LWN: Comments on "Patent exhaustion and open source"

Patent exhaustion and open source

mathstuf — Thu, 25 Apr 2019 23:01:16 +0000

Ah, that one indeed might not (cursory reading of it seems to confirm that). However it still has the "Google is special and can use as they wish" part which makes them "special" and is not something I like agreeing to. Fedora's PCA is fine because, AIUI, it is basically "FOSS is OK, but if it isn't licensed, we're going to treat it as MIT". I don't remember if I ever signed the old ICLA, but I probably did and was before I learned/started caring so much.

Patent exhaustion and open source

mjg59 — Thu, 25 Apr 2019 22:49:44 +0000

I don't think the Google individual CLA includes a copyright assignment?

Patent exhaustion and open source

mathstuf — Thu, 25 Apr 2019 22:06:16 +0000

I don't know. That ends up running around chasing big corporate demands. I personally don't sign CLAs that assign copyright, but I do so for $DAYJOB because it's their copyright anyways. For example, I have patches in protobuf, but if I weren't being paid to do it, I wouldn't be able to bring myself to donate code (even trivial fixes) to Google because the CLA basically says "we're more important, thanks". I don't see these projects clamoring to tear down CLAs that prevent individuals with principles about them from contributing.

If a company doesn't want to contribute because some community project has policies or whatever they don't like, they can go make their own sandbox and play in it. Nothing has stopped them before. Sure, it'd be great to have them contribute, but I'd have no trouble saying "no, this project cannot afford to be hindered by patents, go away" if patents were a) important to the project and b) blocking some corporate contribution because, IMO, it's not worth it.

Patent exhaustion and open source

TruePath — Tue, 02 Apr 2019 00:18:52 +0000

Moreover, even if this works in a limited way I worry it's a bad thing. We want open source to succeed and that means we want companies to be able to use and contribute to it on the facial terms of the license without great risk.

If anytime a company considers contributing to an open source product (especially one that requires they make their modified source available) they have to get a giant review by patent lawyers then they just often won't contribute and will make their own closed source alternative.

Patent exhaustion and open source

TruePath — Tue, 02 Apr 2019 00:11:27 +0000

That's a clever little trick but I doubt it would work with respect to the Microsoft and Github case. Almost certainly, just like with respect to copyright under the CDA, hosting third party content would not be seen as them distributing the underlying code. Rather, the third party who posted it to Github would be the distributor.

Indeed, your claim proves far too much and the law doesn't work like code...judges don't just rubber stamp obvious attempts to exploit the rules and instead apply common sense.

I mean suppose I want to infringe on google's patents for their search engine. On your theory I could simply spin up a google cloud account to host my source and then claim I'm immune from search. If taken seriously this would force the end of virtually every major company offering cloud computing options.

Patent exhaustion and open source

pabs — Sun, 31 Mar 2019 05:24:06 +0000

Another article on the topic:

https://blog.hansenpartnership.com/a-roadmap-for-eliminat...

Patent exhaustion and open source

Wol — Thu, 28 Feb 2019 15:07:04 +0000

> This means that farmers often prefer buying new hybrid seeds with desirable traits (like, high yield or resistance to pesticides) to keeping back part of their previous harvest for re-sowing, which would lead to a worse outcome. IOW, farmers tend to not want to infringe on patents on the seeds because it wouldn't help them commercially.

Those that have bought the agri-business kool-aid. I remember someone on Groklaw (that dates it) saying that his seed, that his family had kept and re-sown for generations, yielded maybe 20% more than his neighbours buying in high-yield seed. Thing is, the commercial seed does well in the conditions it's bred. It doesn't necessarily do well in different conditions elsewhere, while seed that's been kept and re-sown adapts to the local conditions.

Cheers,
Wol

Patent exhaustion and open source

derobert — Tue, 26 Feb 2019 22:37:05 +0000

More importantly, how many open source exFAT implementations shipped by Android phone vendors who have Microsoft patent licenses; those are a much stronger argument.

Patent exhaustion and open source

biergaizi — Sat, 23 Feb 2019 03:29:51 +0000

"XOR cursor patent"... Thanks for your hilarious example, it is the joke of the day ;-)

Patent exhaustion and open source

NAR — Fri, 22 Feb 2019 23:54:06 +0000

I guess Microsoft owning GitHub doesn't necessarily means GitHub *is* Microsoft and that Microsoft's patents are licensed to GitHub. Also, downloading from GitHub doesn't necessarily means that it's GitHub that provides the software - it's the user who pushed the code to GitHub. I mean if I park my car at a parking house and my wife picks up the car from there, it won't be the parking house that gives the car to her - it's me.

Patent exhaustion and open source

hodgesrm — Fri, 22 Feb 2019 15:40:24 +0000

Legal documents and code are not that different. Good contracts for example are designed to minimize edge cases, just as good code is.

Patent exhaustion and open source

anselm — Thu, 21 Feb 2019 13:12:01 +0000

Also, modern commercially-interesting cultivars of many food crops and similar plants are often hybrids that do not “breed true”, i.e., they don't reliably propagate the traits that make them commercially interesting to their progeny. This is a consequence of how genetics works.

This means that farmers often prefer buying new hybrid seeds with desirable traits (like, high yield or resistance to pesticides) to keeping back part of their previous harvest for re-sowing, which would lead to a worse outcome. IOW, farmers tend to not want to infringe on patents on the seeds because it wouldn't help them commercially.

Patent exhaustion and open source

bkuhn — Wed, 20 Feb 2019 16:32:39 +0000

>> That may well be true, but I fear I missed the straw poll whereby it was established.

There was a lot of hallway discussion afterward. Also, we discussed it some at the speakers' dinner for the track. There was a very lively debate there. :)

I realize there was none of this available to you; next time someone from LWN is at our Legal & Policy DevRoom, please do let us know. I can't promise the press will be invited to the speakers' dinner, but I would have made the case that you should be invited to the DevRoom co-organizers.

(I co-organize the DevRoom where this talk occurred, along with Tom Marble, Karen Sandler and Richard Fontana.)

Patent exhaustion and open source

nowster — Wed, 20 Feb 2019 14:17:33 +0000

And just how many implementations of exFAT are there on Github?

Patent exhaustion and open source

Cyberax — Wed, 20 Feb 2019 10:10:43 +0000

> The Q&A period was short. The only probing legal questions that were asked related to design patents (Lindberg replied that he'd not done the analysis) and to Bowman v. Monsanto. That case was about patented seeds, and whether the creation of new seeds was covered by patent exhaustion; the Bowman court held that the patent right to make a new copy wasn't exhausted by the sale of the seeds.
Seeds (and animal breeds) have special protection in the patent law. It recognizes that self-replicating entities need to be treated differently from regular dumb objects.

Patent exhaustion and open source

madhatter — Wed, 20 Feb 2019 07:44:35 +0000

> As for Van's talk, definitely listen to the Q&A.

In fairness, I should note that as soon as Lindberg read the article, he asked for a short note about the Q&A to be added. I declined, so the article as published rather skates over the Q&A, as you note; if I'd added a para about it, I would have said something like this (and I'm grateful to Lindberg for certain elements of this summary):

The Q&A period was short. The only probing legal questions that were asked related to design patents (Lindberg replied that he'd not done the analysis) and to Bowman v. Monsanto. That case was about patented seeds, and whether the creation of new seeds was covered by patent exhaustion; the Bowman court held that the patent right to make a new copy wasn't exhausted by the sale of the seeds. Lindberg argued that the difference was that that the Monsanto licence agreement forbade anyone from replanting seeds, where FOSS licences grant the rights to copy and make.

> a lot of the folks in the room felt that his argument might be sophistry

That may well be true, but I fear I missed the straw poll whereby it was established.

> I suspect his exhaustion theory would need to be adjudicated before we could rely on it heavily in FOSS.

That, I would agree with.

Patent exhaustion and open source

rsidd — Wed, 20 Feb 2019 04:59:32 +0000

If I understand right, this is not about a "non-practicing entity". It is about resale/reuse of an already patent-licensed item. If a patent troll holds the XOR cursor patent, and it's valid, you can't implement the XOR cursor without paying them. But if they have *licensed* the patent to X.org for their codebase, and you then use the X.org code in your own product, they can't sue you. Cf the AOSP example .

Patent exhaustion and open source

sfeam — Wed, 20 Feb 2019 04:44:55 +0000

I don't know whether the argument works at all, but if it works then it works against companies that license the patents and sell products that depend on them. The ownership of the patent does not matter at that point. So no, offloading the patent to a 3rd party would not change things.

Patent exhaustion and open source

magfr — Wed, 20 Feb 2019 04:43:27 +0000

Oh yes, IANAL, so anything I say are just ramblings.
With that said he might have an opening to curb NPEs in there.

Patent exhaustion and open source

magfr — Wed, 20 Feb 2019 04:38:21 +0000

While this is a nice hack the usual protections of having patents belong to an non- practicing entity would still work, i.e. this sadly won't work against patent trolls, only against honest companies.

Patent exhaustion and open source

bkuhn — Wed, 20 Feb 2019 03:43:55 +0000

> Hmm, so MSFT joined OIN because after buying GitHub, their portfolio was basically already pretty much exhausted towards FLOSS anyhow? ;-)

I note the wink, but others might not realize that OIN membership doesn't do much for FLOSS generally as the Linux system definition is quite narrow and OIN is (not surprisingly) not open to including key FLOSS projects that implement technologies where OIN's members and funders have heavy patenting.

As for Van's talk, definitely listen to the Q&A. Van is a very engaging speaker but a lot of the folks in the room felt that his argument might be sophistry. I really hope Van's arguments in this particular talk are right, but they appear to be pretty radical views among patent attorneys. I suspect his exhaustion theory would need to be adjudicated before we could rely on it heavily in FOSS.

Patent exhaustion and open source

KaiRo — Wed, 20 Feb 2019 01:21:39 +0000

Hmm, so MSFT joined OIN because after buying GitHub, their portfolio was basically already pretty much exhausted towards FLOSS anyhow? ;-)

Patent exhaustion and open source

dskoll — Tue, 19 Feb 2019 19:49:29 +0000

Oh, that is amazing! It's not just computer software that has surprising edge-cases... the law does too, apparently.

Patent exhaustion and open source

brouhaha — Tue, 19 Feb 2019 17:43:08 +0000

[me giggling then applauding]
That's awesome! Thanks you for writing this up, and thanks to Van Lindberg for giving the presentation!

I was somewhat familiar with patent exhaustion, but I definitely was not this well-informed about it.