Working Code Podcast - Episode 201: LLM's vs Stack Overflow
After a much enjoyed winter break, we at the Working Code podcast are back at our microphones for season two. And, to kick things off, we're talking about LLMs vs. Stack Overflow. Large Language Model (LLM)-based Artificial Intelligence (AI) is currently at the center of many tech conversations. Put bluntly, it is the thing that we're all paying attention to. But, human attention is finite; and, as members of the Stack Overflow community have reported, the emergence of LLMs has had a massive impact on the site's interest (with question-asking down some 70% in the last two years). Does this foretell an end of Stack Overflow? Or, is this merely a moment in which the pendulum is swinging a bit too far in a new direction?
I've never asked a question on Stack Overflow; but, over the years, I've landed on the site countless times via Google search. It's always been an invaluable resource. And, in the last few years, I've particularly enjoyed the evolutionary insights that the site provides.
Many Stack Overflow threads have comments that span more than a decade. The longitudinal aspects of this information give me a condensed view into how the web has changed, which APIs have come into existence, and how older techniques can be transposed onto a more modern technology landscape. In my experience, this kind of high fidelity information doesn't get surfaced in a ChatGPT session.
I get value out of ChatGPT. But, it doesn't even begin to approach the value-add of having a conversation with another person. At least not yet. It remains to be seen if things will continue to improve. In the meantime, I hope that Stack Overflow—and sites like it—continue to be a mainstay of the developer community.
For full show notes, see workingcode.dev.
Reader Comments
Post A Comment — ❤️ I'd Love To Hear From You! ❤️
Post a Comment →