The Journal of Peer Production - New perspectives on the implications of peer production for social change New perspectives on the implications of peer production for social change
Cultures of sharing in 3D printing: what can we learn from the licence choices of Thingiverse users? image

Reviewing process: [original] [reviews] [signals]

This article contributes to the discussion by analysing how users of the leading online 3D printing design repository Thingiverse manage their intellectual property (IP). 3D printing represents a fruitful case study for exploring the relationship between IP norms and practitioner culture. Although additive manufacturing technology has existed for decades, 3D printing is on the cusp of a breakout into the technological mainstream – hardware prices are falling; designs are circulating widely; consumer-friendly platforms are multiplying; and technological literacy is rising. Analysing metadata from more than 68,000 Thingiverse design files collected from the site, we examine the licensing choices made by users and explore the way this shapes the sharing practices of the site’s users. We also consider how these choices and practices connect with wider attitudes towards sharing and intellectual property in 3D printing communities. A particular focus of the article is how Thingiverse structures its regulatory framework to avoid IP liability, and the extent to which this may have a bearing on users’ conduct. The paper has three sections. First, we will offer a description of Thingiverse and how it operates in the 3D printing ecosystem, noting the legal issues that have arisen regarding Thingiverse’s Terms of Use and its allocation of intellectual property rights. Different types of Thingiverse licences will be detailed and explained. Second, the empirical metadata we have collected from Thingiverse will be presented, including the methods used to obtain this information. Third, we will present findings from this data on licence choice and the public availability of user designs. Fourth, we will look at the implications of these findings and our conclusions regarding the particular kind of sharing ethic that is present in Thingiverse; we also consider the “closed” aspects of this community and what this means for current debates about “open” innovation.

by Jarkko Moilanen (University of Tampere), Angela Daly (Swinburne University of Technology/European University Institute), Ramon Lobato (Swinburne University of Technology) & Darcy Allen (RMIT University)

1. Introduction

A growing literature in economics and social science has explored the practices of information exchange among online communities. A strong theme within this literature is that open cultures—characterised by reciprocal sharing, weak IP, and open flows of information among practitioners—are conducive to technological innovation. In Benkler’s influential analysis, the end result is ‘a flourishing nonmarket sector of information, knowledge, and cultural production… subject to an increasingly robust ethic of open sharing, open for all others to build on, extend, and make their own’ (Benkler 2006, p.7). This phenomenon has been the focus of much recent research on collaborative production models, with numerous studies appearing about wikis, open-access publishing, free software and open science (e.g. Chesbrough 2006, Nielson 2011, Suber 2012, Anderson 2012, Hatch 2013).

One lesson from this literature is that sharing practices are context-dependent. Sharing is a social practice shaped by a range of variables, and sharing practices differ from community to community and from technology to technology (Kennedy 2013). Infrastructural issues, cultural factors and legal frameworks, both explicit and implicit, play a role in shaping the way we share information. It is therefore necessary to consider the diverse norms, values, structures and systems that emerge around particular forms of sharing. Scholars in various disciplines have taken up this challenge by documenting specific (rather than universal) aspects of sharing practice, such as the regulatory frameworks that govern conduct and the variable properties of technological platforms (e.g. Berdou 2011, Currie, Kelty & Murillo 2013, Schweik & English 2012, Suzor 2012).

This article contributes to the discussion by analysing how users of the leading online 3D printing design repository Thingiverse manage their intellectual property, and in doing so exchange information. 3D printing represents a fruitful case study for exploring the relationship between IP norms and practitioner culture. Although additive manufacturing technology has existed for decades, 3D printing is on the cusp of a breakout into the technological mainstream—hardware prices are falling; designs are circulating widely; consumer-friendly platforms are multiplying; and users’ technological literacy is rising. Thingiverse, as the leading website for 3D printing design-sharing, is playing a central part in this process.

As a contribution to the emerging literature on Thingiverse and its role in the 3D printing innovation system (Rayna et al 2014, Kyriakou et al 2012, Rideout 2011), this article analyses some recent IP-related controversies within 3D printing communities and examines Thingiverse’s contested position within this community. In particular, we look at how intellectual property, in the form of 3D printing design files, is ‘shared’ in two ‘directions’: ‘horizontally’ among Thingiverse users; and ‘vertically’ between those users and MakerBot, Thingiverse’s parent company. By ‘sharing’ we mean non-monetary exchange, either one-way or reciprocal, of design files between Thingiverse users. Such activities are usually restricted by traditional or orthodox uses of copyright law but the ‘sharing’ of material and software in these ways is a characteristic feature of maker and hacker communities (Coleman 2012).

We conduct this examination of sharing in Thingiverse using two methods: a legal analysis of the intellectual property provisions in Thingiverse’s terms of use and the disputes which have arisen around them; and an empirical research component comprising an analysis of metadata collected from design files hosted by Thingiverse which reveal the licensing choices made by Thingiverse users. We will discuss what the insights obtained from these two methods may suggest about attitudes towards intellectual property and sharing within this rapidly growing online community of practice.

The paper has three sections. First, we offer a description of Thingiverse and its place in the 3D printing ecosystem. Second, we examine the prevalence of vertical intellectual property sharing between MakerBot and Thingiverse users, noting in particular the issues that have arisen regarding Thingiverse’s Terms of Use and allocation of intellectual property rights. The third section considers horizontal sharing among Thingiverse users. Here we discuss the different types of secondary licences available to Thingiverse users when uploading their 3D printing designs, and present our empirical analysis of Thingiverse metadata. Throughout our discussion we explore the specific kinds of sharing ethics that are present in Thingiverse, the ‘closed’ as well as ‘open’ aspects of its user community, and what all this means for current debates about open innovation: themes that we draw together in the paper’s Conclusion.

2. Thingiverse’s place in the 3D printing ecosystem

Additive manufacturing technology has long been a feature of aerospace, medical, manufacturing and defence industries. Within the last few years the technology has crossed over into the consumer space, with household-oriented printers coming onto the market at ever-lower prices. The continuing boom in public interest has led to significant commercial investment, venture capital speculation, and consolidation of what was previously a fragmented sector. It has also created a tsunami of hype, with business magazine The Economist heralding the arrival of a ‘third industrial revolution’ (21 April 2012).

In this paper, our focus is on one small, yet crucial, part of the 3D printing ecosystem: the online design repositories that allow 3D printing enthusiasts, both professionals and non-professionals, to exchange design files. These repositories play a crucial role in linking experts with DIY enthusiasts who may not have the necessary skills to create complex Computer Aided Design (CAD) files. Rayna et al (2014) provide a brief typology of online 3D printing platforms after identifying 14 examples. The authors suggest the first mover in the space was Ponoko in 2007, and that the market grew following this. Table 1 provides examples of the diversity of 3D printing sites. These sites perform a mix of functions including design supply, hosting, customisation, co-creation and crowdsourcing, as well as offering print-on-demand and other bespoke manufacturing services.

Table 1: 3D printing design sites

Thingiverse ( The most popular 3D printing repository. Owned by MakerBot. Designs are free to upload and download. Has received the most media attention for Digital Millennium Copyright Act (DMCA) take-down notices.
Shapeways ( A leading commercial site that combines repository and print-on-demand functions. Users can create online shopfronts offering made-to-order products, printed by Shapeways that can be delivered to customers in finished form.
Ponoko ( The first website to launch a service in this area, in 2007. Ponoko continues to operate with both free Creative Commons designs, and a paid service for ‘creators and consumers’
Cuboyo ( A commercial site offering paid downloads of user-generated 3D objects. Cuboyo takes a 30% cut of the sale price, with the remaining 70% going to the seller.
MakeXYZ ( MakeXYZ acts as an intermediary between fabricators and end users. The site enables you to ‘find makers in your neighbourhood that are ready to help you make something’. ( A mix of free and paid design downloads. The site is connected with iMakr, which opened a physical store in London in May 2013. Offers 3D print on demand, and 3D design requests.
Repables ( Free, open-access repository founded by Gerrit Coetzee. Aims to be a non-commercial alternative to proprietary repositories like Thingiverse.
Fabster ( A showcase for 3D printing designs, based on a popular Facebook page. Allows companies and artists to display their 3D printed portfolios ‘within a mutual territory for all’. Does not offer downloads.
Yeggi ( Launched in April 2013 as a ‘meta-library’ or aggregator where makers can search for designs across a number of the other depositories listed above—‘We collect data from all 3D communities and marketplaces offering 3D models to print’

Thingiverse, owned by MakerBot, is the largest and most important of the 3D printing repositories. MakerBot’s own history can be traced back to the RepRap project. Founded in 2007 by Dr Adrian Bowyer, a Senior Lecturer in mechanical engineering at the University of Bath, RepRap was an initiative to develop a 3D printer that could re-print most of its own components. The RepRap project released all of the designs it produced under the GNU General Public licence, in line with free software principles. Designers were free to modify RepRap designs so long as they shared their creations back with the RepRap community.

However, three of the organisers of the NYC Resistor Hackerspace in Brooklyn—Bre Pettis, Zach Smith and Adam Mayer—had other ideas. They ‘threw out the self-replication requirement’ of RepRap (Courtland 2013), and focused their energies on developing a consumer-friendly printer. Their company, MakerBot Industries, was founded in January 2009, and had sold several thousand printers by 2011. In 2012 it attracted US$10 million in venture capital funding and the following year the company was bought, with much fanfare, by the 3D printing giant Stratsys (Stratasys, along with 3D Systems, is one of the ‘big two’ 3D printing corporations). Stratasys paid US$400 million in stock, with the consequence that MakerBot’s founders became millionaires overnight.

Thingiverse plays the role of the design hub within MakerBot’s 3D printing ecosystem. Users can post and collaborate on design files for 3D printable ‘Things’, and find new and interesting uses for their MakerBot printers. Thingiverse has become the leading repository of user-submitted design files and the world’s largest online 3D printing community. Unlike some of its competitor sites, Thingiverse is a free site inasmuch as users do not pay a fee to access it nor does it host external advertising. Its commercial function for MakerBot is to add value to printer sales by offering a free and easy way for users to find designs they can print off at home. In the same way that iTunes adds value to iPhone purchases, but generates little profit in its own right, the site is ancillary to MakerBot’s main line of business—sales of 3D printing hardware.

There is a distinct commercial logic underlying Thingiverse’s culture of sharing. Users of the site are part of a thriving, ‘open’ community of practice, but their activities are organised in ways that align with MakerBot’s commercial ambitions. Thingiverse files are ‘encouraged to be licensed under a Creative Commons license’ (MakerBot 2014(b)). To quote Bre Pettis: ‘[i]f you’re not sharing your designs, you’re doing it wrong’ (The Economist 2011). The community feel is further espoused on the website: ‘We’re hoping that together we can create a community of people who create and share designs freely, so that all can benefit from them’ (MakerBot 2014(a)). This communitarian aspect to Thingiverse is a major selling point for MakerBot; it markets its hardware on the basis of this vibrant online community.

As our brief account of Thingiverse has demonstrated, the site mobilises a large and growing user base and attempts to align ideas of community, creativity and commerce—framed within the concept of ‘sharing’—in ways that are commercially useful to MakerBot. In the following two sections, we will interrogate precisely what this sharing of intellectual property in Thingiverse looks like, in two directions: vertically between MakerBot/Thingiverse and users; and horizontally among users.

3. Sharing intellectual property vertically

In this section, we examine the sharing of intellectual property taking place between MakerBot/Thingiverse and Thingiverse users via an analysis of the MakerBot’s policy decisions regarding intellectual property transfer between the company and users, and the disputes arising around these decisions. These situations of conflict over intellectual property, which are not unusual in the context of for-profit Web 2.0 platforms, suggest that MakerBot’s ‘sharing’ rhetoric is confined to users sharing their 3D printing designs ‘upwards’ with the company, while MakerBot is much less willing to share its intellectual property ‘downwards’ with users.

The Replicator 2 controversy

The first sharing controversy involving MakerBot relates to the Replicator 2 3D printer, the infamous fork of the original open-source RepRap project, which MakerBot released in September 2012. Although it incorporated a number of new features compared to its predecessor, the printer was not received positively by the 3D printing community. Unlike previous MakerBot printers the Replicator 2 was closed and did not follow the principles of open hardware—i.e. that the ‘design is made publicly available so that anyone can study, modify, distribute, make, and sell the design or hardware based on that design’ (OSHWA n.d.). The open hardware movement can be conceptualised as an extension of the free/libre/open source software ideology to physical technological artefacts, designed and disseminated in an ‘open’ fashion. The difference with the Replicator 2 was that MakerBot did not publish details of how the printer was designed. To accompany the new printer, MakerBot also released new 3D printing software that did not comply with open source principles, attracting further criticism.

On the MakerBot blog, Bre Pettis defended the company’s actions:

For the Replicator 2, we will not share the way the physical machine is designed or our GUI because we don’t think carbon-copy cloning is acceptable and carbon-copy clones undermine our ability to pay people to do development. (Pettis 2012)

Among the many reasons cited, Pettis focused on the idea that ‘running a business is complicated’ and required MakerBot to keep control of the core technology. It seemed that the Replicator 2 was aimed at a different demographic than those tinkerers involved with the open hardware movement—namely, people ‘who want to make gorgeous models instead of hack the machine’. Unsurprisingly, MakerBot’s actions caused a great deal of controversy within the 3D printing community. The decision to become closed source was criticised by Josef Prusa (2012(b)), a former MakerBot employee, as well as Zach Smith, a founding member of MakerBot who had subsequently left the company. Smith (2012) refers to this departure from open to closed source as the ‘ultimate betrayal’, underscoring the bitter disputes about intellectual property and sharing norms that have become part and parcel of the 3D printing boom.

Occupy Thingiverse

2012 also saw the second controversy involving MakerBot and Thingiverse, the Occupy Thingiverse incident. This controversy has its origins in Thingiverse’s 2012 decision to alter its Terms of Use (to which all users must agree if they wish to open a Thingiverse account) to include the following:

You hereby grant, and you represent and warrant that you have the right to grant, to Company and its affiliates and partners, an irrevocable, nonexclusive, royalty-free and fully paid, worldwide license to reproduce, distribute, publicly display and perform, prepare derivative works of, incorporate into other works, and otherwise use your User Content, and to grant sublicenses of the foregoing, solely for the purposes of including your User Content in the Site and Services. You agree to irrevocably waive (and cause to be waived) any claims and assertions of moral rights or attribution with respect to your User Content. (MakerBot (c))

This marked a significant departure from the previous Terms of Use, which had a more narrow scope, were revocable in nature, and concerned Thingiverse’s rights to use the content uploaded by users across its platforms:

However, by posting, uploading, inputting, providing or submitting your content to, you are granting, its affiliated companies and partners, a worldwide, revocable, royalty-free, non-exclusive, sub-licensable license to use, reproduce, create derivative works of, distribute, publicly perform, publicly display, transfer, transmit, distribute and publish that content for the purposes of displaying that content on and on other Web sites, devices and/or platforms (Walter 2012).

One major difference in the wording is that the previous Terms had a more narrow scope than their replacement, and seemed to restrict the use of user designs to similar platforms to Thingiverse. The consequence of this appears to have been that MakerBot was limited in its ability to incorporate designs uploaded to Thingiverse into ‘closed’ hardware products, such as their 3D printers. The change in language, however, seemed to allow Thingiverse to use those designs for its own commercial purposes, as well as to assert moral rights over any design uploaded to the platform. One of Thingiverse’s lawyers defended the moral rights waiver, stating that these rights were not part of American copyright law and that the purpose of the waiver was to ‘lend certainty’ to the licences upon which MakerBot relies to operate Thingiverse—to prevent, for instance, users claiming that their moral rights against the ‘mutilation’ of a work had been infringed (McCarthy 2012). Interestingly, Thingiverse’s actions were also framed within a discourse of sharing: they suggested the assertion of moral rights by the original users would be ‘fundamentally inconsistent with the intention of Thingiverse, which is to share things and their derivatives’. Reference was also made to the fact that these updated Terms of Use were ‘structured similarly to any large website that hosts user-content’.

Yet Thingiverse’s assertion that moral rights were not part of US copyright law was inaccurate—some operation of moral rights had been recognised in case-law, and then statutorily in the Visual Artists Rights Act 1990 (Rosenblatt 1998), which was enacted to implement some moral rights provisions of the Berne Convention that was eventually signed by the US in 1989. However the bigger issue was the widespread disquiet within Thingiverse’s maker community, many of whom took to the Internet to argue that Thingiverse was no longer ‘open’, and that the new Terms betrayed the community ethos. This user backlash came to be known as Occupy Thingiverse. The movement was initiated by the aforementioned former MakerBot employee Josef Prusa (2012(a)), who wrote an open letter to Bre Pettis which drew attention from other Thingiverse users (Prusa 2012(b)). The Occupy Thingiverse meme quickly gained traction, and many users chose to remove their designs from the Thingiverse site so that they might retain more control over them. Following Prusa’s suggestion, many of these users reposted their designs at GitHub, a popular data repository for open-source projects (Molitch-Hou 2013). Forums were created to discuss alternatives (RepRap 2013).

It is worth noting that Thingiverse’s Terms of Use take precedence over any of the more ‘open’ licence choices that users make on the site. While Thingiverse still gives users a choice of licence (a point we will discuss in more detail shortly), this is a secondary licence on top of the primary licence over uploaded content that this language in the Terms of Use grants to Thingiverse. The licence with Thingiverse is a standard form, non-negotiable contract—users must sign up to Thingiverse and accept these terms, or not have a Thingiverse account at all. The offending language which led to the Occupy Thingiverse movement is still present in Thingiverse’s Terms of Use at the time of writing.

Ongoing patent problems

The Replicator 2 and Occupy Thingiverse controversies demonstrate MakerBot’s willingness for users to share their creations ‘upwards’ as well as horizontally between users, as the MakerBot business model requires. This can be seen plainly in another recent controversy regarding MakerBot patents.

In mid-2014 it emerged that MakerBot had filed various 3D printing-related patent applications in the US (Benchoff 2014). The two patents that received the most attention related to technical improvements in 3D printer hardware—specifically, a quick-release extruder mechanism and a plate-levelling device. While the fact that MakerBot is patenting new elements of its hardware design angers many within the 3D printing community, who remember the open-source roots of the RepRap project, in the context of MakerBot’s increasingly aggressive IP policy such patent applications are not particularly surprising. However, there was an added degree of intrigue in this instance because similar innovations had previously been published under open licences by enthusiasts. Ironically, one of these—the quick-release extruder—appeared on Thingiverse months earlier. This prompted accusations that Makerbot was stealing the ideas of its users. MakerBot was quick to defend itself, however, arguing in a blog post that its patent applications predate the user-designed versions. It offered no apology for its patenting practices, acknowledging that filing patents was not ‘optimal’ but was an intrinsic part of ‘being a competitive company’.

The general reaction from community members to these revelations again was predictably negative, with some members advocating a boycott of Thingiverse, and others resisting the call to boycott the site (Molitch-Hou 2014). Those who were against a boycott noted that the process of posting design files on Thingiverse generated evidence that the user had created and uploaded that design on a certain date—which could be used to oppose patent applications such as those by MakerBot.

Applications for patents are not the same thing as the grant of a patent. The patent may not be granted on a number of grounds. Specifically, successful patent applications must fulfil various conditions, including the elements of novelty, non-obviousness (in the US — ‘inventive step’ in Europe) and usefulness (in the US — ‘susceptible of industrial application’ in Europe). However, chronology here is key. While these allegations of MakerBot ‘stealing’ users’ designs in order to patent them may well turn out to be a red herring, this course of action that MakerBot is taking by filing patent applications demonstrates a further distancing from the company’s ‘open’ roots and another point of conflict with its user community.

Together, these three examples — the RepRap controversy, Occupy Thingiverse, and the hardware patents — demonstrate the vertical sharing of intellectual property that goes on in Thingiverse between MakerBot and users is largely one-directional. While users are encouraged to upload and share their designs on Thingiverse using a Creative Commons or free software licence (discussed in more detail in the following section), MakerBot is no longer sharing its intellectual property in the form of 3D printing design files back with the community, and indeed has even been accused of misappropriating the creations of Thingiverse users.

4. Sharing intellectual property horizontally

So far in the article we have considered various aspects of intellectual property regulation on Thingiverse, including some recent controversies related to this issue. Some of these aspects relate to Thingiverse as a platform—how it is organised, how it mobilises certain kinds of practices, and its role within MakerBot’s sphere of commercial ambition. These aspects are all bound up with what Gillespie (2010) calls ‘the politics of platforms’.We now turn to the question of user practices on Thingiverse — specifically, what kinds of licensing options users choose when they upload their 3D printing designs to the site. In this section we explore what these choices tell us about how cultures of sharing are framed, understood and practiced among Thingiverse users, i.e., in the horizontal, peer-to-peer dimension of sharing, as opposed to the vertical sharing dimension discussed in the previous section.

For users to upload a design to Thingiverse, they must firstly have a Thingiverse account, and agree to the Terms of Use (as discussed above). The Terms of Use are a standard form agreement and there is no scope to negotiate with Thingiverse: users must either take them or leave them. When uploading a 3D printing design file to the platform, the user can add various pieces of information about the design, such as: its name; a brief description of the Thing; a category from a drop-down list, including such terms as art, fashion, gadgets, hobby and learning; and instructions as to how the creator made the Thing. The user can also link their creation to another, pre-existing Thing if it was ‘inspired by, derived from, or a remix of’ that other Thing. Finally, the user also has a choice of secondary licence to attach to the Thing from a drop-down list so that the user can ‘choose how you want your thing to be used by others’.

These secondary licences chosen by users are the focus of our discussion here. Creative Commons licences feature heavily in the drop-down list of options presented to users. Indeed, Thingiverse actively encourages users to list their designs under one of these licences, stating that licensing under this banner means ‘that anyone can use or alter any design’ (MakerBot 2014(b)). While such an arrangement seems conducive to open sharing among users, there are a number of issues here that warrant attention.

Creative Commons (CC) licensing provides a unique combination of conditions giving users a form of copyright that is more tailored to their personal needs. They can be seen as a level to which the user wishes to free their works into the public domain, reflecting the extent to which they reserve, or do not reserve, their rights. In this sense, CC licensing can be seen as an unorthodox use, or ‘hack’, of intellectual property law. Traditionally, copyright has worked by granting the creator of a work a bundle of exclusive rights over that work, governing how the work is shared, copied and modified — and usually this cannot be done without the permission of the copyright holder (which is not necessarily the original creator as these rights can be assigned to others). CC licensing differs from this model inasmuch as the copyright holder can choose to allow the future distribution of copies and modified versions of the original work without users needing to ask specific permission, while requiring that these same rights are preserved in any future modified versions.

As mentioned above, when uploading a file, Thingiverse contributors are asked to attach a secondary licence to their design file, which includes the core suite of six CC licences (Creative Commons n.d.(b)). This possibility to choose a licence acts as a flexible tool to respond to the needs of creators and the demands of consumers. The licences are grouped under four ‘modules’, each representing the extent to which certain rights are (or are not) withheld. The four modules are: Attribution (there are no restrictions on what others can do with the creation so long as the original creator is acknowledged); Non Commercial Use (the original creation can be used in any way so long as it is not commercial); Share Alike (the original creation can be used in any way so long as derivative creations are licensed under the same terms); and No Derivatives (the original creation can be redistributed so long as it is passed along unchanged and in whole). Combinations of these four modules result in six unique licences: Attribution (CC BY); Attribution-ShareAlike (CC BY-SA); Attribution-NoDerivs (CC BY-ND); Attribution-NonCommercial (CC BY-NC); Attribution-NonCommercial-ShareAlike (CC BY-NC-SA); and Attribution-NonCommercial-NoDerivs (CC BY-NC-ND). As previously noted, when uploading a file on Thingiverse, users select this secondary licence from a drop-down menu. The default option is CC BY and the second option down on the list is CC BY-SA. Interestingly, as can be seen on the table below, these are the two most popular licences chosen by users in our sample, and it may well be that many users simply go along with the pre-selected option or the second choice, rather than considering the other options.

The CC licences display varying degrees of ‘openness’. The most open CC licences (based on the extent to which intellectual property is licensed in a non-restrictive way) are those which only require attribution to the original creator when using or remixing the material—the CC BY (Attribution) licence. Accordingly, the most restrictive CC licences would be those that do not allow derivative works (no adaptions and no changes) and do not allow the works to be used for commercial purposes. Also important is the ‘Share Alike’ restriction permitting derivative works but only if they are further licensed under the same terms as the original work. For example, if the original work is not to be used for commercial purposes, then neither is the derivative. This kind of licence is known as ‘sticky’ as it ‘sticks’ to all future, modified versions of the original work. The two least ‘open’ licences with the lowest ‘sharing’ factor would be CC BY-NC-SA (Attribution-NonCommercial-ShareAlike) and CC BY-NC-ND (Attribution-NonCommercial-NoDerivs).

Thingiverse also gives its users the choice to use a CC-Public Domain Dedication licence, although CC has officially ‘retired’ this licence. Its successor is the CC0 ‘No Rights Reserved’ dedication, by which creators of copyrighted material can waive their interests in those works and ‘place them as completely as possible in the public domain, so that others may freely build upon, enhance and reuse the works for any purposes without restriction under copyright or database law’ (Creative Commons n.d(a)).

Moreover, Thingiverse users can choose from a number of other, non-CC, licences, which are derived from free software licences: while broadly similar to CC licences, these are designed to deal with different situations. CC licences typically apply to ‘traditional’ or ‘conventional’ types of copyrighted material such as music, film, photography and literature—and also apply to websites. Free software licences typically apply to software, which has been protected by copyright in the US since 1980 when computer programs were defined as falling within the ‘literary works’ category. The inclusion of software code in categories of copyright protection has been controversial (Samuelson 1988), as has the recognition by US courts since the 1970s that software could also be patentable (which is not the case for more conventional literary works). In response to this software ‘land grab’, free software licences have been created and used to make the works accessible and allow others to build upon them. Thingiverse offers a choice of three free software licences for uploaded files in addition to the CC options: two GNU licences—GNU General Public Licence (GNU-GPL) and GNU Lesser General Public Licence (GNU-LGPL)—and the BSD (Berkeley Source Distribution) licence.

In addition to the choice of secondary licence, Thingiverse users also have the ability to make their files ‘public’ or ‘private’. ‘Public Things’ are publicly available to see and download from Thingiverse’s website. ‘Private Things’ are Things which have not yet been officially published, because they are still in draft form or because the person uploading the file does not wish to make it publicly available for some other reason. There is also a box that can be ticked to signal that a Public Thing is a Work in Progress, to alert others to the fact that it may be updated in some way in the future but is still available publicly in its current form.

By analysing these user choices, both of secondary licence and whether a Thing is public or private, we can gain some insight into the motivations of Thingiverse users through revealed preference, and thereby assess the prevalence of horizontal sharing practices among Thingiverse users. To this end, in 2013 we conducted an empirical analysis of the metadata of 117,450 Thingiverse Things, primarily to discover information about users’ choices of secondary licence and the prevalence of Public and Private Things. Thingiverse has an open API which can be used to collect metadata about Things, but in practice it proved complicated to use for our purposes, mostly due to a lack of code examples and insufficient documentation; so we chose to screen-scrape with a custom-built Ruby program, which extracts information from the site by parsing its web pages.

Figure 1 shows how the number of designs on the site has grown since 2011, as determined by the upload date of Things. There is a noticeable increase around Jan-Feb 2013. One reason for that, aside from growing popular interest in 3D printing technology, might be the fact that in January 2013 Thingiverse launched a new online application, MakerBot Customizer, which allowed users to create easily new things from parametric designs.[1] Another possible reason is the counter-effect of the negative publicity around Thingiverse discussed above.

Figure 1: Number of Things in Thingiverse

moilanen figure 1

The data scrape, conducted between 16 and 18 August 2013, collected and stored metadata about 117,450 Things—both Public and Private—dating from Jan 2009 to Aug 2013. Private Things could be coded for their status only, while Public Things offered a wealth of other data, including the database identifier, author handle, secondary licence choice, creation date, how many times the Thing had been commented on, how many times it had been viewed, and any tags the creator had attached to it. This scrape also revealed number of ‘makes’ (how many times other Thingiverse users have reportedly printed the item), the number of collections (how many collections, i.e. user-curated categories like ‘Keys’ and ‘Vases’, in which the given Thing appears) and number of remixes (how many times the Thing has been used as starting point for derived works). There are of course limitations to such an approach, which provides a wide-angled overview of IP practices rather than the deep analysis that might result from other methods, such as community ethnographies or sample studies.[2] These limitations notwithstanding, the scrape uncovered a number of insights into Thingiverse’s operation and use.

Table 2: Top 5 secondary licence choices among Thingiverse users, 2013

Licence type Percentage of Things
Attribution (CC BY) 36%
Attribution-ShareAlike (CC BY-SA) 36%
Attribution-NonCommercial (CC BY-NC) 10%
Attribution-NonCommercial-ShareAlike (CC BY-NC-SA) 8%
GNU General Public Licence (GPL 2.0) 4%

Our first finding, based on an analysis of metadata from the 68,618 Public Things hosted on Thingiverse, was that the CC licences were by far the most popular licences used. The top 4 licences were all CC, representing 89.84 per cent of all Public Things (see Table 2). Interestingly, the two most popular choices—Attribution (CC BY) and the ‘sticky’ Attribution-ShareAlike (CC BY-SA)—both allow commercial usage. As mentioned above, the order of popularity here is fairly similar to the order in which licence options are listed in Thingiverse’s licence drop-down menu, so it is possible that many users do not venture far down the list or are happy to go with the initial options.

Further analysis of the remixed objects tells us more about how these licence choices interact with user practices. Licence choices for these objects are illustrated in Figure 2. Note that, while the number of CC BY (blue circle) and CC BY-SA (grey circle) licensed Things are almost the same, CC BY licensed Things are remixed far more often (10,569 times) than CC BY-SA (7,225) Things. Conversely, many more CC BY-SA (grey circle) licensed Things have been made than CC BY (blue circle) licensed Things. In other words, non-sticky licences seem to be preferred when it comes to remixing 3D printing designs. This would be in line with the ‘experimental’ quality of these designs, whereas sticky licences may be associated more with finished works ready for presentation to, and construction by, other users.

The most frequently remixed item we found was a customisable iPhone case, licensed under CC BY, which had been remixed 2,153 times. The licence used in the example is not sticky, but does require attribution. In theory this gives remixers more freedom in selecting other licences for their work, particularly compared to the commonly used CC BY-SA licence, which requires that derivative works be licensed under ‘same or similar’ terms.

Figure 2: Licence choices—by makes and remixes

moilanen figure 2

This remixing pattern can be compared with use of the ‘collection’ function. CC BY (blue) and CC BY-SA (grey) licensed Things are almost equal in terms of how many times they are included in user collections, suggesting that choice of licence here does not play a major role in which objects users find most interesting to add to their collections, unlike the situation with number of makes and number of remixes discussed above. In comparison, BY-SA licensed items (11,964 times) are significantly more frequently printed out than BY items (7,833 times). Licence choice does not seem to make a difference to the likelihood of a Thing being added to a user’s collection. It is hard to discern what the users’ motivations behind adding a Thing to one of their collections is without more information from those users, but this finding may suggest that it is users’ view of a particular Thing as being innovative, useful or otherwise interesting to them which is important here, rather than the secondary licence attached to that Thing. Moreover, it may well be that a collection constitutes different components of a complex creation—or preliminary attempts at such a creation—created by different users using different licences for each part.

In terms of actual reproduction using a 3D printer, however, ‘sticky’ licenced models are far more likely to be made than BY items. It is possible that there is a link between a Thing’s quality and the licence chosen by the creator, with high quality Things appearing under more restrictive licences than low quality Things due to the time and effort involved in creating the former.

Figure 3: Licence choices—by makes and collection count

moilanen figure 3

Analysis of Thingiverse tags can also reveal insights into intellectual property norms on the site. These are listed in Figure 4 below. The most common tag that users attach to Things is customized (19,206 times). This tag would seem to refer to Thingiverse users handling Things in a way closer to ‘remixing’ than ‘building from scratch’, although overall few Things use this tag in what is framed as being an iterative, collaborative platform. Nevertheless, strong conclusions cannot be drawn from this finding given tagging Things is optional rather than mandatory.

Figure 4: Most common tags used with Thingiverse objects

moilanen figure 4

The findings so far broadly conform to the image of Thingiverse as a platform dominated by derivative works and collaborative projects among users. All this supports the notion of Thingiverse as an open sharing service, in the horizontal peer-to-peer sense at least. Yet other findings from our study tell a different story.

Our scrape of the site revealed that nearly 42 per cent of the files hosted on Thingiverse are Private—in the sense that they have not been made available to internet users according to the process described above, whereby a user selects a ‘Public’ status for their creation and the file is published on Thingiverse. In other words, they are not shared through the Thingiverse platform—or if they are shared, the sharing is limited to a small group of collaborators who have access to the login and password details of the particular Thingiverse user account, or via other, non-Thingiverse channels.

Why are there are so many Private Things? One explanation may be that many people are experimenting with the platform and are not ready to release their work to the community. If a user does not fully complete the upload process their item will show up as a Private Thing. However, we should bear in mind another conflicting finding from our scrape: only 6 per cent of Things have an in-progress status.[3] We also note that over time the ratio of Private Things to all Things has been growing slightly, although this may be a random fluctuation. More longitudinal research is needed to determine whether this increase in Private Things is significant.

Figure 5: Types of Thingiverse Things

moilanen figure 5

We were unable to collect any more information about why Thingiverse users were marking their Things as private, despite our efforts to contact individual users. Further qualitative research would be useful here. From the information we have, it seems that sharing by Thingiverse users among themselves may not be as prevalent or dominant a practice as the rhetoric suggests, and should perhaps be considered within the wider context of MakerBot’s commercial strategy: as mentioned earlier, the company is keen for its users to share yet less willing to share back since it transitioned to ‘closed’ design and software.

This finding that a large proportion of total Thingiverse Things are actually ‘Private’ and not shareable by other users is not entirely isolated from other horizontal, peer practices. For instance, Thingiverse has been the object of various Digital Millennium Copyright Act (DMCA) takedown notices regarding files its users had uploaded to the site which allegedly infringed the intellectual property of others (Rideout 2011; Weinberg 2013, Thompson 2012, Brean 2013, Andersen and Howells 2014, Kahler 2013). While some of the entities issuing these notices were large companies, in one case the notice came from another 3D printing user who had uploaded his design to rival design repository Shapeways. The very existence of these DMCA takedown orders shows another limit to horizontal sharing on Thingiverse. Even if Thingiverse users are happy to share their designs via CC licences, this does not mean that others are also happy for this to happen.

The empirical analysis presented here demonstrates that while it is true to say a great deal of intellectual property sharing occurs among users in a horizontal fashion in Thingiverse, the picture is complicated by the fact that a large proportion of Things overall uploaded to the site remain ‘Private’, and thus cannot be shared with other users via the platform.

5. Conclusion

The design repository Thingiverse has had a rapid and spectacular rise, and is now a vital component in global 3D printing culture and practice. As our study of the site has demonstrated, Thingiverse sits at the nexus of a number of intellectual property tensions and disputes. As such, it is a useful site from which to view wider IP power struggles within peer production arenas.

We have demonstrated that Thingiverse’s IP policies are profoundly contradictory. On the one hand, Thingiverse and the practices it enables are the ultimate by-products of user innovation, fertilised by open-source culture. On the other hand, Thingiverse is a commercial, proprietary platform, owned by a large, global corporation, which has been widely criticised for its own intellectual property misdemeanours. Recent controversies around Terms of Use reveal the contentious nature of Thingiverse’s actions, and the ethical gulf that divides MakerBot and Thingiverse from their users.

Yet the behaviour of Thingiverse users is also somewhat contradictory. As our analysis has shown, only a proportion of users license their content in ways that take full advantage of open licensing norms. Creative Commons licences are used for 89 per cent of all Public Things. Yet users also keep a surprisingly large proportion of their designs private. Whether by accident or design, this adds a thick layer of ‘off-stage’ activity to what is intended, at least judging from MakerBot’s rhetoric, to be an open, transparent system dedicated to sharing.

Looking into the horizon, we anticipate a number of future developments for Thingiverse. Digital Millennium Copyright Act takedown notices, which suggest some kind of unauthorised sharing has taken place, are likely to become an increasingly common occurrence, and the site is likely to deepen its reliance on the kind of complex, automated regulatory systems that characterise other major user generated content platforms, notably YouTube. This is, in part, a natural consequence of the 3D printing boom. As awareness of the technology rises, hardware gets more sophisticated, and design options proliferate, the level of concern among rights holders is likely to increase. The rise of 3D scanners is another factor to watch. Take-up of scanning technology, and its integration into handheld devices, will likely reduce the reliance on user-created designs, thus relocating (although possibly not reducing) some of the existing IP tension, with ethical choices being devolved to software rather than human designers.


The authors thank Jenny Kennedy and Amanda Scardamaglia for their helpful and insightful comments on this piece.


Anderson, C 2012. Makers: The New Industrial Revolution, Crown Business: New York.

Andersen, J & Howells, J 2014. ‘The Intellectual Property Rights Implications of Consumer 3D Printing’, Aarhus University Department of Business Administration School of Business and Social Sciences thesis, Denmark.

Benchoff, B 2014. ‘MakerBot Files Patents, Internet Goes Crazy’, Hack A Day, 24 May. Available from: [Accessed: 19 October 2014]

Benkler, Y 2006. The Wealth of Networks: How Social Production Transforms Markets and Freedom, Yale University Press: New Haven.

Berdou, E 2011. Organization in Open Source Communities: At the Crossroads of the Gift and Market, Routledge: London.

Biggs, J 2014. ‘MakerBot Responds to Critics Who Claim It Is Stealing Community IP’, TechCrunch, 28 May. Available from: [Accessed: 19 October 2014]

Brean, DH 2013. ‘Asserting Patents to Combat Infringement Via 3D Printing: It’s No “Use”’, Fordham Intellectual Property, Media & Entertainment Law Journal, Vol. 23(3), pp. 771-814.

Chesbrough, HW 2006. Open Innovation: The New Imperative for Creating and Profiting from Technology, Harvard Business School Press: Cambridge.

Coleman, G 2012. Coding Freedom, Princeton University Press: Princeton.

Courtland, R 2013. ‘Resources Profile: Bre Pettis’, IEEE Spectrum, 25 June. Available from: [Accessed: 10 May 2014]

Creative Commons n.d.(a). ‘About CC0 – “No Rights Reserved”’. Available from:> [Accessed: 9 May 2014]

—— n.d.(b), ‘About the Licenses’. Available from: [Accessed: 9 May 2014]

Currie, M, Kelty C & Murillo LFR 2013. ‘Free Software trajectories: from organized publics to formal social enterprises?’ in Journal of Peer Production, July, Issue 3.

Gillespie, T 2010. ‘The Politics of “Platforms”’, New Media and Society Vol. 12(3), pp. 347-364.

Hatch, M 2013. The Maker Movement Manifesto: Rules for Innovation in the New World of Crafters, Hackers, and Tinkerers, McGraw-Hill: New York.

Kahler, A 2013. ‘I got a DMCA takedown notice from Makerbot/thingiverse for this’, Google+. Available from: [Accessed: 10 May 2014]

Kennedy, J 2013. ‘Rhetorics of Sharing: Data, Imagination and Desire’, in Geert Lovink and Miriam Rasch (eds), Unlike Us Reader: Social Media Monopolies and Their Alternatives, Institute of Network Cultures: Amsterdam, pp. 127-136.

Kyriakou, H, Englehardt, S, and Nickerson, J V, 2012. ‘Networks of Innovation in 3D Printing’ Presented at Workshop on Information in Networks, September, New York.

MakerBot, 2014(a). Community. Available from: [Accessed: 7 May 2014]

MakerBot, 2014(b). About. Available from: [Accessed: 7 May 2014]

MakerBot, 2014(c). Terms of Use. Available from: [Accessed: 7 May 2014]

McCarthy, R 2012. ‘Our Lawyer Explains The Thingiverse Terms of Service’, MakerBot. Available from: [Accessed: 7 May 2014]

Molitch-Hou, M 2013. ‘Repables: Thingiverse Made Open (Again)’, 3D Printing Industry, 12 September. Available from: [Accessed: 8 May 2014]

——, 2014. ‘Has MakerBot Become TakerBot?’, 3D Printing Industry, 28 May. Available from: [Accessed: 19 October 2014]

Nielson, M 2011. Reinventing Discovery: The New Era of Networked Science, Princeton University Press: Princeton.

OSHWA n.d., ‘Definition (English)’. Available from: [Accessed: 8 May 2014]

Patent Genius, 2011. ‘Automated 3D build processes’. Available from: [Accessed: 8 May 2014]

Pettis, B 2012. ‘Let’s try that again’, MakerBot. Available from: [Accessed: 8 May 2014]

Prusa, J 2012(a). ‘Occupy Thingiverse Test cube’, MakerBot Thingiverse, Available from:> [Accessed: 8 May 2014]

——, 2012(b). ‘Open Hardware meaning’, Josef Prusa. Available from: [Accessed: 8 May 2014]

Rayna, T, Striukova L, Darlington J 2014. ‘Open Innovation, Co-Creation and Mass Customisation: What Role for 3D Printing Platforms?’, in T D Brunoe, K A Neilson & S B Taps (eds), Proceedings of the 7th World Conference on Mass Customization, Personalization, and Co-Creation (MCPC 2014), Aalborg, Springer, 425-435.

RepRap 2013. ‘Thingiverse Alternative poll on Reddit’. Available from:,149017,237703 [Accessed: 8 May]

Rideout, B 2011. ‘Printing the Impossible Triangle: The Copyright Implications of Three-Dimensional Printing’, The Journal of Business, Entrepreneurship & the Law, Vol. 5(1), pp. 161-177.

Rosenblatt, B 1998. ‘Moral Rights Basics’, Harvard Law School, viewed 7 May 2014, [Accessed: 7 May 2014]

Samuelson, P 1988. ‘American Software Copyright Law’, in Columbia-VLA Journal of Law & the Arts, Vol. 13, pp. 61-75.

Schweik, C M & English, R 2007. ‘Tragedy of the FOSS commons? Investigating the institutional designs of free/libre and open source software projects’, First Monday, Vol. 12(2). Available from: [Accessed: 2 December 2014]

Smith, Z 2012. ‘MakerBot vs. Open Source—A Founder Perspective’, Hoektronics. Available from: [Accessed: 8 May 2014]

Suber, P 2012. Open Access, MIT Press: Cambridge.

Suzor, N 2012. ‘Order supported by law: the enforcement of rules in online communities’, Mercer Law Review, Vol. 63(2), pp. 523-595.

The Economist 2011. ‘More Than Just Digital Quilting’, 3 December. Available from: [Accessed: 2 December 2014]

Thompson, C 2012. ‘3D printing’s forthcoming legal morass’, Wired. Available from: [Accessed: 10 May 2014]

Walter 2012. The MakerBot/Thingiverse move to the Dark Side (Updated), Hackerspaces Flux. Available from: [Accessed: 7 May 2014]

Weinberg, M 2013. What’s the Deal With Copyright and 3D Printing?, Public Knowledge white paper, Institute for Emerging Innovation, 29 January.

[1] Parametric design is a design method in which the output is generated by a set of rules or an Algorithm, normally by using a computer program.

[2] Since data collection was not continuous, the collected data offers only a snapshot view at a particular point in time. Our selected method enables automated processing of large data sets with minimal human intervention and resources, lending itself to generalisations and a broader view; however, the downside is that data often is shallow and does not capture fine nuances or hidden meanings. Community-based ethnographies could offer a more in-depth view, but not at the same extent and certainly not with same small amount of resources and time. Sample studies would also enable detailed exploration of selected items, but the generalisation of results gained would be difficult or impossible.

[3] These are not included in the ‘Private’ status percentage