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

Getting forked by Microsoft

philiplaine.com

Post reply on HN
Print view View original

481–490 of 1001 posts

Re: Getting forked by Microsoft

#481

Post by YetAnotherNick »

saulpw wrote: post #445
unsungNovelty wrote: post #348

Earlier quoted context omitted.

> My personal thought is that we need a new kind of license: community open source. No corporations, just community. You are going exactly against the OSS philosophy. OSS shouldn't restrict the use of software just because you don't like it. It was created to fight exactly this. This is also why source available BS (like BSL) is against OSS. OSS is literally about being about hacking and changing software to suit you…

The OSS philosophy was conceived to help end users, not for-profit corporations. Then for-profit corporations co-opted the "Open Source"(tm) label to ensure they could benefit from all this free labor. You and many others are falling for it, and doing their work for them by scolding OSS developers for "going against the OSS philosophy". So screw this corporate "OSS philosophy", and stop telling people what they "shou…

The point is what do creators want to get out of their open source project. If it is the opportunity to sell, they can make it source available. If they don't want money, having open source license is better as it could mean more contributions.

Re: Getting forked by Microsoft

#482

Post by palata »

vvillena wrote: post #457
palata wrote: post #405

Earlier quoted context omitted.

I am genuinely interested: everybody here says that they removed the copyright headers. But when I browse through the Spegel sources, they do not contain a copyright header... To me it's the Spegel author's fault: there should be a copyright header in every single file, such that Microsoft would have to keep it.

It's in the LICENSE file. With a MIT license, you assign a copyright to the project, or to a certain set of files. The Spegel license attributes copyright to "the Spegel authors", while Peerd attributes it to "Microsoft Corporation". If some of the peerd code was lifted from Spegel, it's blatant stealing. Code attribution is the only thing a MIT license asks people to honor, and Microsoft couldn't even do that.

> If some of the peerd code was lifted from Spegel, it's blatant stealing.

Could we say "it's incorrect attribution"?

> and Microsoft couldn't even do that.

Did you consider it may have been done by an engineer who, in good faith, thought they were giving proper credit by adding it to the README? Would you want that engineer fired because of the bad attribution?

It's not like Microsoft is making millions out of this. Sure, they should fix the attribution. It's a mistake.

Most startups/small companies I've seen rely heavily on open source and fail to honour every single licence. This is bad and nobody cares. Here, Microsoft mentioned the project in the README (which is not enough, but not nothing), and I'm pretty sure that they can fix it if someone opens an issue. But overall, companies like Microsoft do honour licences a lot better than startups in my experience.

BigTech is evil for many reasons, but maybe we could consider that this is just an honest mistake.

Re: Getting forked by Microsoft

#483

Post by RajT88 »

giancarlostoro wrote: post #118
vasco wrote: post #6

Earlier quoted context omitted.

If a megacorp wants your help to explain ANYTHING to them, you better be paid handsomely per hour. Wtf are people doing charity for trillion dollar empires.

It's also very possible they had been working on it already and wanted to compare notes, I certainly would if I were working on something internal and found a similar project, but I agree, ask them for a consultation fee. I don't see why they wouldn't pay it. Both projects also share in license, so I have less of an issue with it personally. They're both MIT licensed.

> I don't see why they wouldn't pay it.

I have seen plenty of dev managers refuse to pay for something if they didn't have to.

Re: Getting forked by Microsoft

#484

Post by ygjb »

palata wrote: post #453
MyOutfitIsVague wrote: post #434

Earlier quoted context omitted.

The details are less important. The code that is copied needs to be attributed, either with comments, or a license file that states which files came from the project, or something else, but the specific code does need to be recognizable by a reader as coming from that other source. Comments and copyright headers are the easiest way to do this.

Still, to me it's not even clear if "substantial parts of the code" were copied. What the article shows is really small snippets of pretty generic code. Ok, it keeps the original comment and the overall form. But if it's 15 lines, it may even count as "fair use", couldn't it? Remembering how LLMs use the concept of "fair-use" by stealing everything everywhere... My point is that Peerd seems like it's loosely based on…

> Still, to me it's not even clear if "substantial parts of the code" were copied. What the article shows is really small snippets of pretty generic code. Ok, it keeps the original comment and the overall form. But if it's 15 lines, it may even count as "fair use", couldn't it? Remembering how LLMs use the concept of "fair-use" by stealing everything everywhere...

Fair use allows for commentary, news reporting, criticism, teaching, research, and scholarship and there are guidelines. Most cases where fair use is sought as a defense requires litigation to clear it up. The other alternative when forking an extremely permissive MIT license is to just follow the license.

> Probably they should keep a copy of the original LICENSE file somewhere, sure. And if one asks politely, maybe they will do it.

They are required to do so by the original license of Spegel. Does Microsoft ask politely when people violate MS licensing by say, pirating their software, or do they work with 3 letter agencies and a massive enforcement team to ensure their licenses are followed?

> My point is that Peerd seems like it's loosely based on Spegel. Maybe a fork that was heavily modified. Not sure if they should track all the code that looks like it was not modified enough and attribute it everywhere.

Yes. Every other tech company I have worked at, including Mozilla, a company that publishes almost everything they do as open source, has had folks dedicated to ensuring license compliance.

> Again: they did credit the original project. So it feels a bit aggressive to say that they "stole it without giving any credit".

They didn't provide credit in the way that the license requires. This isn't a case where a new community member forked or copied code into their first open source project. This is one of the biggest companies in the world with a well-known history of taking and using OSS without proper attribution. I like and use many MS products, but they absolutely do not deserve the benefit of the doubt.

Re: Getting forked by Microsoft

#485

Post by evantbyrne »

palata wrote: post #415
evantbyrne wrote: post #403

Earlier quoted context omitted.

I understood and this is incorrect. > The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. It needs to appear somewhere regardless of where exactly the license was placed in the source repository.

Right. So they should just copy this licence somewhere in a subfolder, saying "parts of this project derive from Spegel, with licence: "? They can still do it now, and probably they should (someone can even open a PR?).

The MIT license does not seem to dictate the exact location of inclusion. Logically, I would think you would want to associate it with the specific parts of code that you are copying. In the past, I've listed licenses together in the root license file for forks, and other times when the included code was a minor part of the overall project placed forked licenses within impacted files.

Re: Getting forked by Microsoft

#486

Post by cestith »

krupan wrote: post #368
p_ing wrote: post #131

Earlier quoted context omitted.

Apple famously migrated away from bash (stuck on 3.2 in macOS 15) to zsh to avoid the GPLv3 'problem'.

There was zero chance of them having problems shipping bash and I'm glad you put problem in quotes.

They still ship bash. It’s just not the default shell anymore.

Re: Getting forked by Microsoft

#487

Post by tedivm »

palata wrote: post #398
robmccoll wrote: post #291

Earlier quoted context omitted.

You are mistaken. The license explicitly allows it subject to the terms of the license: > The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. Microsoft didn't follow these terms. They copied "substantial portions of the Software" and didn't include the notice.

Which notice? None of the *.go source files I have opened in Spegel contain a notice. Microsoft cannot remove a notice that does not exist... In my opinion, it's the Spegel author's fault: they should have added a notice in every single file!

The "notice" is the literal license file. It is illegal to strip someone else's license from their work. It doesn't matter that they replaced MIT with MIT, because they stripped the author and attribution from it.

https://github.com/Azure/peerd/blob/main/LICENSE

If you read that file you'd think that Microsoft was the copyright holder, but they very clearly aren't.

Re: Getting forked by Microsoft

#488

Post by stogot »

koiueo wrote: post #19

> I default to using the MIT license as it is simple and permissive What's good about being "permissive"? I keep hearing this argument, but I still don't understand, what's the incentive for authors of one-man projects to choose anything "permissive". Do you enjoy your project getting forked, walled off and exploited for profit by someone who has never done you any good? AGPLv3 still allows forking, still allows maki…

I think for me, I’ve been a beneficiary of using MIT licenses (in minor ways, no large or famous projects) and so when I publish code I prefer sharing as MIT.

Maybe I should reconsider, but I never thought anyone would remove an MIT license. That sounds like plagiarism (though they did put a thank you in their repo)

Re: Getting forked by Microsoft

#489

Post by Conan_Kudo »

palata wrote: post #426
ptx wrote: post #396

Earlier quoted context omitted.

It's in the license file: https://github.com/spegel-org/spegel/commits/main/LICENSE

Right. So Microsoft should just have a copy of this LICENSE file somewhere? Can't we just open a PR to add it to the repo? Did the author do that and did Microsoft decline the PR? Feels like Microsoft was not necessarily trying to steal work (they link the original project in their README).

It needs to be present in the headers of each file that they took from. Attribution matters and in mixed projects you need that clarification at the file level.

Re: Getting forked by Microsoft

#490

Post by golergka »

koiueo wrote: post #19

> I default to using the MIT license as it is simple and permissive What's good about being "permissive"? I keep hearing this argument, but I still don't understand, what's the incentive for authors of one-man projects to choose anything "permissive". Do you enjoy your project getting forked, walled off and exploited for profit by someone who has never done you any good? AGPLv3 still allows forking, still allows maki…

> exploited for profit by someone who has never done you any good

Yes, that's the whole point of open source? Most contributions to the most popular libraries and frameworks (not necessarily end products) are from employees on their paid corporate time to begin with.

Post reply on HN