The State of Open Source 3D Printing: Make Copycats and Freeloaders into Assets

Josef Průša, the founder and CEO of Prusa Research, published an important and thought provoking article titled “The state of open-source in 3D printing in 2023” where he called for an open discussion about a new license for 3-D printing. This is the third in a series of articles to move that discussion forward for the benefit of the global 3-D printing community.

First, it is clear the mine-field of intellectual property (IP) law is a mess. Delays in granting patents are often longer than the innovation cycle (patents are obsolete before they are granted). Patents are written mostly by lawyers and are thus far from the quality or completeness the open hardware community is accustomed to in our documentation. Engineers at conventional additive manufacturing firms are warned to not to even look at patents to prevent willful infringement. Then there is a raft of academic literature that has shown that patents reduce innovation – the exact opposite of the intended purpose of patents. 3-D printing is even cited as a technology that reduces the cost of innovation and thus the need for a 20-year monopoly.  Existing open source licenses in 3-D printing are also messy and nothing quite works right for the open hardware community. It is not super clear, for example, if an OpenSCAD script that generates a physical 3-D design of a useful product should be protected via copyright, patent, or both, and exactly how one should do that if they want it to be free and open source. Does this change if you get an AI chatbot to write the script? Or if you ask the AI to generate the script to solve a specific problem rather that dictate the design? At least for now, it appears an actual monkey is protecting all of us from being restricted by a tidal wave of AI IP. Today, non-human actors like AI systems are not entitled to copyright protection. Time will tell if the courts prevent them from turning the patent system into an even bigger morass than it is now.

Although there are open hardware licenses (e.g. the CERN OHL), most developers opt for Creative Commons or the GNU General Public License (GPL), which was first developed for free software. The GNU GPL guarantees end users the four freedoms to run, study, share, and modify the software. This freedom includes selling the software as long as adaptations are shared with the same license.

There are many 3-D printing companies built on the backs of open source designs that either do not release source code, delay releasing it, or only release parts of it. Worse, this is often only after the community is forced to shame them. In addition, many of these companies are not giving appropriate attribution, either out of laziness, sloppiness, or malice.

Průša argues that the GNU GPL is very vague, written in a complicated way, open to various interpretations and ‘developed by academics for academic purposes. Respectfully, and I do have enormous respect for Mr. Průša, none of that is entirely accurate. First, all such licenses are a chore for normal people to read because it is ‘lawyer speak’. It was originally developed by software activist Richard Stallman to be used to protect software in the real world, not as any kind of academic exercise.  As the GNU GPL has been so popular in software circles for so long, it has a large advantage over newer licenses because it has been tested in court and we now know that it works. You can successfully sue people for breaking it. Now, doing this practically is complicated as regulations differ in each country, and license disputes can be long and expensive. No one wants real 3-D printing companies to waste their time or resources on that. In fact, ignoring an ‘IP strategy’ saves open source 3-D printing companies enormous amounts of time and money that would otherwise be wasted on lawyers that add nothing to real innovation.

It would be great if a larger overarching organization could provide some cover. Prusa Research has reached out to the Free Software Foundation, that ignored them – even for money. What the 3-D printing community really needs is someone like the legendary consumer-advocate Ralph Nader, who sued GM for harassment, won and then used the money to set up the non-profit Center for Auto Safety doing the very thing that GM was harassing him about. If an open source 3-D printing advocate were to win a large case and then use the proceeds to set up a contiguous gauntlet of open hardware protection, the entire community would benefit as it would allow many companies to flourish in the space.

Making copycats and freeloaders assets

Until this happens, we still have the problem of copycats and freeloaders. Prusa points out that there are 1:1 clones of hardware or software on the market that currently do not bring anything back to the community. This is legal by all current fully open source licenses. Having more well-documented 3-D printers, even if they are simple clones out in the wild, does benefit the community in modest ways. At the very least it should drive down component costs for suppliers and have spare parts more easily accessible for everyone. If even a small fraction of the copy-cat users share free 3-D printable designs, they are also adding value back to the open source community. There are millions of free designs now, but each one makes every printer more valuable. Companies that became open source platforms like Arduino are successful in part because of the widespread cloning. Counterfeit companies that use trademarked open source names, however, are breaking the law and should be squashed. It would, of course, be better if there were reciprocal contributions from all companies. Until, however, we get our next Ralph Nader, it might be useful to think about how we can use technology and policy to make copycats and freeloaders an asset to the open source 3-D printing community.

Comparison of an open source (left) and commercial version (right) of a blaster toy (rockets2, 2021; Amazon, 2021b).

For example, better automated troubleshooting and issue management would benefit from a network effect of having more users, even if they are not original customers, and open source companies could brand that, which would help with advertising. Maybe a newbie’s first 3-D printer is a cheap copycat, but then they upgrade to the real deal to get new innovations, materials or buy upgrade kits, etc.

Similarly, there may be opportunities to mature 3-D printing materials. Finding optimal printing parameters for a new 3-D printing feedstock is very time consuming by trial and error, even for experienced researchers, and is even a good bit of work when using an optimization algorithm. These parameters will also shift for the model being manufactured, particularly for complicated and challenging-to-print designs. Mounting cameras on 3-D printers and having the default script be to send images of prints in process, along with the printer settings, can help train machine learning algorithms that could then be used to optimize printing settings for specific models, or even help us make smart 3-D printers that will correct mistakes mid-print. Software could be structured that this setting would need to be enabled to take full advantage of the communities’ learning and optimal print parameters.

It would, of course, be better if freeloading firms actually tried to make the printers better and shared their improvements, but in these ways, even exact copies with no effort put into improvements would help the entire community and the open source firms playing by the spirit of the rules.