This post originally appeared on the Shapeways blog.

Last month Creative Commons hosted a meeting with members of the 3D printing community to discuss how Creative Commons licenses can continue to work well with 3D printed products and files.  Shapeways was thrilled to be included in the  initial meeting and our hope is that this acts as a catalyst  for a more impactful collaboration between 3D printing and Creative Commons.

The purpose of the initial meeting was for the community to gain a better understanding of how the Creative Commons licenses work with 3D printing today as well as identifying key areas for improvement moving forward. We have raised some questions about the dynamic between Creative Commons and 3D printing in the past, and part of the initial meeting was to revisit these topics with other groups.

While this was the first major Creative Commons and 3D printing meeting, we expect that it will not be the last.  The discussion was wide ranging and freewheeling, and it is probably safe to say that there are a number of opportunities for collaboration going forward.  This is an especially opportune time to begin the collaboration as Creative Commons is in the initial phase of a 5 year plan to reinforce and grow the Commons.

We will continue to discuss the collaboration as it evolves and the initial meeting was successful in helping us to flag a few key areas for focus.  Through discussion, attendees were able to confirm the areas of concern that were generalizable beyond one or two parties.  You can read Creative Commons’ debrief here.  Two of those areas, a micro and a macro, deserve mention here.

Micro-Challenge: Attribution

The question of attribution presents a specific challenge in the context of 3D printing.  It is a “micro” challenge in the sense that it is highly specific, but its impact can be widespread.

image

Fundamentally, the question is: what does it mean to give attribution in the 3D printing community (this was a large part of our earlier discussion about 3D printing and Creative Commons)?  When someone uses a CC-BY license, what do they expect that attribution to look like?

This question actually breaks down into two parts: attribution in the file and attribution in the object.

The file attribution part may be the most straightforward.  There are a number of new 3D printing file formats which are currently under development.  Many of these formats include the ability to attach specific license information to files.  With this capability, giving attribution may be as simple as adding (and maintaining) the correct metadata in a file.  While there are a number of implementation challenges, file metadata does at least appear to be the place to focus on for file attribution.  Building visibility for this feature among developers of new file formats goes a long way towards solving this problem.


image

Attribution in the object is less straightforward.  Some 3D printed objects integrate the designer’s name directly into their surface, making attribution automatic.  However, those cases are the minority.  For other objects, what is the best way to give attribution?  Thingiverse provides physical attribution tags, which can be great for displaying objects at trade shows or at galleries.  However, they may not work in other contexts.  For example, it is probably unrealistic to expect people to hang physical attribution tags from 3D printed bracelets or earrings.  Developing a system on how to handle these types of situations will take time and collaboration.

Macro-Challenge: Applicability of Copyright-Based Licenses

You can expect to hear a lot more about this issue going forward.  This concern is larger than 3D printing, although for various reasons 3D printing is an area that tends to throw it into stark relief.  Briefly, Creative Commons licenses are based in copyright.   However,  unlike the photos, movies, songs, and stories that formed the original core of the Creative Commons, not all 3D printed objects are eligible for copyright protection (this paper takes a deeper dive into why that is and what it means).  For those objects, the terms of the Creative Commons license are not enforceable.

The implications of this incompatibility are potentially far reaching, and the initial meeting was not the time to describe them all as other topics took precedent.  Even explaining the problem can be complicated, and one thing that we will be working on going forward are more and better ways to communicate this challenge to the community.

Keep an eye out for more to come soon. For now, we are excited to be developing a community of stakeholders who will drive the collaboration with Creative Commons to bring you the most valuable information.  We will continue to keep you up to date on these efforts and plan to invite community participation as these efforts expand.  Until then, please comment with any questions or concerns.

This post originally appeared on the Shapeways blog.

Imagine you are selling a porcelain coffee cup in your Shapeways shop.  It is a complicated piece with a complex handle and an intricate design on the outside of the cup.  The lip of the cup is also funkily askew.  After the cup has been selling for a few weeks, you get a letter informing you that – unbeknownst to you – the lip of your cup may infringe on someone else’s design patent.  If the letter is correct, what kind of damages should you have to pay to the person who owns that patent?  Should you have to pay them all of the profits you have made on the cup, or just the percentage of the profits that can be attributed to the lip covered by the patent?  Shapeways, along with Engine, just filed a brief with the U.S. Supreme Court urging them to get this rule right for 3D printing designers.

The U.S. Supreme Court is trying to decide which rule to apply in the context of a long running dispute between Apple and Samsung over smart phones.  Apple had a handful of design patents on elements of the iPhone such as its shape and icons, and accused Samsung of infringing on those patents.  Apple won the dispute, which meant that the court had to decide what damages Samsung should pay.

In regular utility patent infringement cases, infringement damages are generally tied to the value that the part of the object covered by the patent brings to the entire object.  So if, for example, a car company infringes on a patent for a bolt that holds the tire onto the car, the damages are proportional to the relative importance of that bolt to the entire car.  In other words, the fact that part of a car infringes on a patent does not mean that the patent holder gets all of the profit generated by the entire car.

However, a lower court interpreted the rules differently for design patents.  There are many elements of a Samsung phone that make it valuable for a purchaser that are totally unrelated to its rounded edges (for example, its ability to make and receive phone calls).  However, the court decided that because the Samsung phone incorporated elements that infringed on Apple’s design patents Samsung should have to give Apple all of the profits from the phone.  The U.S. Supreme Court is now trying to decide if that rule makes any sense.

What Does This Have to do with 3D Printing?

Although Shapeways designers do not sell smart phones in their shops (at least not yet….), they do offer many objects that could potentially be covered by a design patent.  That means that they could potentially be infringing on design patents – even on design patents they do not realize exist.  If and when that happens, it is important that the remedies for infringement are proportional to the value that the elements protected by patent bring to the final object.

In our brief, which was primarily drafted by Katherine McNutt and Hyosang (Mark) Kim of the Juelsgaard Intellectual Property and Innovation Clinic at Stanford University Law School, we argue that design patents should use the same sort of proportional damage system as the utility patent system. Just as someone who owns a patent in a bolt should not be able to recover the profits for an entire car, someone who owns a patent in a lip should not recover the profits for the entire cup.

Fundamentally, this sort of proportionality is the just result in disputes over complex objects that include many elements unrelated to a single design patent.  It also avoids problems where the first patent holder who sues recovers all of the profits, leaving any additional patent holders with limited options for recovery.

It also helps disincentive the creation of design patent trolls.  If a patent holder can lay claim on a designer’s entire profits simply because they own a design patent that may cover a small element of an object, the value of a design patent infringement claim increases exponentially.  The prospect of recoveries well beyond the value of the patent’s contribution to the final object could encourage the same sort of patent troll behavior that we have seen in traditional patents.

Although the brief was filed this week, it will be some time until the Supreme Court makes a decision in this case.  When they do, we’ll be sure to update you on the case and try to explain what the decision means for our designers.

Open source hardware traces its roots directly back to the open source software community. Despite these roots, hardware is very different from software.  As a result, while some of the norms of open source software port easily to the world of hardware, some are an imperfect fit.  One of the areas of imperfect analogy is in the world of licensing. Fundamentally, this is because open source hardware is not as completely wrapped up in copyright as is open source software.  As a result, open source hardware licenses generally lack one of the two major features of open source software licenses.

Software Licenses

Software is, as a category of work, eligible for automatic copyright protection the moment it is written.  As a result, software can be thought of as “born closed.”  By default, in most cases using (which requires copying) someone else’s software without permission violates their copyright.  In order to change this default state, creators of software use licenses to allow other to use their software without risk of a copyright infringement lawsuit.  Fundamentally, these licenses can be thought of as binding promises not to sue a user for copyright infringement as long as the user complies with the terms of the license.

These license serve at least two purposes.  For users, licenses clear the threat of a copyright infringement lawsuit by laying out clear rules of use. Users can be confident that as long as they follow the rules of the license, they cannot be sued for copyright infringement by the original author of the software.

For creators, licenses let the software out in the world while still allowing the creator to exercise some control over the software via license conditions.  The creator gives up some control over the software, but can be confident that future users are required to comply with whatever limitations incorporated into the license (be it sharing future work, buying the creator a beer, or whatever).  

The single license achieves both of these purposes: assuring users that they will be free of infringement liability, and assuring creators that future users will have to comply with restrictions that the creator cares about (this can also be thought of requiring future users to ‘follow the rules’).

Hardware Licenses

Unlike software, hardware is not completely protected by copyright. While some parts of hardware are eligible for copyright protection, many functional elements simply are not.  While they may be eligible for patent protection, since (unlike copyright protection) patent protection is neither free nor automatic, generally speaking many parts of hardware will be unprotected by any type of intellectual property.  This is especially true for open source hardware, where the creators are unlikely to elect to spend time and money obtaining patents on their goods.

In many ways this is a positive thing.  Unlike software, hardware is largely “born free” and does not require a license to make it available to the larger world.  This is the reason that you can legally hack hardware you own in a what that does not really have a parallel in the world of copyright.  However, this very freedom complicates open source hardware licenses because it makes them harder to enforce.

That does not make open source hardware licenses without value.  They can still accomplish the first purpose of open source software licenses, which is to relieve end users of anxiety related to IP violations that might occur with use.  To the extent that the hardware is protected by some sort of intellectual property rights (and many piece of hardware have at least some protected component that this would be relevant for)*, the license can give users confidence that using the hardware within the limits of the license will prevent them from being sued for infringement.

However, most open source hardware licenses are less good at achieving the second purpose.  As noted above, while many parts of any given piece of open source hardware may be protected by copyright, many parts will not. That is especially true for functional parts of the hardware.  Since these parts are not protected by IP, using them does not require permission from a rightsholder (because that rightsholder does not exist because there are no rights to hold).  As a result, there is no penalty for violating the terms of the license that purports to limit the use of these parts.

The outcome of this is that in many cases end users will be able to violate the restrictions that a creator places on a piece of open source hardware without fear of an infringement lawsuit.  In these instances, the open source hardware license has failed in that second purpose outlined above because they do not allow creators to control the behavior of downstream users.

Can This be Fixed?

To give a lawyerly answer, that depends on your definition of “fixed” is.  One way to fix this problem would be to make it easier to get some sort of IP protection on hardware.  That would make it more likely that licensees would be able to enforce the terms of their licenses. However, because the law works for everyone not just people doing open source hardware, any expansion of IP protection on hardware would also be used by people who don’t care about openness to make hardware more closed.  As I’ve written a number of times, it would be tragic if the legacy of open source hardware was to make hardware more closed.

The other way to fix this problem is with education.  The inability to impose conditions on end users is only a problem if creators have the expectation that they can impose conditions on end users.  That expectation creates a likelihood of frustration for creators who think that a bad actor is violating their license, just to find out that their license does not really apply.  However, if creators understand the purposes and limitations of open source hardware licenses, that expectation can be eliminated (or at least reduced) and the problem is essentially resolved.

That’s the direction I’m inclined towards.  Admittedly it requires altering norms and expectations that have been built up by open source software for decades.  However, it won’t be the the only time that open source hardware has to alter some of what it inherits from open source software.  


*This highlights a second major educational challenge around open source hardware. With open source software, teaching people about IP is fairly easy: they can assume that all software is protected by copyright. Hardware is much more complicated.  While any given  piece of hardware is likely at least partially protected by copyright, what that copyright actually covers will be somewhat idiosyncratic to the hardware itself.  As a result, it is going to be hard to develop everyday rules of thumb for what open source hardware licenses really do and do not control.  I’m hopefully that at some point a taxonomy of some sort will emerge, but that probably won’t happen anytime soon.

image

The version of this announcement that ran on the Shapeways blog is below.  I just want to preface it with a huge thanks to my wife Jessica for doing all the real editing of this paper (as she does for all of the papers I write), Sherwin Siy for looking over this for me, Sophie Khan and Geoffrey Mann for letting me use their images to illustrate “non-representational” scans, Henrik Rydberg for designing the layout, Parker Higgins for inspiring footnote 11, and Shapeways (especially Marty Meyer) for letting me keep doing these papers.  I also suggest quite strongly checking out Professor Margoni’s paper linked to below for an EU perspective and Paul and Laura’s forthcoming paper for a real law review article treatment of this stuff.  I’ll add a link to it here once I have one that is not behind a paywall (if such a link exists).  Also, Justin Hughes’ paper on photography and copyright is really great.

We are excited to announce a new whitepaper, 3D Scanning: A World Without Copyright*.  As the name suggests, the paper examines how 3D scanning intersects with copyright law.  We are big fans of 3D scanning here at Shapeways, and so we thought it was important to start a discussion around how copyright might impact all of the scans that are coming into the world.

It may come as a surprise, but in many cases 3D scans will not be protected by copyright.  That does not mean that scans are not important, but it does mean that people making and distributing scans should understand what rights they do – and do not – have in those scans.

Why aren’t the scans protected by copyright?  One of the key requirements for copyright in the United States is originality. Even if it takes a large amount of skill to create a scan, if making the scan does not involve originality it is simply not eligible for copyright protection.

The vast majority of scans fall squarely in that category.  By definition, most 3D scans attempt to create a perfect digital replica of the model being scanned.  Injecting “original” content that deviates from the object being scanned into that digital file would undermine the purpose of the scan.

Again, lacking originality does not mean lacking skill.  Making accurate 3D scans can be hard work, and there is a real difference between someone who knows what they are doing and an amateur.  But without room for creative interpretation, the resulting scan is not eligible for copyright protection.

Even without copyright, people making scans still have the ability to profit from and control their scans.  Today many professional scanners charge for scanning services, and someone in possession of a scan file can charge someone else for access.  This system seems to work; the ranks of professional scanners is swelling as 3D scanners become cheaper, better, and easier to use.  These types of professional arrangements are strong because they are built on contracts which do not require copyright in order to be legally enforceable.

A lack of copyright on scan files will also make it easier for designers to access and build upon digitized works.  It means that the person who scans a 2,000 year old Roman sculpture does not suddenly pull part of that sculpture out of the public domain.  Once the sculpture is digitized anyone is free to print their own copies to have at home or in classrooms, increasing access to our collective digital heritage.  Furthermore, those scans can be freely remixed into new interpretations of classic works.  These scans can form the building blocks of all sorts of new creativity – new creativity that might itself be protected by copyright.

It is also important to recognize that just because most 3D scans are beyond the scope of copyright does not mean that all 3D scans will be.  Scans that do intentionally inject original, creative elements such as Geoffrey Mann’s Shine and Sophie Kahn’s portraits are likely to be eligible for copyright protection.  And, of course, 3D scanning is relatively new and the law around it could evolve right along with the technology.

Obviously we are going to keep our eye on this area of law and will do our best to update you as developments occur.  If you are intrigued, I would encourage you to check out the new whitepaper here.  For a more EU-centered take on 3D scanning and copyright (especially in the context of access to cultural heritage), try this paper by Professor Thomas Margoni.  And, since our whitepaper is primarily written for a non-legal audience, if you are looking for a law review article that addresses many of these same issues keep your eyes peeled for an article by Paul Banwatt and Laura Robinson of Matter and Form in a forthcoming edition of the Intellectual Property Journal.

This post originally appeared on the Shapeways blog.

With the relaunch of the Shapeways Designer For Hire program, we are seeing more and more designers and clients coming together to create amazing 3D printed objects. While that is unquestionably a good thing, more people coming together also means more opportunities for misunderstandings. (Buzzkill sentences like that are a reason that lawyers aren’t always invited to parties.) Lauren covered some of big areas of misunderstanding with her 8 Questions Your Client Doesn’t Know to Ask You. This post is going to only address one area: who owns the files and the copyrights at the end of a design job?

Thinking about this in advance can help head off a number of disputes, and getting an agreement in writing can be even better. You may see “getting it in writing” and immediately think to yourself “but contracts are boring.” You are right. Contracts are boring. But like other boring things that manage to stick around like seat belts and automated data backup, they can also be really useful in a pinch.

Getting this sort of thing in writing serves at least two purposes. First, and probably most importantly, it forces the designer and the client to get on the same page before time and money are spent. If they don’t agree, it is much better to know that before a job starts than to wait until the end. Second, once both sides have agreed in writing, it makes it easier to resolve disputes that do come up at the end of a job.

Who Gets the Files?

The first question to address is who gets the files at the end of the job.  Many designers assume that they will keep the files for a model, while clients assume that they will get the files at the end of the job. This obviously creates the opportunity for misunderstanding.  Hashing this out at the beginning of a job can avoid lots of heartache at the end.

When you are discussing files, make sure that you are specific. There is a big difference between getting an .stl file for a model and getting something with a bit more data like an .obj file.

Also, there is a distinction between having possession of the files and being able to use the file going forward. Regardless of who has the files, how they are used is probably governed by copyright. Which, naturally, raises the question of…

Who Gets the Copyright?

Especially if the model is nonfunctional and decorative, it is probably protected by copyright. In most cases, the designer will own the copyright in that model when it is created (assuming the copyright exists).

The copyright gives the copyright holder the ability to control things like how many prints are made from the file, or if the file can be modified in the future. Talking about what – if anything – happens after the commissioned print is made can help shed light on how important doing these types of things are to the client.

If the client wants more than a set number of printed models at the end of the job, they will probably need both the 3D model file and some sort of license to the copyright that protects the model.  There is a lot of flexibility to what the terms of that license can be. It can restrict the client to making exact copies via a designated printing service (like, say, Shapeways), require the client to report to the designer every time a print is made, or allow the client to modify the file freely. If the client really wants full control of the model, they could even require that the designer transfer the copyright to the client.

What is the Answer?

Every designer-client relationship is different, so there is no single answer that will apply to every situation. If you have a specific question, it could be useful to consult a lawyer before moving forward with the agreement.  AIGA’s model contract can also serve as a great jumping off point for discussions between a designer and client. As a bonus for designers, once they decide which parts of that contract they want to use it can just become their default client agreement.

Ultimately, the most important thing is to discuss these questions at the start of the relationship. If the client and designer are on the same page, it will be a quick discussion. If they disagree, it gives them the opportunity to work it out at the start of the relationship. Sometimes that disagreement can be resolved by the designer charging an extra fee for additional rights or access to files. Usually the designer and client can come to agreement, although sometimes they cannot. In either case, it is always better to discover that at the start of a job than at the end.