Hacker News Reader

Comment by Will_Do | Hacker News Reader

This seems to also have the BSD + Patents beloved by Facebook[0]. The virus is spreading.

[0]: https://github.com/facebook-atom/atom-ide-ui/blob/master/PAT...

Note: Please don't use the term "BSD + Patents" for the Facebook license. There's an actual Open Source license "BSD+Patent", which is based on the well-accepted Apache 2.0 patent grant: https://opensource.org/licenses/BSDplusPatent .

If by "spreading" you mean Facebook is publishing new open source projects using that license combination, yes, it's "spreading".

Atom-IDE is a Facebook project. It seems to be actually derived (in part) from their Nuclide project, which has a completely proprietary non-open source license.

Is it a Facebook project though? The announcement says in collaboration with Facebook. That's an unusual definition of the word collaborate.

If only there was some sort of history of the code... and the history of people that contributed to the code over time...

Something like this maybe? https://github.com/facebook-atom/atom-ide-ui/graphs/contribu...

I looked at top 12 people, none of them claim to work for GitHub, majority of them claim to work for Facebook.

If only there was some way to tell.

Okay, this should be the top comment here. My developer tools using extensions that give away patent protection in a one side license is unacceptable.

This is absolutely infuriating. Amazing how one file added to the repo can kill the project.

What can we reasonably do to stop this? This goes against the very notion of open source.

Either you stop using it, or you convince a federal that it's unenforceable or unconscionable.

Developers have a lot of power here. Encourage others not to use anything that's BSD + Patents. If, in 2 years time, Vue is more popular than React and Facebook perceives PATENTS is the reason why, it could well vanish.

This is a great demonstration of the danger of BSD + Patents. First it's Facebook using it for a great frontend framework. Now Github is using it for an IDE. It truly a threat to open source.

I wouldn't bet against Vue been more widely used worldwide than React, it's huge outside the anglosphere and pretty popular inside.

Im on a project thats replacing react because of this issue. I don't agree with that descision, but lawyers.

We're going with Inferno. Vue feels like a better Angular, not a better React. It's not even comparable to React. It's just too clunky.

One pleasant surprise in this Facebook purge has been the discovery of seamless-immutable. It's far easier to work with, especially if you're using Typescript, which makes Immutable an absolute pain.

Interesting, I'm on same stack Vue/TS had not seen seamless-immutable thanks!

No, it is not. If you sue people with software patent, I’m happy you lose that ability.

Yeah but the protection only applies to facebook. You can still sue github and everyone else. We can clearly conclude that facebook is selfishly doing this for their own benefit and not because of an altruistic reason like a world without patents.

So in reality what you're saying means: "If you sue facebook with any patent, I’m happy you lose that ability. If you sue any other company other than facebook, I’m happy you keep that ability."

And now github is using the same license too :)

I don't have any patents why should I give a fuck about people who want to steal money from facebook?

This covers all patents, including non-software patents, so you lose the ability to sue for non-software patents as well. Are you happy with that?

But that is not even the point -- when facebook invokes the nullification clauses in the patent grant, you're simply back to no patent grant, at all. How is having a patent grant not strictly better?

No, I'm not happy with that, but that is also not what's happening here. You're still welcome to sue Facebook if you want, but you'd lose all patent licensed to you in the PATENT file.

I'm also not happy with spreading FUD. Bringing up non-software patent is a strawman.

Don't use it. There exist a few Facebook fanboys but nobody I know likes Facebook.

What is the issue with a free patent grant (which is revoked under certain conditions)? Atom itself is licensed under MIT and doesn't give you a patent grant.

EDIT: I understand that it would be best to get a free patent grant, but most open-source projects have no patent grant at all. Why are we not complaining that Microsoft might hold patents for VSCode that they suddenly will start suing companies over?

There exist good licenses that extend a patent grant in a reasonable manner. Facebook went off and wrote their own version that is asymmetrical in their favor. Also, lawyers seem to agree that most OSS that don't include a patent grant are giving an implicit patent grant, and this explicit grant is worse for the developer than the implicit one (while being better for Facebook.)

Some lawyers argue that BSD/MIT includes an implicit patent grant, but this has never been tested in court. Facebook's lawyers obviously don't think it's clear that BSD includes an implicit patent grant, which is why they have an additional license. I don't see how "we hope that a judge would interpret the licence as an implied patent grant" is better than an actual patent grant (with some constraints).

Also, IANAL, but I never understood why the revocation of the explicit patent grant would cancel out the implied patent grant. If your argument is that "the BSD licence text implies a patent grant", shouldn't that be true regardless of any additional licenses?

It's not about whether or not the implicit grant exists, it's that the implicit grant doesn't go far enough for Facebook's liking.

What you refer to as "some constraints" is an abdication of your ability to defend your company against Facebook infringements for as long as you use software that includes this horrible clause.

IANAL, but I've read that when there's no explicit patent grant written, an implicit patent grant can be inferred, since MIT says you have the ability to use it (which would require a patent grant if covered by patents). If that is the case, no (explicit) patent grant would be better than this grant.

Reply on news.ycombinator.com