The worst I’ve got so far hasn’t been hallucinated “books”, but stuff like functions from a previous major version of the api mixed in.
I’m most of the time on the opposite side of the AI arguments, but I don’t think it’s unreasonable to use an LLM as a documentation search engine. The article itself also points out copilot’s usefulness for similar things, but seems the opinion lost the popular vote here.
Yeah, I guess that can happen. For me, it has saved much more time than it has wasted, but I’ve only used it on relatively popular libraries with stable apis, and don’t ask for complex things.
If you don’t know what the code does, you’re vibe coding. The point is to not waste time searching. Obviously you’re supposed to check the docs yourself, but that’s much less tedious and time consuming than finding it, if the docs are hard to navigate.
Unlike vibe coding, asking an LLM how to access some specific thing in a library when you’re not even sure what to look for is a legitimate use case.
From a local source: https://yle.fi/a/74-20162439
Last week we had a hornet crash, and now two helicopters in one.
I too have made the mistake of flipping an if statement. It was clearly supposed to happen to non-HP cartridges.
Most of the hate is justified IMO, but a couple weeks ago I died on the hill arguing that an LLM can be useful as a code documentation search engine. Once the train started, even a reply that thought software libraries contain books got upvotes.