Stack Overflow's daily new-question volume has collapsed from roughly 200,000 at its 2017 peak to under 1,000 today, coinciding with the rise of ChatGPT.

47:58this slow decline. So in 2022 we're at about 115,000 questions per day and then slow decline. In 2024 we're at 40,000 questions per day. By 2025 it's only 15,000 questions per day and now less than a thousand questions per day. So, um, first first F in chat. Throw your Fs in chat for for Stack Overflow. Um, I I actually like I don't like seeing this because, uh, that means there are a lot

48:27of questions that people have that aren't getting indexed on the web, right? They're having a one-off session with ChatGpt. >> They solve their problem, but then when someone else comes along to solve that same problem, they have to pay token pricing, right? like they don't they don't get to benefit from a forum with all of these questions that have been asked. >> Well, that's Stack Overflow is trying to

48:48do that. They're they're they're they released like Stack Overflow for agents >> where like agents can post what they had, but it doesn't seem to be catching on. At least not yet. But you're right. It's just like, oh, you have to everybody solves the same problems over and over again. And is is that a problem? >> Yeah. I mean, to me that that seems like a step backwards. Um,

49:09>> yeah, >> because it's it's more compute, it's more powered, it's more resource, it's more potential time spent because an agent is spinning to try and solve that same problem. >> Yeah. >> So, what you mention, I hadn't heard of what you mentioned, but that could be cool where where agents are putting their their questions and answers on a on a forum. >> Um, >> what if we just generated all of that

49:29every single time we needed it and boiled the oceans because of it? like h >> you know what makes me think is weird about this is that so often I would post something to Stack Overflow and I would get >> why are you doing it this way? Um and and sometimes that's the that's frustrating but sometimes that's the best thing if you're debugging something with like a co-orker or or posting it on

49:55Stack Overflow. often the answer to your problem will be like you're you're coming apart approach you're approaching this wrong and I don't know that like the quick fix from a lot of these AI agents is often let's take a step back and and redo this often the fix is like all right well here here's how you can fix it in the like the hole that you're in you know we