Yesterday the Copyright Office released all of the answers to the questions they had after the unlocking hearings.  Turns out that Stratasys, Public Knowledge, and I all agree!  We all told the Copyright Office that there is no meaningful way to make some sort of distinction between “commercial” and “non-commercial” users of 3D printers for the purpose of the exemption.  What’s going on here?

Both during the hearing and in its follow up question, the Copyright Office seemed interested in exploring granting some sort of partial exemption for unlocking 3D printers.  It is always dangerous to look too deeply into the tea leaves of questions to guess what people are thinking, but it seemed like the Office was considering an unlocking exemption for a category of small, personal printers or users while not granting the exemption for larger commercial printers or users.  The Office’s follow up question asked participants if this sort of distinction was viable.

All three of us answered no, essentially making the same types of points: Professionals use lower cost desktop printers for work and everyday people use expensive “industrial” printers for fun (via services such as my employer Shapeways).  Printers can switch back and forth between what they print (or mix batches at the same time) and users can similarly switch what they are doing. 

At some level, all of us are making the same gamble: instead of tacitly accepting some sort of compromise from the Copyright Office where some printers get unlocked and others don’t, we are pushing the Copyright Office to make a real choice: either all 3D printers get an exemption or none do.  No legal distinctions that don’t make sense, no half measures.  Obviously, PK and I are pushing for the exemption while Stratasys is pushing against it, but our strategic thinking is probably similar. We’ll see in the next few months which gamble (if any?) pays off.

The Makerbot Twist

There has been a Makerbot subplot running through this entire proceeding, and the Stratasys response added another chapter.  Stratasys owns Makerbot and a significant amount of the discussion during the Copyright Office hearing discussed Makerbot specifically.

One of the things that I took from the hearing was that Stratasys was seriously considering adding DRM to their Makerbot line that locked Makerbots to Stratasys-approved filament. However, a week later the new Makerbot CEO (who, not incidentally, is also the son of the Chairman of Stratasys proper so presumably has deep ties inside the company) told Adafruit that he had no plans to DRM Makerbot filament.

Obviously the timetable is a bit compressed, but Stratasys’ response to the Copyright Office could have provided them an opportunity to clarify their plans with regard to Makerbot and DRM on the record.  The fact that they didn’t take that opportunity doesn’t inherently mean that they are being duplicitous (there are a million legitimate internal logistical reasons that the statement didn’t make it into the filing).  But it certainly does seem fair to note it as a missed opportunity.  The impression that Stratasys gave the Copyright Office was that the option to add filament DRM to Makerbot printers was important to them. This could have been a good opportunity to correct that impression, and it was one they did not take.

image credit: flickr user Mark Dumont.

After the hearing, the Copyright Office submitted an additional question about unlocking 3D printing.  The question was:

During the hearing for this class, opponents of the proposed exemption expressed concerns that parts made with substandard materials could enter the commercial supply chain and pose risks to the public, citing the example of airplane parts printed with substandard filament. The Office welcomes comments addressing whether an exemption could or should differentiate between “commercial” versus other types of uses and, if so, how those different categories of use might be defined.

The deadline, originally last Monday, was extended to today.  In part, this was because the Copyright Office was a bit slow getting transcripts of the hearings (which, it is worth repeating, they didn’t make easy for anyone to record and did not record themselves - not even by tapping into the soundboard!) up on their site.  My admittedly somewhat snippy answer is below.  I’ll update with a link to the PDF that has footnotes once they are up on the site. (edit 7/1: here’s a post with all three replies and a discussion of what they mean)

June 29, 2015

Jacqueline C. Charlesworth

General Counsel and Associate Register of Copyrights

United States Copyright Office

Library of Congress

101 Independence Ave. SE

Washington, DC 20559

Re:    Docket No. 2014-7

    Exemptions to Prohibition Against Circumvention of Technological Protection

    Measures Protecting Copyrighted Works

    Proposed Class 26: Software-3D Printers

Dear Ms. Charlesworth:

Thank you for your follow up questions.  It is unlikely that the Librarian could draw a meaningful distinction between types of 3D printer users without inadvertently including or excluding many users.  As such, I do not recommend differentiating between types of users for the purposes of this exemption.  Instead, the Register should recommend this exemption for all 3D printer users.

Blurring Distinctions

To the extent that there was ever a formal distinction between consumer and professional users of 3D printing, that distinction has long since eroded.  Even Stratatsys has positioned what may have once been considered a “consumer-oriented” printer - its Makerbot line - as a “prosumer” machine.  This move towards prosumer branding of lower cost printers was noted as early as 2012 - making it old in 3D printing terms - and has only accelerated since.

Indeed the existence of the prosumer - whether you see the term as a portmanteau of professional and consumer, producing and consumer, or proactive and consumer - serves to highlight the challenges inherent in drawing distinctions between users in the world of 3D printing.  Relatively low cost desktop machines are used in commercial and industrial settings everyday.  The e-NABLE community uses them to create production prosthetic hands.  Medical device designers use them to prototype designs and print anatomically correct surgical models.  In fact, Stratasys’ own Makerbot has an entire section of its website devoted to professional use cases for Makerbot printers.  Conversely, services such as Shapeways and Techshop make traditional higher end business machines available at a low cost to users focused on more personal, non-commercial applications.  

As the pace of innovation increases, even the functional distinctions between less expensive and more expensive machines begin to blur.  Technologies that were once only available in machines costing tens or hundreds of thousands of dollars are migrating into machines available for three or four figures.  The result of this is that any given model of machine is likely to be used by a broad spectrum of users for a broad spectrum of uses.  Even machine cost cannot act as a reliable proxy for differentiating between machines for commercial or personal users.

Attempting to draw distinctions based on what is being printed is no more straightforward.  An object can move from idea to prototype to final product on the same machine.  Indeed, some objects that begin as personal projects may evolve into commercially available products over time.  Similarly, a single object can contain elements printed on numerous 3D printers of varying levels of technical capability.

One of the advantages of 3D printing is that it blurs the line between prototype and final product.  Another is that it puts the production of finished, professional products into the hands of everyday users.  Without the need to retool between jobs, a single printer can easily begin the day printing personal trinkets and end the day printing components of a commercial product without ever changing the material used to print.  In fact, sometimes a single printer has batched jobs and is printing both at the same time.  

No Reason to Exclude the Commercial Supply Chain from the Exemption

Some non-copyright-related objections to this exemption request focus on the use of 3D printers specifically  in the commercial supply chain.  To the extent that the Copyright Office is considering those non-copyright-related concerns as legitimate in the context of this exemption evaluation, it would seem to be especially nonsensical to prevent participants in the commercial supply chain from using third party consumables in their printers.  

Setting aside the compelling reasons for all 3D printer users to be able to use consumables of their choice as they see fit, in many ways commercial supply chain manufacturers are the best positioned to evaluate the legal and safety ramifications of their preferred consumables.  As a group, such manufacturers are more likely to have the type of technical and legal expertise to evaluate the impacts of using third party materials in their production.  While this level of technical and legal expertise should not be a requirement to use third party consumables, it would be somewhat nonsensical to exclude the parties most likely to have such expertise from this exemption.

Industry-Specific Concerns Are Best Addressed Elsewhere

3D printers are general purpose machines.  That means that they can be used across industries for an almost unimaginably wide variety of purposes.  Because a single machine can print multiple objects simultaneously, one 3D printer can be used across multiple industries at the same time.

Opponents have raised speculative concerns about the use of 3D printing in specific industrial applications such as airplane construction.  With respect to the Copyright Office, such concerns are well beyond the expertise of this proceeding.  The Copyright Office and the Librarian of Congress do not possess the expertise to evaluate the potential threats raised by the use of third party consumables in the construction of airplane parts.  Nor do the Copyright Office or the Librarian of Congress have the expertise to evaluate the value of technological protection measures designed to protect copyrighted works in preventing the types of industrial sabotage described by opponents.

These types of concerns are beyond the scope of this proceeding in part because there are other parts of the Federal Government with a mandate to oversee them.  For example, in the context of airplane safety, nothing in this proceeding will undermine the Federal Aviation Administration’s mandate to protect the integrity of airplane parts.  Similarly, anti-counterfeit laws, product liability, and supplier contracts will remain intact to maintain the integrity of commercial supply chains.  None of these non-copyright-based protections hinge on the status of this exemption.  To the extent that the Copyright Office feels the need to evaluate these threats, it is noteworthy that no representatives of industries such as airplane manufacturing, or of the commercial supply chain more broadly, felt compelled to raise concerns about the proposal.

Concrete Benefits Outweigh Hypothetical Harms

While the Copyright Office and the Librarian are poorly positioned to evaluate airplane safety or supply chain integrity, both are well positioned to evaluate the impact that this exemption will have on copyright.  Viewed in that context, the concrete benefits of this exemption outweigh the hypothetical harms described by opponents.  Opening the door to consumable competition from third parties will have a direct impact on innovation in this field.  Competitors can take steps to compete directly with manufacturers to provide existing types of consumables, driving down prices and driving up quality.  Innovators can also move to bring new products to market, increasing what is possible with 3D printing.  Any user can retain confidence that owning a 3D printer means being able to decide what types of materials are used in it without threat of 1201 liability.

Balanced against these benefits are harms that stray far afield from the purpose of both copyright law and this proceeding.  By focusing on doomsday predictions related to supply chain integrity, opponents tellingly avoided raising concerns about infringement to their software.  Unable to credibly describe a concern related to copyright infringement, opponents instead fell back on speculative fears unrelated to copyright law.  The lack of copyright-specific harms raised by opponents should give the Copyright Office comfort in recommending this exemption to the Librarian.

In light of these factors, I respectfully request that the Register recommend that this exemption be granted for all users of 3D printers.



Michael Weinberg


One of the most exciting things about widespread access to 3D printing is how it has started to push cultural institutions to begin digitizing their 3D collections.  Now, in addition to being able to see free high quality 2D scans of paintings like a 15th Century Italian Pentecost  and 18th Century Japanese Woodcuts, you can see (and sometimes download, print, and modify) high quality 3D scans of the Cooper Hewitt Mansion, Abraham Lincoln’s face,  and Musette the Maltese Dog.  With objects reaching back thousands of years scattered across cultural institutions around the world, it isn’t hard to imagine a future where the world’s cultural heritage objects are available to anyone with a 3D printer (or, say, a Shapeways account).


We’ve been able to access high quality 2D world heritage items online for years. (Image from the J. Paul Getty Museum Open Content Program, which is awesome).


Why not high quality scans of 3D objects? (image courtesy MetMuseum)

But a question about copyright is lurking in the background of this glorious future.  Specifically, a question about copyrights in the scans of the objects themselves: are 3D scans protected by copyright?  If the answer is yes, scanning could drag parts of cultural heritage objects away from their home in the public domain and lock them up behind proprietary walls for decades.  That would make it much harder for people to access their own cultural heritage.

Fortunately, at least one court in the United States has found that scanning an object does not create a new copyright in the scan.  That means that scanning a 9th century Hanuman mask doesn’t wrap the scan in a new copyright.   However, a paper from earlier this year by Thomas Margoni illustrates that the copyright status of scans is not as clear in the European Union.  That lack of clarity alone could slow the dissemination of objects housed in Europe’s finest cultural institutions.  Hopefully, the EU will move to clarify that 3D scans of objects do not create entirely new layers of copyright protection.

Remember, in this context we are not talking about copyrights in the objects themselves.  For the sake of simplicity, let’s just focus on the thousands of years of cultural production prior to around 1920 that is well in the public domain.  In these cases we are talking about someone who did not create the original object scanning it and then claiming a new copyright on the scan – and only the scan – itself.

Background: Originality

Originality is a key to understanding why a scan should or should not be protected by copyright.  Originality is a general requirement to obtain copyright protection, although the bar for what qualifies as “original” is famously low.  That being said, while the bar is low it does exist.


It takes a lot of work to put together the phone book. That doesn’t mean it is protected by copyright. (image credit: flickr user James Cape)

Note that in this context originality is not synonymous with “complicated” or “labor intensive.”  Instead, it suggests that the author of the work made creative choices about how to create the work. In a famous US case, the Supreme Court denied copyright protection for the phone book.  The court acknowledged that putting together a phone book takes lots of time, effort, and resources.  But it denied copyright protection because there isn’t room for creative expression in how you assemble a phone book.  The form pretty much dictates that you list everyone in alphabetical order and that each entry starts with a name and ends with a phone number. Given a pool of names and phone numbers, everyone’s phone book is going to look pretty much the same.

The same type of theory can be applied to scanning.  It can take a lot of work and technical expertise to accurately scan a 3D object.  But at the end of the day, the goal is to create as accurate a scan as possible.  Some people may be better or worse at achieving that goal, but the nature of the task does not leave a lot of room for creative interpretation.  Without creative interpretation there is no copyright protection.

That distinction is reasonably straightforward in the US.  However, Margoni’s paper highlights that fact that it is not as clear in the EU.  EU-wide laws designed to harmonize copyright leaves the test for originality up to each member state, and those member states have each structured that test slightly differently.  That means that at least some types of scanning in some EU member countries could be protected by an additional copyright.

Why This Matters


Everyone is sad when cultural heritage objects are locked up. (image credit: flickr user Ania Mendrek)

It would be bad to protect 3D scans with a new copyright because it adds another wall of rights around the object being scanned.  This is especially harmful in the context of scans of world heritage objects.  World heritage objects are part of our collective inheritance.  Adding additional rightsholders creates a barrier for everyone who wants to access that inheritance.

Beyond copyright’s capacity to simply block use, additional layers of protection also undermine confidence in use.

Copyright lasts for a long time, and copyright rules can make it hard to determine the protection status of a given object.  But, at a minimum, a statue from 1900 – or 1900 BC –  is clearly in the public domain.  “Made by a civilization unfamiliar with electricity = public domain” is a rule of thumb that everyone should be able to rely on without consulting a copyright attorney.  If there is the potential for an additional scanning copyright, every time you came into contact with a 3D scan of a world heritage object you would have to ask a host of questions: Who made this actual scan?  How did they decide to license it?  Will I have to worry about someone who made a different scan suing me for copyright infringement?  Regardless of the answer, the mere existence of each of these questions make it less likely that people will make use of the scans.

World heritage objects belong to everyone.  There are already plenty of people trying to pull them into an ownership box without adding an additional layer of copyright protection to scans.  There is no reason to make it more likely that one person will have a veto over how these objects are used.  The Margoni paper is an important step towards understanding how 3D scanning may be treated in the EU.  The next step is making sure that those rules arc towards openness and accessibility for all.

This post originally appeared on the Shapeways blog.

Podcast -Your Rights and 3D, A conversation with Shapeways about IP

Bridgette Mongeon had me on to talk about 3D printing and IP from a designer’s perspective.  It was interesting (for me) to get into the kinds of concerns that professional sculptors have with services like Shapeways in terms of security and accessibility of support.

Today was the last public step in the 3D printing unlocking proceeding (some background here).  Essentially, this is about allowing people who own 3D printers to use whatever their material they want in their printer.  Since some manufacturers use chip-and-software-based verification to lock printers to material coming from them, the entire process was happening as part of the triennial “1201 process,” so named for 17 USC 1201 that authorizes it.

That last step was a public hearing.  On side pro-unlock was Sherwin Siy from Public Knowledge and me from, well, me. I started this process while I was at PK but now that I am at Shapeways was participating in my personal capacity.  On side anti-unlock was a representative from Stratasys.  Stratasys is a large commercial 3D printer manufacturer.  They also own a desktop 3D printing company called Makerbot that you may have heard of.

Rebecca Tushnet already has her always fantastic liveblog/transcript of the proceeding up so I’ll just link to it here.  Some quick thoughts:

  • There were a lot of questions about the copyright-protected work in question and about how the DRM mechanism would actually operate.  Since the petition was (intentionally) written at a fairly high level of abstraction in order to cover all printers, this became a bit complicated.  
  • The Copyright Office seemed interested in exploring distinctions between a consumer/prosumer market and a commercial/industrial market.  This makes a bit of sense since the original petition skewed towards the consumer market.  However, it is hard to articulate a reason why it would be reasonable to grant this exemption for the consumer market and not the industrial market.
  • There was also some discussion about why the underlying copying in question was not an infringement.  This was a bit complicated because I don’t actually believe that there is infringement going on.  However, I do worry that the absence of infringement would not prevent a printer manufacturer from bringing a lawsuit or threatening a lawsuit.  Fortunately, Sherwin was able to talk about both 17 USC 117 and fair use.
  • There were questions about ownership vs. licensing of the software in the machines.  The Copyright Office asked for specific examples of how specific printer manufacturer handle this issue.  I didn’t have specific examples.  It probably would have been better if I had researched the licensing terms of a handful of printers before the hearing.
  • Stratasys/Makerbot essentially restated the concerns that they raised in their opposition filing.  These concerns boil down to things like people making counterfeit objects with sub-standard third party material and users hating Stratasys when third party material makes ugly prints with a Stratasys machine.  The Copyright Office was encouragingly skeptical of this justification for 1201 liability.  As I wrote in my obnoxious reply comments, I share most of those concerns.  It is just that 1201 is not the law that we have to prevent counterfeit 3D printed parts from going into airplanes.  That’s why we have product liability law. And contracts.  And trademark.  
  • In defending their interest in using DRM to exclude third party filament, Stratasys made it pretty clear (to me at least) that they have been seriously considering adding verification chips to Makerbots.

  • Stratasys was rolling deep.  In addition to the person they had testifying, they had at least 4 lawyers sitting in the back.  Those lawyers seemed perfectly nice, but their presence pretty much put to bed any ideas I had that a 3D printer company would never bring a 1201 action against someone who decided to use an unapproved material. 

What happens now?  Assuming the Copyright Office doesn’t have any followup questions, once the hearings are done tomorrow they start deciding how to handle all 27 exemption proposals.  They come up with recommendations (at least in theory in consultation with NTIA - it was encouraging to see a representative from NTIA asking questions today) and give them to the Librarian of Congress.  He then considers the recommendations and decides how to handle each request.  When will that happen?  Who knows?  Maybe this fall?

Will this exemption be granted?  I have no idea. The Copyright Office felt less hostile to the idea than usual and asked good questions probing how the Stratasys concerns connected to copyright.  But they also seemed concerned that Sherwin and I didn’t have a detailed technical analysis of every possible DRM mechanism that could be used in a 3D printer.  So we’ll see.

Oh, and if you want to watch a video of the hearing you are out of luck. The Copyright Office doesn’t know how to record things on video. At some point they will release a transcript. Until then Professor Tushnet’s blog is the closest you are going to get (and it is pretty damn close).