If I wanted to use OpenAI, I would. If I wanted to use StackOverflow, I would. Now I just only get to use OpenAI no matter what.
This hellscape is forming way too fast.
This hellscape is forming way too fast.
It doesn't indicate that generative AI is going to be shoehorned into StackOverflow's websites. It would seem counterproductive, in fact, to do that, since the gist of this seems to be that StackOverflow provides a large wealth of organized, validated human-generated knowledge, which is exactly the sort of thing you want to train LLMs on. Feeding AI-generated data back into that would diminish the value of the data SO hosts for that purpose.
https://meta.stackexchange.com/questions/399440/testing-a-ne...
I specialize in Amazon Redshift.
I've written a lot of PDFs about Amazon Redshift - serious stuff, deep technical investigations and explanations, published along with the source code which produces the evidence which the PDF is based on - and when people asked questions where I'd written up the answer, I pointed them at the appropriate PDF.
After some months, I received a direct message, which looked to me to be a pro-forma, a standard message sent in this situation, from the staff that I was promoting my site and I should not do so. It was well written and polite.
That's fine - I have no problems with that, it's their web-site.
What I did not like, however, and what came over as slimey, was that the staff had also deleted every post I had made.
This was not mentioned, at all, in the well written and polite message, which then of course became disingenuous. If you're going to do something serious like that, you need to tell people, not let them discover it for themselves.
This was for all posts, where I'd explained something directly or pointed to a PDF - presumably it's a standard action SO take in this situation.
I deleted my account and left.
I'm not sure what "position of power" you could be referring to. Power to do what, with respect to what? OpenAI has useful tools that Stack Overflow wants to apply to its own use cases, and Stack Overflow has good data for training LLMs on. Seems like a straightforward alignment of incentives.
I assume stackoverflow's metrics (traffic, number of new questions and answers) are down by an amount they are not happy with, so they are eager to strike any deal before their ship sinks.
At least that's how I read the news piece. Personally, I'm as often on stackoverflow, as I've ever been, whereas my chatGPT usage is down to almost zero.
And even greater motivation to just cooperate with StackOverflow for mutual benefit, rather than engage in a ridiculous arms race with them.
> I assume stackoverflow's metrics (traffic, number of new questions and answers) are down by an amount they are not happy with, so they are eager to strike any deal before their ship sinks.
I'm not sure I'd understand the connection to this even if that were true. The value StackOverflow seems to be bringing to the table is specifically a large dataset of human-curated technical knowledge. Both parties in this arrangement would have strong interest in ensuring that StackOverflow continues to generate this data through its user-centric Q&A website. I'm not sure how a deal with OpenAI would prevent their "ship" from "sinking" if that were the situation they were in.
> Personally, I'm as often on stackoverflow, as I've ever been, whereas my chatGPT usage is down to almost zero.
Same here. ChatGPT is a nice novelty, but I haven't found all that much productive use for it. Most people I know who do use it regularly are using it for either correcting their spelling/grammar, or as a conversational-interface search engine, neither of which I find to be superior to proofreading my own writing or evaluating information from its original sources after doing a conventional search.
But there might be a value-add for StackOverflow in the latter case: finding specific answers to complex questions can be a hit-or-miss proposition, and ChatGPT might at least provide a more efficient way of finding the articles that answer your questions, if implemented properly.
Of course, implementing it properly would likely involve designing the LLM to track the sources of the data it's tokenizing, and present a 'bibliography' for each of its answers, rather than just blindly compositing data from all sources into single probability values.
> provide attribution to the Stack Overflow community within ChatGPT
...and that didn't seem important enough for OpenAI to bother to mention it on any of their media channels that I've seen.
so, who knows?
It feels like it's a whole lot of nothing to me, and exchange they're letting OpenAI having all of their Q/A data.
I doubt it will make any significant difference to S/O for most people; and anyone who thinks putting S/O links in a chatGPT response is going to drive traffic back to S/O is kiddddddddddding themselves.
On top of this, you could say the same about any disrupting technology.
There are very few novel questions and the ones that are there use outdated apis.
moderation there is done so poorly it has become a meme of sorts, so even if and when it improves, any improvement in perception will lag... and because users choose to use the site based on their perception of its value rather than its true value, it has sort of become a vicious cycle
The decision to close my question in spite of it having a clear technical difference made no sense at all. It honestly felt like a bot that just noticed that a lot of the content of the question was related to other questions-a bot without the ability to understand why the question is literally different.
Why is SO like this these days. Is it just because there is such a large history of content in the site, that it’s easy for people who don’t want to think to just mark questions closed?
The reason is that while using SO you generally reach similar errors and then read answers and try to make sense out of the problem you are having, that's fantastic, but being able to explicitly state your problem and make followup questions on it is even better.
Yesterday I had to engage with a project using Redux. It has been a while since I touched that technology so I went forward and gave a summary of it to ChatGPT asking if I was correct on my assumptions, from there onwards I made a couple more explanations, a couple questions and I was done. I think this ability to further prod with questions is too good of a feature to pass on.
I use OpenAI because StackOverflow answers are just the absolute wrong answer. A combination of gaslighting (you shouldn't be having this problem), dogmatic enforcement of good ideas that started as guidelines and problematic example code that should not be trusted. You are better of with a reddit thread or a blogpost and much better of with actual documentation. StackOverflow is the thing that causes the bugs and the tech debt in the first place.
At least now OpenAI's competition has a fighting chance, because their models won't be poisoned by SO