, plastic) as these are gener-
ally easy to cut and reform.
3.5.1 Layer case study 1: Paper clock
In this case study, the designer worked to construct a three-dimensional sup-
port structure to guide 28 servo-controlled display elements, see Figure 11. At
first glance, paper would seem like a material that is too soft for the structure.
However, the designers of this clock employed a layer bonding strategy to
form a rigid meta-structure from cardstock. When the cardstock layers are
assembled they form a rigid composite. The support structure was made
from layers of cut paper. Each cut has the necessary pass-throughs for drive
pins or the clock read out display elements. In the back, several layers are
used to make a housing for the servos. Each layer is bonded with an array
of adhesive coated small foam blocks, approximately 2 mm in thickness. These
also offset each layer to give the structure volume with a small number of
layers. This structure has a high strength-to-weight ratio and is easy to assem-
bly. This approach is faster than digitally controlled additive manufacturing
processes (Sass & Oxman, 2006). While it might not survive extended use, it
serves for concept evaluation.
3.5.2 Layer case study 2: Duct-tape canoe

The example of a duct tape canoe highlights such balance of tension and
compression, see Figure 12. A PVC wire frame formed the ribbing for the skel-
eton of a canoe. PVC is flexible to bend in off-axis loading, yet strong under
Figure 11 (left) A single layer of the digital to analog clock; (center) structural subassembly; (right) fully assembled digital to analog clock
(drive servos not shown) (alstroemeria)
Figure 12 (left) PVC structural core for the canoe; (center) wrapping the PVC core in duct tape; (right) functional, assembled canoe
(hyroc346)
direct compressive axial loading. It is suitable to form a rigid frame. Edges
were bound with duct tape. Duct tape was wrapped around the exterior to
form a tensile sheet to increase the structural rigidity and to act as a waterproof
layer. The key insight of this principle is that well planned meta-structures can
provide low cost means to produce large, rigid structures. These structures can
be made from materials that are easy to work with and allow for rapid itera-
tion and proof of concept testing.
4 Validation of DIY design principles

This study employs the preliminary data extracted from a controlled exper-
iment to evaluate use of the principles in design. This study took place in two
design course projects. The projects were run in two subsequent years so
there were no repeat participants. Participants took one week to intensively
design and construct a prototype. The control effort began and ended before
the experimental course to ensure that there was no coincidental knowledge
of the principles transferred between groups. For the control group, a tradi-
tional stage-gate prototyping methodology was adopted. In the experimental
group, the principles were introduced via presentation and examples. Partic-
ipants of the experimental course also provided self-evaluation via survey af-
ter the prototyping effort. There was a total of 550 individual participants in
the study. Participants were those students who opted to provide data for the
study. The design tasks were completed by teams of five individuals. The
control course had 61 teams, and the experimental course had 49 teams.
This difference is due to a slight variance in the class size between years.
One uncontrolled difference was the design challenge, though both problems
were similar in scope. The control course problem was design of a cooling
system for automated milk delivery. The experimental course problem was
to design a demonstration for given phenomenon in mechanics. These two
projects were slightly varied as they took place in courses with different
focus. However, they are both part of the standardized ‘Designettes’ curric-
ulum (Telenko et al., 2016). Therefore, scope, requirement stringency, and
variety among solutions were equivalent from the two courses. Project
budget, time allocation, and prototyping requirements of the two courses
were also the same. Participants all had equal opportunity and the same
design requirement to deliver a prototype. Figure 13 shows an example so-
lution from the experimental course. Repeated fabrication processes apply
to the wheels, structural supports, and inertial energy storage flywheels,
which were all laser cut. Include structural voids applies to the chassis, see
Figure 13.
The experimental group implemented significantly more principles in their

prototypes. This was determined by two independent raters who listed the to-
tal number of principles implemented in each physical prototype. Prototypes
were analysed by reviewing photographs of the prototypes and through review
Figure 13 Sample prototype
from experimental class, for
a gyroscopically stabilized
bike
of each team’s final reports. Two reviewers evaluated all the prototypes for use
of the principles without knowledge of the core hypothesis or which condition
prototypes came from, they were instructed to review a picture of each proto-
type and check off which of the known five principles, if any, were present via a
survey. Pearson’s correlation between the raters was r > 0.8. This analysis pro-
vides an evaluation of the connectivity of principles. This difference is signifi-
cant with Students t-test at p < 0.01 (shown in Table 4). The control groups did
on average apply two or three of the five principles in their prototypes. This
indicates that the principles can be arrived at by intuition or experience; how-
ever, exposure to the method can increase the degree of implementation.
Teams in the experimental group implemented roughly the same number of
principles in each prototype as were found in Instructables on average for
the extended study. There was no significant difference between the experi-
mental group prototypes and Instructables articles in number of principles.
There were significantly fewer principles applied in control versus Instruct-
ables articles. Instructables articles address a highly diverse set of problems.
Its authors can logically be expected to be at reasonably aware of methods em-
ployed in other Instructables. This provides an additional comparison against
the control group.
Prototypes in the experimental group were also rated by a panel of external

judges at the end of the course. Of all teams in the experimental group, eleven
were chosen to enter a final competition round as part of the course. This de-
cision was based on design and execution. Teams in the final competition
Table 4 Implementation of principles across groups
Group Average number of principles Standard error
Experimental teams 4.0 0.13

Control teams 2.9 0.14
Instructables entries 3.9 0.04
Table 5 Principle use between teams in the final competition and other teams
Group Average number of principles Standard error
Teams in final competition (n ¼ 13) 4.4 0.14

Other teams (n ¼ 37) 3.9 0.16
employed more principles in their prototyping, on average, than other teams.

These results are summarized in Table 5. The difference is significant with Stu-
dent’s t-test at p ¼ 0.05.
A third metric used to evaluate design outcome was the number of teams that
completed physical prototyping. A certain percentage of teams only provided
a design concept rather than a working prototype. This is a performance mea-
sure for overall success of the process. In this case completion of a prototype
can be used as a proxy for performance as delivery of a completed prototype
was an explicitly listed performance in the requirement of the exercise. A direct
comparison is made between the normalized percentage of teams that
completed a prototype between the experimental and control course. In the
control group, 51% of teams provided evidence of prototyping (documenta-
tion of a physical build). In the experimental group, 91% of teams provided
evidence of prototyping. Using a transformed z-test for difference of propor-
tions of two defective proportions the p value is < 0.001. Despite this calcula-
tion, there are two potentially limiting factors on the significance and
generalizability of the results. Firstly, while all other key variables were essen-
tially equivalent, there is a difference in design problem provided for each
experimental and control conditions. This variance may have impacted design
outcome. Secondly, the number of unique design prototypes in this component
of the study may be considered of low-to-moderate sample size for qualitative
analysis, therefore the correlation of exposure to the principles with increased
use of the principles in practice must be taken as indicative. This is in accor-
dance with best practices in design science, and replication with further study
is encouraged. This segment of verification acts only to support the core ethno-
graphic research.
4.1 Evaluation of Surveys

A survey was completed by participants of the experimental group. Self-
assessment survey results, see Table 6, indicate specific effects of implementing
the principles. Survey questions were designed to evaluate usage of a principle
versus the perceived utility of each principle. Use of each principle was re-
ported as a binary yes/no by participants. Utility was reported for an associ-
ated variable with a Likert reply according to the range: (2) strong
disagreement, (1) disagreement, (0) neutral, (1) agreement, and (2) strong
agreement. Prompt text and results are given in Table 6. There are several
Table 6 Results of surveys, significant results in bold font
Repurpose commercial products Components were easy to find Component quality was high
(yes/no response) (Likert response) (Likert response)
Used repurpose principle 0.1 0.6

Did not use 0.1 0.4
Student’s t-test, p 0.482 0.313
Satisfice component quality Fabrication required minimal effort Fabrication quality was high
Used basic crafting principle 0.8 0.5

did not use 0.6 0.7
Stencil fabrication patterns Schematics were accurate Schematics were helpful
Used blueprint principle 1.2 1.4

Did not use 0.7 0.7
Standardize fabrication process Fabrication required minimal effort Fabrication quality was high
Used repetition principle 0.5 0.7

Did not use 0.2 0.4
Layer structural assemblies The prototype was durable The prototype was light
Used structure principle 0.9 0.6

Did not use 0.3 0.3
significant comparisons in the survey response (bolded), others are indicative

but not significant. This may be due to the relatively small number of survey
participants. The utility of schematics to reduce effort and increase accuracy,
and structural layering to increase structural integrity were rated as significant
(see Table 6).
5 Discussion
The goal of this work was to identify whether there are actionable principles in
the DIY database, Instructables.com that can inform design. This study
explored the analogy between design prototyping and the construction of a
DIY product. When a designer is developing a new product, they share some-
thing in common with the DIY community. That is, they need to learn the
operating principles of the thing they look to build, and to seek a means of
developing it with limited resources. Existing research on prototyping
(Viswanathan & Linsey, 2012; Yang, 2005) indicates that early-stage, low
cost prototyping is critical to successful design. It is often a practical reality
that designers in industry are given very little budget or time to produce func-
tional prototypes. The cost savings methods of fabrication in this work pro-
vide insights to support industrial design practice. Two methods stood out
as highly prevalent in the Instructables database: repurposing commercial
products, and satisficing component quality.
The concept of re-purposing is critical. It allows the designer to benefit from

the cost reducing mathematics of scale. That is if the designer can identify
functional blocks to satisfy major requirements of a design that already exist
in commercial products, it is likely these products will be sold at much lower
cost than a custom build. The phenomenon of diminishing cost in large scale
production is well established. Likewise, design for manufacturing literature
has explored make/buy decision analysis. Experimental results indicate that
designers may or may not capitalize on this opportunity when prototyping.
There are many corollary and emergent phenomena that enable the flexibility
of re-purposing. When repurposing electromechanical products, a challenging
aspect can be adding control. Widely available embedded microcontrollers
that support programming in object-oriented languages (e.g. raspberry Pi, Ar-
duino) enable easy replacement of an artefact’s control system. Another chal-
lenge in re-purposing can be integration and module interfacing. Again, the
prevalence of desk-top additive manufacturing supports rapid manufacture
of interface modules. These are further supported by open source code data-
bases, free and open source modelling software, and digital repositories. The
greater challenge becomes identifying new and creative systems to produce.
The key take-away is that to accelerate innovation, designers can support pro-
totyping by searching for commercial products to replace key functional mod-
ules in the design.
Another key insight of the work is the principle of satisficing. Working from
concept to functional prototype often requires many iterations. Designers
can fail to predict use context and failure modes accurately (Norman, 1983).
To reduce the cost risk of unnecessary over-engineering, the DIY approach
is to start from the absolute minimum cost set of materials and simplest pro-
duction approach. Only when components fail, typically, will the DIY practi-
tioner upgrade to a higher cost, higher quality component or add a new
function. In this way, evolutionary prototyping is achieved with a low-cost
pathway to proof-of-concept.
The remaining principles were less prevalent in the database. Stencil fabrica-
tion patterns enables precise manufacture using common tools. This principle
had not been predicted by the researchers. The principle, standardize fabrica-
tion process, was seen to be born both of necessity (cases where the maker had
only one machine was available) or to reduce machine startup effort. Using a
single fabrication machine should also simplify the planning process (e.g. pro-
ducing different types of drawing or CAD design files). Layered structural as-
semblies borrow from state-of-the art architectural practices (Sass, Chen, &
Sung, 2016) and highlight ways that machines with relatively small fabrication
volumes can still be used to produce large structures. This is also supported by
the theory of tensegrities which give a guideline for enhancing the strength-to-
weight ratio of a structure by pre-loading oppositional tensile and compressive
members.
Ultimately, this work urges an open reflection on means of achieving quality in

design prototypes at a faster pace and with less cost. The single overarching
theme is a willingness to optimize what is available over seeking what is ideal
to achieve a specific goal. A designer who stops a project time and again to
wait 3 or 4 weeks for a custom part to arrive from a vendor or fabricator,
may ultimately fall behind the designer who is able to identify a nearby
resource that can serve the same function and acquire it within hours or
even minutes. This concept of deconstructing the environment and suiting it
to our purpose is fundamentally human. In fact, historic humans actively de-
constructed their environment by crafting tools, shelter, and other life-
sustaining artefacts from their surroundings (e.g. usage of timber in carpentry,
usage of stone in masonry). The same principle is at work here e exploration
of ways to deconstruct the surrounding environment into new and useful com-
binations. In many ways, the product development process remains rooted in
the manipulation of raw-resources found in the natural environment (wells,
mines, etc.). The fundamental take-away is to begin exploring, ‘What innova-
tions are possible when we perceive objects in the constructed environment as
if they were a natural resource?’
5.1 Limitations and future work

The research approach taken was to test for repeated principle use in the data-
base. This approach does not allow for identification of approaches that have
potential to positively impact design but were only present in a very small
number of cases (i.e. outliers). One benefit of this approach is that it substan-
tially reduces the risk of a ’false positive’ or identifying a principle that is not
generalizable. Future research could potentially explore evaluating the poten-
tial for non-repeated principles. One possibility to achieve this goal would be
to extend the crowd sourcing analysis to include sorting and selection of en-
tries as well as analysis of entries. To carry this work further, additional anal-
ysis might include using automated topical analysis to evaluate Instructables
or other such online repositories, or deeper ethnographic work that follows
the individual creators of particularly astounding articles. Finally, the authors
would encourage the community to pursue replication trials to evaluate
whether exposure to these principles successfully correlates with increased
application as well as what forms of representation are more effective at
achieving this goal.
Acknowledgements
The authors are grateful for the comments from anonymous reviewers that
helped improve the paper.
Appendix A. Final principles vs initial categories
Keyword Original observation Instances

(n ¼ 70)
Repurpose integrate and repurpose commercial consumer products (i.e. hacked) to reduce effort 6
identify similar designs as a starting blueprint for design 4
use generic connectors and interfaces to save integration effort 3
explore the basic working theory of the product to avoid common mistakes 2
minimize damage to an original commercial product during repurposing 2
minimize system part count, by searching for pre-made subassemblies 2
consider secondary effects of integration, how to simplify interfaces etc. 1
kits can also be used as a compromise between custom fabrication and repurposing 1
the repurposed function may be radically divergent from its original function 1
perform disassembly carefully to support re-use of a COTS product 1
replace multi-functional components with pre-made COTS 1
start by learning theory, benchmark state-of-the-art, deconstruct the working principle 1
adapt to existing input/output protocols 1
Satisfice adaptive quality, use the cheapest and lightest available parts, increase quality at failure 11
points
adapt the approach, match the approach to skill set of the designer 3
suit the materials to the task, over engineering is costly 2
let functional parameters guide part selection 2
run a brief failure modes and effects analysis, use it to design basic failsafe features 2
test safely, step by step, rather than over-engineering 1
take time to price out various options, select the cheapest to start 1
modelling detail is matched to complexity to save time 1
cost to performance trade-off 1
assembly: refined accuracy reduces correction efforts later 1
design to available tools, rather than ideal tools 1
part selection: most readily available parts 1
make the thing openable (permit disassembly) to allow changes later 1
incorporate feedback on operational status into the design to reduce failure modes 1
make 50 design sketches first, to ensure the idea is simple 1
develop several trial prototypes, testing each at as low of cost as possible 1
Stencil test function at each stage of assembly, use sequence of assembly to guide interface 8
detailing
use manufacturing rigs made from readily available materials, to support accurate 4
assembly
position elements and fit test the assembly before permanent affixing 4
self-alignment in fabrication, physically overlay to existing artefacts or systems 3
pre-cutting and scoring, human-in-the-loop fabrication can achieve precision with 1
planning
use mock-ups (or CAD) to layout and validate bill of materials before ordering supplies 1
colour code or use sorting bins for ease of organization, and to manage assemblies 1
plan out highly detailed diagrams before assembly 1
Layer start with rough dimensioning of the outer shape, then filling, smoothing, finishing 5
make a structural base first, then fasten (as in tensegrity), then lighter parts, wiring last 5
use reinforcing components, particularly structural lattices for large parts 4
start by selecting casing, or employing shells, of sufficient volume, complete layout 1
detailing
Standardize use a repetitive manufacturing process to save effort and machine cost 2
as complexity increase custom parts are preferred 1
develop a process to standardize outcome and reduce effort 1
perform fabrication treatment on multiple components simultaneously 1
References
alstroemeria. Digital/analog clock - arduino þ paperCraft. Retrieved 21 June,
2017, from http://www.instructables.com/id/DigitalAnalog-Clock-Arduino-
PaperCraft/.
Anderson, C. (2012). Makers: The new industrial revolution. Random House.
Arnheim, R. (1994). The way of the crafts. Design Issues 29e35.
Atkinson, P. (2006). Do it yourself: Democracy and design. Journal of Design His-
tory, 19(1), 1e10.
Camburn, B., Dunlap, B., Gurjar, T., Hamon, C., Green, M., Jensen, D., et al.
(2015). A systematic method for design prototyping. Journal of Mechanical
Design, 137(8), 081102.
Camburn, B. A., Sng, K. H. E., Perez, K. B., Otto, K., Jensen, D., Crawford, R.,
et al. (2015). The way makers prototype: Principles of DIY design. In Interna-
tional design engineering technical conferences and computers and information in
engineering conference. Boston, MA.
Cooper, C. B., Dickinson, J., Phillips, T., & Bonney, R. (2007). Citizen science
as a tool for conservation in residential ecosystems. Ecology and Society,
12(2), 11.
Cox, G. (1999). The digital crowd: Some questions on globalization and agency.
Design Issues 16e25.
Devendorf, L., & Ryokai, K. (2014). Being the machine: Exploring new modes of
making. In Proceedings of the 2014 companion publication on Designing interac-
tive systems, ACM.
Dow, S. P., Glassco, A., Kass, J., Schwarz, M., Schwartz, D. L., &
Klemmer, S. R. (2010). Parallel prototyping leads to better design results,
more divergence, and increased self-efficacy. ACM Transactions on
Computer-Human Interaction (TOCHI), 17(4), 18.
dragonator. A fully 3D printable GlaDOS robotic ceiling arm lamp. Retrieved 21
June, 2017, from http://www.instructables.com/id/A-fully-3D-printable-Gla-
DOS-Robotic-ceiling-arm-la/.
drj113. (2017). A word clock. from. http://www.instructables.com/id/A-Word-Clock/.
Edwards, C. (2006). ‘Home is where the art is’: Women, handicrafts and home im-
provements 1750e1900. Journal of Design History, 19(1), 11e21.
Eiben, C. B., Siegel, J. B., Bale, J. B., Cooper, S., Khatib, F., Shen, B. W., et al.
(2012). Increased Diels-Alderase activity through backbone remodeling guided
by Foldit players. Nature Biotechnology, 30(2), 190e192.
Elsen, C., H€aggman, A., Honda, T., & Yang, M. C. (2012). Representation in
early stage design: An analysis of the influence of sketching and prototyping
in design projects. In ASME 2012 international design engineering technical
conferences and computers and information in engineering conference. Chicago,
IL: American Society of Mechanical Engineers.
Faas, D., Bao, Q., & Yang, M. C. (2014). Preliminary sketching and prototyping:
Comparisons in exploratory design-and-build activities. In ASME 2014 inter-
national design engineering technical conferences & computers and information in
engineering conference. Buffalo, NY.
Francis, J. J., Johnston, M., Robertson, C., Glidewell, L., Entwistle, V.,
Eccles, M. P., et al. (2010). What is an adequate sample size? Operationalising
data saturation for theory-based interview studies. Psychology and Health,
25(10), 1229e1245.
Fu, K. K., Yang, M. C., & Wood, K. L. (2016). Design principles: Literature re-
view, analysis, and future directions. Journal of Mechanical Design, 138(10),
101103.
Fuller, R. B. (1962). Tensile-integrity structures. Google Patents.
Green, M., Seepersad, C. C., & H€ oltt€
a-Otto, K. (2014). Crowd-sourcing the eval-
uation of creativity in conceptual design: A pilot study. In ASME 2014 inter-
national design engineering technical conferences and computers and information
in engineering conference. American Society of Mechanical Engineers.
Grey, F. (2011). Citizen cyberscience: The new age of the amateur. CERN Courier.
H€aggman, A., Honda, T., & Yang, M. C. (2013). The influence of timing in
exploratory prototyping and other activities in design projects. In ASME
2013 international design engineering technical conferences and computers and
information in engineering conference. American Society of Mechanical
Engineers.
Hippel, E.v (2005). Democratizing innovation. Cambridge, Mass: MIT Press.
hyroc346. Duct Tape and PVC Kayak. Retrieved 21 June 2017, from http://
www.instructables.com/id/Duct-Tape-and-PVC-Kayak/.
Koen, B. V. (1988). Toward a definition of the engineering method. European
Journal of Engineering Education, 13(3), 307e315.
Kuznetsov, S., & Paulos, E. (2010). Rise of the expert amateur: DIY projects,
communities, and cultures. In Proceedings of the 6th Nordic conference on
human-computer Interaction: Extending Boundaries, ACM.
Liddament, T. (1999). The computationalist paradigm in design research. Design
Studies, 20(1), 41e56.
Life, R. S. (1994). Qualitative data analysis.
Lindtner, S., Hertz, G. D., & Dourish, P. (2014). Emerging sites of HCI innova-
tion: Hackerspaces, hardware startups & incubators. In Proceedings of the
32nd annual ACM conference on Human factors in computing systems, ACM.
Mellis, D. A., & Buechley, L. (2014). Do-it-yourself cellphones: An investigation
into the possibilities and limits of high-tech diy. In Proceedings of the 32nd
annual ACM conference on Human factors in computing systems, ACM.
Menold, J., Jablokow, K., & Simpson, T. (2017). Prototype for X (PFX): A ho-
listic framework for structuring prototyping methods to support engineering
design. Design Studies, 50, 70e112.
Mota, C. (2011). The rise of personal fabrication. In Proceedings of the 8th ACM
conference on Creativity and cognition, ACM.
Moussette, C., & Banks, R. (2011). Designing through Making: Exploring the
simple haptic design space. In Fifth international conference on tangible,
embedded, and embodied interaction.
Neeley, W. L., Jr., Lim, K., Zhu, A., & Yang, M. C. (2013). Building fast to think
faster: Exploiting rapid prototyping to accelerate ideation during early stage
design. Portland, OR: ASME-IDETC.
Norman, D. A. (1983). Design rules based on analyses of human error. Commu-
nications of the ACM, 26(4), 254e258.
Norman, D. A., & Verganti, R. (2014). Incremental and radical innovation:
Design research vs. technology and meaning change. Design Issues, 30(1),
78e96.
oomlout. How to Make a Three Axis CNC Machine (Cheaply and Easily).
Retrieved 21 June 2017, from http://www.instructables.com/id/How-to-
Make-a-Three-Axis-CNC-Machine-Cheaply-and-/.
Paulos, E. (2013). The rise of the expert amateur: DIY culture and the evolution of
computer science. ACM SIGARCH Computer Architecture News, ACM.
Perez, K. B., Anderson, D., & Wood, K. L. (2015). Crowdsourced design princi-
ples for leveraging the capabilities of additive manufacturing. In Proceedings of
the 20th international conference international conference on engineering design
(ICED15). Milan.
Q-Branch. The EyeWriter. Retrieved 21st June, 2017, from http://www.instructa-
bles.com/id/The-EyeWriter/.
Reich, Y., Konda, S. L., Monarch, I. A., Levy, S. N., & Subrahmanian, E. (1996).
Varieties and issues of participation and design. Design Studies, 17(2), 165e180.
RobHopeless. Build a Laser 3D Printer. Retrieved 21st June, 2017, from
http://www.instructables.com/id/Build-a-Laser-3D-Printer-Stereolithogra-
phy-at-Ho/.
Rosner, D., & Bean, J. (2009). Learning from IKEA hacking: i’m not one to
decoupage a tabletop and call it a day. In Proceedings of the SIGCHI Confer-
ence on Human Factors in Computing Systems, ACM.
Saakes, D. (2009). Big lampan lamps: Designing for DIY. In Proceedings of the
seventh ACM conference on Creativity and cognition, ACM.
Salmond, W. (1997). A matter of give and take: Peasant crafts and their revival in
late imperial Russia. Design Issues 5e14.
SarahandDillon. Use the "force" to drive a remote controlled device with a DIY 3D
capacitor!. Retrieved 21st June, 2017, from http://www.instructables.com/id/
Use-the-Force-to-drive-a-Remote-Controlled-Devic/.
Sass, L., Chen, L., & Sung, W. K. (2016). Embodied prototyping: Exploration of
a design-fabrication framework for large-scale model manufacturing. Com-
puter-Aided Design and Applications, 13(1), 124e137.
Sass, L., & Oxman, R. (2006). Materializing design: The implications of rapid
prototyping in digital design. Design Studies, 27(3), 325e355.
Saul, G., Xu, C., & Gross, M. D. (2010). Interactive paper devices: End-user
design & fabrication. In Proceedings of the fourth international conference on
Tangible, embedded, and embodied interaction, ACM.
Singh, V., Skiles, S. M., Krager, J. E., Wood, K. L., Jensen, D., & Sierakowski, R.
(2009). Innovations in design through transformation: A fundamental study of
transformation principles. Journal of Mechanical Design, 131(8), 081010.
Stone, R. B., Wood, K. L., & Crawford, R. H. (1998). A heuristic method to iden-
tify modules from a functional description of a product. In Proceedings of
DETC98, Atlanta, GA.
Telenko, C., Wood, K., Otto, K., Elara, M. R., Foong, S., Pey, K. L., et al.
(2016). Designettes: An approach to multidisciplinary engineering design edu-
cation. Journal of Mechanical Design, 138(2), 022001.
tigers58. Indoors Fractal HDTV Antenna. Retrieved 21 June 2017, from http://
www.instructables.com/id/Indoors-Fractal-HDTV-Antenna/.
Tilstra, A. H., Backlund, P. B., Seepersad, C. C., & Wood, K. L. (2015). Principles
for designing products with flexibility for future evolution. International Jour-
nal of Mass Customisation, 5(1), 22e54.
Torrey, C., Churchill, E. F., & McDonald, D. W. (2009). Learning how: The
search for craft knowledge on the internet. In Proceedings of the SIGCHI con-
ference on human factors in computing systems, ACM.
Triggs, T. (2006). Scissors and glue: Punk fanzines and the creation of a DIY
aesthetic. Journal of Design History, 19(1), 69e83.
Viswanathan, V., & Linsey, J. (2011). Design fixation in physical modeling: An
investigation on the role of sunk cost. Washington, DC: ASME-IDETC.
Viswanathan, V. K., & Linsey, J. S. (2010). Physical models in idea generation:
Hindrance or help?. In ASME 2010 international design engineering technical
conferences and computers and information in engineering conference Montreal,
Canada: American Society of Mechanical Engineers.
Viswanathan, V. K., & Linsey, J. S. (2012). Physical models and design thinking:
A study of functionality, novelty and variety of ideas. Journal of Mechanical
Design, 134(9), 091004.
whoand. Smartphone Wireless Controllable DIY AFM (Atomic Force Micro-
scope). Retrieved 21st, June, 2017, from http://www.instructables.com/id/A-
Low-Cost-Atomic-Force-Microscope-%E4%BD%8E%E6%88%90%E6%
9C%AC%E5%8E%9F%E5%AD%90%E5%8A%9B%E9%A1%AF%
E5%BE%AE%E9%8F%A1/.
Wood, K. L., & Linsey, J. S. (2007). Understanding the art of design: Tools for the
Next Edisonian Innovators.
Wu, H., Corney, J., & Grant, M. (2015). "An evaluation methodology for crowd-
sourced design.". Advanced Engineering Informatics, 29(4), 775e786.
Yang, M. C. (2005). A study of prototypes, design activity, and design outcome.
Design Studies, 26(6), 649e669.
Youmans, R. J. (2011). The effects of physical prototyping and group work on the
reduction of design fixation.". Design Studies, 32(2), 115e138.

Principles of Maker and DIY Fabrication: Enabling Design Prototypes at Low Cost

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

Principles of Maker and DIY Fabrication: Enabling Design Prototypes at Low Cost

Uploaded by

Copyright:

Available Formats

Principles of maker and DIY fabrication:

Enabling design prototypes at low cost

Keywords: prototypes, design practice, distributed design, case studies, social

1 The Do-It-Yourself design movement

2 Design Studies Vol -- No. -- Month 2018

2 Design principles identiﬁcation and characterization

Prototype fabrication principles are extracted from projects on the Instructa-

Principles of maker and DIY fabrication 3

The Technology article category of Instructables.com was selected for analysis

The initially identiﬁed set of principles appeared to consist of common themes,

4 Design Studies Vol -- No. -- Month 2018

Unique Practices Identified

Figure 1 Data saturation for 0

practices Number of Entries Evaluated

Figure 2 Sample of synthesis

2.2 Design principles characterization

Principles of maker and DIY fabrication 5

Principle Count of critical entries Total percent of entries with application

Repurpose commercial products 430 91%

the database. Results were evaluated through inter-rater agreement. Pearson’s

A second validation consisted of deploying surveys to an online crowdsourced

6 Design Studies Vol -- No. -- Month 2018

A general observation of this procedure is that questions must be clear and

Table 2 Sample of quality assessment and question redesign process

Principles of maker and DIY fabrication 7

The Fleiss’ Kappa is used to develop a ‘Synthesized Rater’. In eﬀect, only

Agreement results are shown in Table 3. Table 3 provides a summary of agree-

8 Design Studies Vol -- No. -- Month 2018

Repurpose commercial products 29 1

3 Maker fabrication principles case studies

3.1.1 Repurpose case study 1: Atomic force microscope

Principles of maker and DIY fabrication 9

3.1.2 Repurpose case study 2: Eye tracking interface

10 Design Studies Vol -- No. -- Month 2018

3.2 Principle 2: Satisﬁce

3.2.1 Satisﬁce case study 1: Capacitive 3D mouse

3.2.2 Satisﬁce case study 2: Laser stereolithography

3.3 Principle 3: Stencil

Principles of maker and DIY fabrication 11

3.3.1 Stencil case study 1: Word clock

3.3.2 Stencil case study 2: Fractal antenna

12 Design Studies Vol -- No. -- Month 2018

low-cost direct design-to-fabrication approach that enables production of a

3.4 Principle 4: Standardize

3.4.1 Standardize case study 1: 3 axis CNC router

3.4.2 Standardize case study 2: GLaDOS robotic lamp

Principles of maker and DIY fabrication 13

3.5 Principle 5: Layer

14 Design Studies Vol -- No. -- Month 2018

3.5.2 Layer case study 2: Duct-tape canoe

Principles of maker and DIY fabrication 15

4 Validation of DIY design principles

The experimental group implemented signiﬁcantly more principles in their

16 Design Studies Vol -- No. -- Month 2018

Prototypes in the experimental group were also rated by a panel of external

Table 4 Implementation of principles across groups

Group Average number of principles Standard error

Experimental teams 4.0 0.13

Principles of maker and DIY fabrication 17

Group Average number of principles Standard error

Teams in ﬁnal competition (n ¼ 13) 4.4 0.14

employed more principles in their prototyping, on average, than other teams.

4.1 Evaluation of Surveys