6 ms·
That's such a vague and novel area, that I don't think lawyers will recommend/allow commenting on it unless required by court of law. There is no precedent on
by blackoil 4y ago
That's such a vague and novel area, that I don't think lawyers will recommend/allow commenting on it unless required by court of law.
There is no precedent on if a computer reading your code or looking at the image is fair-use or not.
- 2Gkashmiri 4y agowhy not train it on copyright material also? whats the prima-facie reason for not doing so? i mean if you are doing "all public repos", why not everything else?
- ziml77 4y agoThey did train it on copyrighted work. The GPL is a license for copyrighted work. If it wasn't copyrighted, a license would be useless. The code being covered by copyright and the code being publicly accessible are two different things.
- 2Gkashmiri 4y agoyou know what i mean... i am talking about training it on windows OS code and adobe photoshop source code and other "proprietary software" code
- tpmoney 4y agoBecause scraping a public website is different from breaking into adobes private source control servers?
- 2Gkashmiri 4y agothere is public DB of windows code out there, if "cant" is the word then let users submit the injest code and let it train on it. if "wont" is the word then who gave them the permission?
- ShamelessC 4y agoThe reason is that GitHub, being GitHub, has a lot of search indices and existing analytics on the code that is hosted on GitHub. It was easier, at least - but probably gives some nice guarantees about the statistics of "public" code, with the norms and conventions you're "used to", because you're used to the internet's coding norms. Company code can be pretty bad, even at Microsoft, often riddled with hyper-verbose variable names and strange design patterns.
- blackoil 4y agoOfficially, Github does not have access to those repos. It manages them doesn't mean all team have access to them.