[Discuss] OSHW questions

Wed Nov 21 09:52:36 UTC 2012

Hi Gabriella

My comments below. Although I am a lawyer, IANAP(atent)L, more of a contracts/copyright licensing kind of guy, so my comments about the patents should be read in that light!

On 19 Nov 2012, at 02:01, gabriella levine <gabriella.levine at gmail.com<mailto:gabriella.levine at gmail.com>> wrote:

In recent weeks I've started to formulate some answers to some questions but these are some of the questions I've been grappling with.

0. Are there guidelines for how to "do open source hardware the right way?"
As in, how to document, where to document, what to publish, how to effectively get community feedback etc...

We're still feeling our way around this. The free and open source software world can provide some examples and case studies, but I think it's dangerous to assume that open hardware will develop in the same way as open software. However, the basic principle, of release early, release often are, I think, key to the development of any open community. This does cause a tension if you think that patenting is helpful in an open hardware context.

1. can I patent a technology then can we also call it open source? or is this then free tech, not open, because we would always potentially own the rights to it, in which case that isn't honest about it being open source?

I think you can, although for the tech to remain open source, it would have to be issued with a very liberal and irrevocable, royalty-free licence. Arguably, that licence would have to be so liberal as to amount almost to an abandonment of the patent in the first place. See http://www.ifosslr.org/ifosslr/article/view/57 for problems with patents and the GPL, for example.

1a. what would be the implications of getting a patent and can we make our own stipulations to a patent (ie we can never withhold information about it even though we own it?)?

An important (and often overlooked) part of the patent bargain is that in exchange for getting a limited monopoly, you have to publish the patent, so that anyone skilled in the art can understand the invention, and, potentially, reproduce it. So (at least in theory, assuming that the patent system works as it should and that over zealous drafters are not succeeding in obfuscating the invention) the you cannot by definition withhold information, as it is available as part of the public record. You can, of course, make stipulations in a licence (or at least conditions so that so long as a third party complies with the conditions, they are within the scope of the licence and therefore you cannot pursue them for infringement).

One stipulation you might like to consider is a copyleft-like principle that explicitly states that anyone making an invention which would otherwise infringe your patent may not distribute it or any derivate unless they also release appropriate design documentation, and that is released under the same form of licence.

 2.would a patent or a provisional patent be something worth considering, for protection against someone taking my technology and patenting it?

This is one way of applying a copyleft-like principle. However, the problem that i have is that (1) patents are expensive to file and maintain; and (2) that the secrecy necessary during the development process are at odds with the collaborative and open approach.

2a.What does a patent or provisional patent do: protect against anyone else that tries to do what we’re doing?

I'm not sure what a "provisional patent" is - I assume it's similar to "patent pending" in UK terminology. A patent, once granted, grants the owner exclusive right to manufacture (or use, in the case of a process) the subject matter of the patent in the jurisdiction in which it is granted for the duration of the patent (and subject to a few exceptions), and to license others to do that. Once granted, you can claim damages against people who were exploiting the patent during its pending period.

 3. could part of the technology be patented and part be open source?
 Yes.

4.is<http://4.is/> my technology "un-patentable" after 1 year of documentation on line about it?

If the subject matter of the patent is available online with sufficient detail do disclose that it's patentable subject matter, than it's likely to be unpatentable (or at least, someone will have have good grounds to challenge the grant of the patent later)

4a. what consists of publishing / documentation? online blogs, posts, or journals..?

Yep. I would say if it's indexed( e.g. google)  in any way, then in certainly is publication. However, even it's in a publicly accessible backwater of the web which is not crawled, then you're potentially open to a challenge.

5.Even as open source hardware technology, do I have to be careful and do patent research to make sure no one has patented something similar?

>From the UK perspective, it's sensible to do that, although patent searching is more of an art than a science, and it's a very difficult job, so people rarely bother. I understand in the US that undertaking a search can be a bad idea, because it's then easier to construct an argument that you are a knowing infringer of another patent, and find yourself open to multiples of damages.

If you contribute to a project under an agreement which requires you to warrant that you have no knowledge of any potentially conflicting patents, having done a search could be a bad thing (e.g. you may find a patent you think is problematic, but after analysis, you think it's ok, but it turns out your analysis is subsequently incorrect).

6. What is the benefit of an open hw corporation using license that is hosted by another company?

Not sure what you mean by this. If you are talking about reusing an existing licence, then that is a good idea. Let's try to avoid licence proliferation! This is doubly true if the licence has any element of copyleft, as copyleft licences are generally incompatible with each other (although we are trying to address this explicitly with TAPR and CERN OHL).

6a. Who would then own the IP? What does it mean to own IP? Is oshw IP always worth $0 because it is free and open to the public?

The IP (by which, in this case, I take it you mean the patent) will belong to the inventor or his/her assignee. It's not necessarily worth $0, depending on the licence, as, especially in relation to a  copyleft licence, you can also issue the tech to someone else under a proprietary licence. This is relatively common in the open source software world - for example, MySQL has a dual proprietary/GPL licence. Also, GPL violations cases often end with the violator paying damages and receiving a proprietary licence. (Richard Stallman doesn't like this, though: he calls it "selling exceptions to the GPL").

7. Would there be a benefit to having a specific open source license for different companies with slightly different needs or one general license that works for many cases? What licenses already exist for oshw and what are the differences between them?

This has happened in software, and it's annoying. So the Eclipse, Mozilla and CDDL are very similar, but have different patent clauses.

With hardware, we are (roughly) aiming for a very open Apache-like licence, an licences which are more equivalent to weak copyleft and strong copyleft (although there are issues with the extent of enforceability of the copyleft). This is where the CERN OHL and the TAPR licence come in.

-how do I know what is right for my technology?
It's not necessarily what's right for the technology - it's also what sort of community you want to build around the technology, and how concerned you are about someone closing the tech and making it proprietary. I touch on this in my article here:

http://www.ifosslr.org/ifosslr/article/view/69/131

Id love to hear feedback and would also love to share some of the stuff I've been learning

Please share!

All the best

Andrew