Live data from Hacker News

Scrapers Take Down GitHub: December 11 Outage Timeline

statusgator.com

1–8 of 8 posts

Re: Scrapers Take Down GitHub: December 11 Outage Timeline

#4
post #3

They may not need to go so far as this to protect their service, but I would accept that any access required occasional authentication. Unfettered, unlimited anonymous access seems dubious in this “I’m taking all your content” AI training world.

It would also be, in my opinion, very against the spirit of Open Source to require an account to download code for a piece of software.

Re: Scrapers Take Down GitHub: December 11 Outage Timeline

#5
post #3

They may not need to go so far as this to protect their service, but I would accept that any access required occasional authentication. Unfettered, unlimited anonymous access seems dubious in this “I’m taking all your content” AI training world.

It would also be, in my opinion, very against the spirit of Open Source to require an account to download code for a piece of software.

As if Microsoft/GitHub (closed source platform) care about open source or free software…

Re: Scrapers Take Down GitHub: December 11 Outage Timeline

#6
post #3

They may not need to go so far as this to protect their service, but I would accept that any access required occasional authentication. Unfettered, unlimited anonymous access seems dubious in this “I’m taking all your content” AI training world.

It would also be, in my opinion, very against the spirit of Open Source to require an account to download code for a piece of software.

The spirit of open source doesn't account for garbage scrappers using mass botnets.

Re: Scrapers Take Down GitHub: December 11 Outage Timeline

#7

GitHub published a postmortem for yesterday's outage already, and the cause surprised me: They blame scrapers, or rather ONE specific scraper, for taking down GitHub for 7 hours.

You are not supposed to download anything from Github. /s

Re: Scrapers Take Down GitHub: December 11 Outage Timeline

#8
post #3

They may not need to go so far as this to protect their service, but I would accept that any access required occasional authentication. Unfettered, unlimited anonymous access seems dubious in this “I’m taking all your content” AI training world.

It would also be, in my opinion, very against the spirit of Open Source to require an account to download code for a piece of software.

But download all of them? There is a line here where it becomes absurd to me.