Skip to main content
10 events
when toggle format what by license comment
Feb 7, 2025 at 15:29 comment added NoDataDumpNoContribution "I can only see further decline of participation and traffic as the result." To be honest, I see that also without this experiment.
Feb 6, 2025 at 10:05 comment added zovits It feels this reply has more insight and thought in it than the AI experiment.
Feb 5, 2025 at 18:55 comment added JoshuaZ Yeah, ok that's pretty compelling evidence that it is much larger than any other site in terms of activity.
Feb 5, 2025 at 18:32 comment added Resistance Is Futile @JoshuaZ For historical records you can find some information here meta.stackoverflow.com/q/333743 Stack Overflow used to be a lot busier place with over 7000 daily questions asked.
Feb 5, 2025 at 18:06 history edited starballMod CC BY-SA 4.0
deleted 10 characters in body
Feb 5, 2025 at 18:04 comment added starball Mod @JoshuaZ stackexchange.com/sites#questionsperday eyeballing, easily more questions per day than all the other sites combined.
Feb 5, 2025 at 17:48 comment added JoshuaZ A lot of this seems reasonable. The claim that Stack Overflow is the only site with major traffic is not obvious. Is there data supporting this?
Feb 5, 2025 at 16:52 comment added Resistance Is Futile @NotThatGuy This wouldn't be training data, but more verification of how good their AI is when applied.
Feb 5, 2025 at 15:55 comment added NotThatGuy "are AI companies willing to pay" - I imagine SE is paying an AI company (possibly having some in-house staff customising established tools), not the other way around, for increase volume. Or there's a partnership with some future profit goals in mind. I can't imagine AI companies would be this desperate for training data, when there's so much data available elsewhere (even if plenty of that is morally and/or legally questionable).
Feb 5, 2025 at 15:21 history answered Resistance Is Futile CC BY-SA 4.0