Ask HN: What is the license of GitHub's blog and The README project?

31 points by linkdd ↗ HN
I don't see a license on their articles, which makes me think they are simply copyrighted.

If so, is it legal to translate an article (and link to the original article) on another website?

29 comments

[ 3.7 ms ] story [ 77.8 ms ] thread
Why not ask Github?
[flagged]
I sent a mail an hour ago to the author of the article. There is no "answer I don't like", if I'm not allowed to do it, then so be it.

Please don't assume things when you don't know the person.

The author could be in a different timezone, so don't get overly disappointed unless you haven't heard back from them in a few days. :)
I consider "no response" to be a negative answer by default :)
Anyone offering free legal advice on HN today?
We share advices on many fields all the time. Legal advices are no exception.

Especially when it's this simple, "water is wet", "prefer postgresql over mongodb", "no you can't legally copy content without being authorised by the authors, even if translated".

Legal advice is an exception. In many jurisdictions, it's a regulated field (like health) and providing legal advice without holding the requisite license is illegal.

Offering your own opinion on the law is fine, but IMHO answering someone else's question with opinion on the law is probably the wrong side of the line.

Disclaimers don't help, either, the regulation exists whether you acknowledge being outwith the licensing or not.

Legal advice is usually more narrowly defined than just "what is the license of these files", which would usually count as legal information rather than legal advice. Discussing legal information isn't usually covered in the unauthorized practice of law rules.

So the beginning of OP's question wouldn't pose a problem.

The second part (asking about the translation etc) could pose a problem, but it's phrased generally so it may or may not.

Legal advice would be applying the legal information to a very precise and specific scenario. For example, translation and republication permission is not always dependent on permission from the copyright holder if a fair use exception applies in the relevant jurisdiction, and even with permission there could be other obstacles to various forms of copyright reuse like a relevant patent from a different rights holder or a previously signed contract promising not to do this. And these details, as well as the risks of various different legal and illegal choices, may depend on the jurisdiction.

We don't really have enough information to know if exactly what OP wants to do is legal, beyond an imprecise answer of "not generally" which as mere legal information is typically fine for non-lawyers to say even where legal practice is restricted.

Disclaimer: I am not a lawyer, and this is not legal advice - not because I am including this disclaimer but because it doesn't fit the definition of legal advice.

"Not a lawyer btw" disclaimers generally do help, if they include the necessary wording to change your advice to "seek professional advice" and make it clear that it is general information not applicable to the audience's specific situation.

Some discussion here - https://law.stackexchange.com/questions/683/usa-is-i-am-not-...

There has been a history of "legal advice" being applied too broadly in practice, so a great deal of caution should still be exercised - https://www.yalelawjournal.org/forum/the-overreach-of-limits...

(comment deleted)
On the bottom of the page it says "© 2023 GitHub, Inc.". I don't see anything else.
The actually relevant question is whether you'll be sued. And if it's a good translation, you plainly state that its a translation of the blog article and link to it the answer is obviously not.
I doubt you'd be sued at all, even if GitHub/Microsoft really didn't want you doing it. More likely they'd just send a cease and desist or a DMCA takedown request first. Path of least resistance.
Legal? Probably not. Would GitHub care? No. Would GitHub be happy their content is being translated and pointed back to the source? Yes.
Pretty sure that given the copilot situation (and LLMs in general I guess), microsofts stance on all things copyrighted is "if its publicly accessible then it is fair game to use, provided you claim it as original work and absolutely do not give credit to the source". Just roll a dice a few times when copying it, or something. /s
[flagged]
Did I miss an age or copyright is still different from patent?
I might mix both.

At the end of they day it should be about products and features and not code

> You really one of the people who want to keep writing stupid code?

No. Why? are you a stupid code advocate yourself? Why wouldnt you want to write good code?

> Besides that there is still a huge difference between someone (human/ai) looking at work and getting inspired/trained on vs. reproducing the same thing.

Those models that are reproducing the same thing was my whole point. Copilot, GPT-3, stable diffusion have all been shown to do it. Copilot excessively so.

> Despite that you also think that just writing some normal code and not a real novel algorithm should be patentable?

Not me, no. Why? Do you really think that? Why would you want to do such a thing!

> I don't even want to have software patents it's so diametral to sharing and progressing together.

Patents arent copyright.

Stupid code in sense of code which is only here for because you need to write it not because it's interesting.

There is no inherent value of knowing how exactly writing a angular form filter or how to retrieve an entity from the backend.

90% of my code (probably more) is not inherently Peace of art. It's code I needed to write to get something done.

Ai should be able to do this and it should be great that it can do this.

> Ai should be able to do this and it should be great that it can do this.

Sure, I agree, but not by ignoring laws. The ends do not justify the means.

It's just ridiculous, being able to completely ignore software license terms because "you" did not read the code, "you" got the code from a statistic model from those sources with licenses. It's license laundering.

Just don't make code online accessable if you don't want your code used.

The same thing as with images.

> Just don't make code online accessable

An AI model that correctly references copyright holders and the applicable licenses would be preferential to not having the code available, IMHO.

I dont see why you would be advocating for people not to share their code in the first place. That seems couterproductive, and against the spirit of open source in general.

> You really one of the people who want to keep writing stupid code?

I'm genuinely curious, do you think Copilot writes good code?

Hi there - I oversee The ReadME Project at GitHub. Our stories are copyrighted, but we are open to syndication/translation if the story is clearly credited with a link that drives readers to the original source.
Thank you. I ended up contacting the author who gave me permission quite quickly :)