an implied license that can't be terminated is better than a explicit license that can: https://news.ycombinator.com/item?id=15051717
The key point here is: all of that still applies regardless of what other contracts exist between the two parties - including the explicit patent grant. The PATENTS file can not modify or nullify anything except itself, either by its existence or by its subsequent revocation. No court would say that it does, because the implications for contract law in general go far beyond patent and copyright.
Disclaimer: IANAL. I'm just a guy who understands logic, and who has been involved with software patents from an engineer's perspective.
I think where you and DannyBee differ is with:
"permissions granted under LICENSE may not be unilaterally withdrawn except for reasons specified in that document itself"
where DannyBee is not taking the patent licence to be a permission that can be implicitly granted under LICENCE when there is a separate explicit PATENTS grant.
He states in the followup comment "No license can be implied if there is an explicit license." which I'm guessing you would say only applies if the explicit licence is within or allowed for by LICENCE.
Do you think that is that a fair representation of your points of view?
Now, why would it be any different if one of those grants is implicit? Again, that would be even crazier, etc. Allowing contracts to interfere with each other like that (in this case the copyright license being affected by a patent issue) would make contract law even more of a nightmare than it is already. I strongly believe courts would reject that not only for the sake of logic or justice but out of sheer self-preservation.
Of course, we're all speculating until this exact issue is tested in court. I just think the people assuming the same as DannyBee are leaping toward the least sane and least likely conclusion, and asking others to do the same.
The law is more about precedent than logic. Unless you can point to specific opinions from lawyers or judges on this, this sounds more like your opinion than any kind of proof.
At issue is that Facebook putting both the License file and the PATENTS file in the same repository seems to many people like an overt act that extinguishes any implicit patent grant in the same way the "Clear BSD" license does (although a difference there is a modified license file). It feels to me like like Facebook is trying to have it both ways -- seeming to be a good citizen in the free and open source community but at the same time creating a one-sided deal with extra conditions. If they wanted to be unambiguous, they could have used the Clear BSD license and then added the PATENTS file.
> I've seen plenty of real lawyers agree.
For reference, here is an opinion from an actual US Patent lawyer citing actual US legal judgement about implied patent licenses implying how an implicit grant in this case would be broader in scope than the explicit Facebook PATENTS license:
In 2005, Dan Ravicher explained that, in the USA, recipients of software under the GNU GPL version 2 receive an implicied patent grant, based on the following US case law.
De Forest Radio, 273 U.S. 236 (1927)
"No formal granting of a license is necessary in order to give it effect. Any language used by the owner of the patent, or any conduct on his part exhibited to another from which that other may properly infer that the owner consents to his use of the patent in making or using it, or selling it, upon which the other acts, constitutes a license.”"
Hewlett - Packard Co. v . Repeat - O-Type Stencil Mfg. Corp. , Inc., 123 F. 3d 1445 (Fed. Cir. 1997).
"Generally, when a seller sells a product without restriction, it in effect promises the purchaser that in exchange for the price paid, it will not interfere with the purchaser's full enjoyment of the product purchased. The buyer has an implied license under any patents of the seller that dominate the product or any uses of the product to which the parties might reasonably contemplate the product will be put."
Bottom Line Mgmt., Inc. v. Pan Man, Inc., 228 F. 3d 1352 (Fed. Cir. 2000)
"Unless the parties provide otherwise, the purchaser of a patented article has an implied license not only to use and sell it, but also to repair it to enable it to function properly. This implied license covers both the original purchaser of the article and all subsequent purchasers."
Still this does not address the issue of an explicit grant extinguishing an implicit grant. It seems like that may still need to be tested in court. The practical issue then is, why bet your company on an unknown like that when there are perfectly acceptable alternative UI libraries (even better ones technically like Mithril, Inferno, and so on)?
It provides weak support for your position, because most of the cases it cites are for things that clearly follow different rules and precedents than patents do. Pretty sloppy for an "actual US Patent lawyer" if you ask me. More importantly, "broader in scope" does not imply "exclusive of". The implicit license might indeed be broader and there's still no whiff of either extinguishing the other. As I've said repeatedly, when contracts (of which licenses are a subset) are written to supersede something else, they say so. There's no implicit nullification.