Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Microsoft forked MIT licensed repo and changed the copyright [fixed]

github.com

Post reply on HN
Print view View original

421–430 of 626 posts

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#421

Post by bayindirh »

tptacek wrote: post #404
bayindirh wrote: post #398

Earlier quoted context omitted.

I'm glad that my assumption proven to be wrong, by Microsoft itself, nonetheless. However, is being wrong is always equal to being a knee-jerk? Does the reputation of the company in question doesn't play a role here? Or, shall we be stateless, and evaluate every events without any prior experience? I don't think that holds a lot of water in real world, either.

Another clue that it's knee-jerkism is that the accusations make no sense. It's hard to see what Microsoft plausibly stood to gain by modifying an MIT license. I suspect a lot of the accusations here are being made by people who simply don't know what an MIT license means. For that matter: even with a restrictive license like the GPL, it's hard to make sense of this as a heist, especially since it's right there in th…

> It's hard to see what Microsoft plausibly stood to gain by modifying an MIT license.

Actually, not being able to find a plausible gain in five minutes doesn't automatically clear Microsoft (or any company) in my mind. From top of my head, I can list three technologies which I find suspicious in the long run: LSP, WSL, SecureBoot.

All in all, I just don't trust Microsoft, and think about the worst of their actions first. They're the only company (ORACLE being a firm second) which evoke this reaction for me, and this their own making over the years.

Just because I don't trust a company and assume the worst of them, and telling this openly makes me a knee-jerk person, so be it.

As I said, I'm happy to be proven wrong, but I refuse to be stateless, and look every action of this company with completely neutral eyes.

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#422

Post by Zababa »

phendrenad2 wrote: post #89
37ef_ced3 wrote: post #44

The larger issue is that anyone using GitHub is donating their work for re-use without attribution through Copilot. In return, you receive hosting from GitHub. The writing is on the wall. You MUST host your own code on a stand-alone website. Here is an example. This Go program (a compiler) generates and serves its own website: https://NN-512.com It runs on a Linode shared CPU cloud instance that costs $5 per month: h…

That "issue" is unrelated. You've just hijacked the thread to talk about it. Also the point you're making is controversial, and definitely isn't widely agreed on. As a human, I can read public code on Github, and use my internal neural network (brain) to regurgitate sections of code, and don't need to attribute anyone (who can say which codebase I'm recalling code from? I certainly can't). So a neural network doing t…

You're making a false equivalence between copilot and a human brain. Neural networks are programs, programs don't have the same rights as humans. Implying the opposite is even more controversial than saying that copilot should respect licenses.

Also, humans sometimes have limitations on what code they should have seen. It's common in the emulator community to forbid anyone that has seen proprietary code from working on black-box implementations to avoid legal issues.

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#423

Post by glandium »

Kye wrote: post #399
bmitc wrote: post #391

What's surprising to me about all the negative reactions Microsoft receives, especially on here, is that from my point of view, they are currently the most open out of all the big companies (Google, Facebook, Apple, etc). I cannot remember a time Apple has admitted to anything or made any effort to have anything that could be considered open, but they do no wrong in most people's eyes, where even a simple mistake by…

Releases of all open source things Apple uses, by OS: https://opensource.apple.com/releases/ These mostly link to repositories under one of their GitHub organizations: https://github.com/apple-oss-distributions/ Apple's open source projects: https://opensource.apple.com/projects/ Which mostly link to repositories under their main GitHub organization: https://github.com/orgs/apple/repositories

Note these are thrown-over-the-wall source dumps, with no valuable VCS history, nothing more recent than at least 2 releases ago, and sending a PR will likely only receive crickets (said as someone who recently found an easily patchable bug in dyld, contemplated the idea of opening a PR and was told by an ex-colleague who now works at Apple that my best shot would be through feedbackassistant.apple.com).

BTW, those repos don't take issues.

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#424

Post by cdcarter »

dapids wrote: post #401
jeffwilcox wrote: post #321

I lead the Microsoft Open Source Programs Office team. I'm sorry this happened. We have merged a pull request that restored the correct LICENSE file and copyright, and are in touch with the upstream author Leśny Rumcajs who emailed us this morning. We'll look to revert the entire commit that our bot made, too, since it updated the README with a boilerplate getting started guide. The bug was caused by a bot that was d…

Do you mind addressing the reason why an employee did exactly what you are claiming the bot did in error? https://github.com/microsoft/cups/commit/8100595a3a3a6d5c7d0... Again, this is a person, not a robot. How does this play into a "software bug"?

If you read the repo history carefully, you'll see a bot was responsible for rewriting the "LICENSE" file from an Apache license to the Microsoft (c) stamped MIT license. This human commit simply copied that same language to the file named "LICENSE.txt". It's unclear why they did that, but that human was not responsible for introducing the license text into the repo.

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#425

Post by will4274 »

dapids wrote: post #401
jeffwilcox wrote: post #321

I lead the Microsoft Open Source Programs Office team. I'm sorry this happened. We have merged a pull request that restored the correct LICENSE file and copyright, and are in touch with the upstream author Leśny Rumcajs who emailed us this morning. We'll look to revert the entire commit that our bot made, too, since it updated the README with a boilerplate getting started guide. The bug was caused by a bot that was d…

Do you mind addressing the reason why an employee did exactly what you are claiming the bot did in error? https://github.com/microsoft/cups/commit/8100595a3a3a6d5c7d0... Again, this is a person, not a robot. How does this play into a "software bug"?

? Instead of a software bug, it was a human error. Is it really surprising that with a company of Microsoft's size, some employees fuck up? Likely the employee was trying to say that the contributions in this fork that were not present in upstream are covered by the new license, but failed to do so properly (by leaving the original license intact and identifying precisely which files the new license applied to and which it didn't).

Courts will take a far more generous view than you are here. If Microsoft is not profiting from the change, and fixes it promptly when pointed out, the courts will shrug at any case - no harm, no foul. It's not even clear to me that it's illegal to have the wrong license on GitHub, assuming the shipping product does not violate the correct license. As nobody has pointed to any infringing Microsoft product... What are we talking about?

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#426

Post by tptacek »

bayindirh wrote: post #421
tptacek wrote: post #404

Earlier quoted context omitted.

Another clue that it's knee-jerkism is that the accusations make no sense. It's hard to see what Microsoft plausibly stood to gain by modifying an MIT license. I suspect a lot of the accusations here are being made by people who simply don't know what an MIT license means. For that matter: even with a restrictive license like the GPL, it's hard to make sense of this as a heist, especially since it's right there in th…

> It's hard to see what Microsoft plausibly stood to gain by modifying an MIT license. Actually, not being able to find a plausible gain in five minutes doesn't automatically clear Microsoft (or any company) in my mind. From top of my head, I can list three technologies which I find suspicious in the long run: LSP, WSL, SecureBoot. All in all, I just don't trust Microsoft , and think about the worst of their actions…

If you don't have anything well-considered to say about an event on HN, the best thing to do is not to say anything at all. Believe it or not, your personal distrust of a giant company isn't all that interesting to the rest of HN.

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#427

Post by yawaramin »

jeffwilcox wrote: post #321

I lead the Microsoft Open Source Programs Office team. I'm sorry this happened. We have merged a pull request that restored the correct LICENSE file and copyright, and are in touch with the upstream author Leśny Rumcajs who emailed us this morning. We'll look to revert the entire commit that our bot made, too, since it updated the README with a boilerplate getting started guide. The bug was caused by a bot that was d…

Thanks for explaining. Out of curiosity, does Microsoft have any external-facing GPL-licensed projects? Are there any restrictions to using (i.e. open sourcing something developed internally or forking something from outside MS) GPL-licensed projects? Specifically, would teams be able to get approval to fork GPL repos?

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#428

Post by Barrin92 »

bmitc wrote: post #391

What's surprising to me about all the negative reactions Microsoft receives, especially on here, is that from my point of view, they are currently the most open out of all the big companies (Google, Facebook, Apple, etc). I cannot remember a time Apple has admitted to anything or made any effort to have anything that could be considered open, but they do no wrong in most people's eyes, where even a simple mistake by…

what surprises me the most about Microsoft related hate is that, despite the fact that they had (still have pretty much) control over the OS market they never locked their system down or charged people in any capacity.

People take it for granted that you can just run what you want on your windows machine but they could have turned it into smartphone style locked down system. Imagine if every app developer in history had to fork 15-30% of their sales over to Microsoft just for building on Windows, they could have made a trillion bucks or something.

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#429

Post by throwawaymanbot »

jeffwilcox wrote: post #321

I lead the Microsoft Open Source Programs Office team. I'm sorry this happened. We have merged a pull request that restored the correct LICENSE file and copyright, and are in touch with the upstream author Leśny Rumcajs who emailed us this morning. We'll look to revert the entire commit that our bot made, too, since it updated the README with a boilerplate getting started guide. The bug was caused by a bot that was d…

No post body was provided.

Re: Microsoft forked MIT licensed repo and changed the copyright [fixed]

#430

Post by fxtentacle »

tptacek wrote: post #233
fxtentacle wrote: post #225

Earlier quoted context omitted.

In some cases, they changed other licenses to MIT. That would then allow them to train their AI models on this source code, or redistribute a closed-source monetized GPL rip-off. Team A) makes a mistake and changes the license. Team B) uses the fork under its new license and forgets to check the original branch's license. Two honest mistakes leading to a de-GPL-ed library.

Here, they're changed the name on an MIT license. What financial benefit do they stand to accrue from doing so? That's the question I'm asking.

I was talking not only about this specific instance of it happening, but Microsoft had similar mistakes throughout the past 6 months. Here's the one from CUPS, a Linux printing library:

https://github.com/microsoft/cups/commit/ad69bcc78bdea3fea3f...

It used to be Apache License, then it became "MIT License (c) Microsoft Corporation". Thanks to the attention that this thread got, it has now been fixed:

https://github.com/microsoft/cups/commit/3859d70160010c61fd7...

But that source code was online with the wrong license for more than 6 months. Imagine if you had hosted Windows source code with a misattributed MIT license for 6 months... They would also bring out the pitchforks ;) Or even worse: well-paid lawyers.

Post reply on HN