Took me a while to realise I can just type search queries into ChatGPT. e.g. simply "london bridge history" or whatever into the chat and not only get a complete answer, but I can ask it follow-up questions. And it's also personalised for the kinds of responses I want, thanks to the custom instructions setting.

ChatGPT is my primary search engine now. (I just wish it would accept a URL query parameter so it could be launched straight from the browser address bar.)

Trying that example, I’d much prefer just going to the Wikipedia page on London Bridge than trying to guess what phrases ChatGPT will respond well to in order to elicit more info. It’s initial response for me didn’t even mention one of the most interesting facts that people lived and worked on the bridge.

YMMV. For my case on software development, I don't even look on stackoverflow anymore.

Just type the tech question, start refining into what is needed and get a snippet of code tailored for what is needed. What previously would take 30 to 60 minutes of research and testing is now less than a couple of minutes.

Fortunately it's not like StackOverflow has been used as training data for LLMs, right?

Well, yes. Point is, GPT-4 read the entire StackOverflow and then some, comprehended it, and now is a better interface to it, more specific and free of all the bullshit that's part of the regular web.

And I don't have to wade through Stack Overflow and see all the times mods and others have tried to or succeeded in closing down very useful questions.

I know there are a lot of google programmers out there, but was using search engines for programming ever a good idea? Don’t get me wrong, I’ll look up how to do absolutely simple things every day but I basically always look in the official documentation.

Which may be why I’ve been very underwhelmed by GPT so far. It’s not terrible at programming, and it’s certainly better than what I can find on Google, but it’s not better than simply looking up how things work. I’m really curious as to why it hasn’t put a more heavy weight on official documentation for its answers, they must’ve scraped that a long with all the other stuff, yet it’ll give you absolutely horrible suggestions when the real answer must be in its dataset. Maybe that would be weird for less common things, but it’s so terrible at JavaScript that it might even be able to write some of those StackOverflow answers if we’re being satirical, and the entire documentation for that would’ve been very easy to flag as important.

Yes there are and it's infuriating. Colleague of mine had problems with integrating some code into an app that was built on a newer version of a framework because "there aren't a lot of examples yet". One web search and I found the frameworks own migration guide detailing the exact differences that would need to be accounted for.

Glad you have time and patience to read documentation.

Such luxury is increasingly rare for software developers nowadays.

This is funny, because I find it much less cumbersome to type a few search terms into a search engine and explore the links it spits out.

It depends on the subject but search engines are on the decline. With so many fake website written by AI I can only see it get worse.

The most extreme I can think of is when I want to find when a show comes out and I have to read 10 paragraphs from 5 different sites to realize no one knows.

> The most extreme I can think of is when I want to find when a show comes out and I have to read 10 paragraphs from 5 different sites to realize no one knows.

I found that you can be pretty sure no one knows if it’s not already right on the results page. And if the displayed quote for a link on the results page is something like “wondering when show X is coming out?”, then it’s also a safe bet that clicking that link will be useless.

You learn those patterns fast, and then the search is fast as well.

I don't disagree but having to have a learning phase for patterns sounds a bit like people clinging to an old way of things.

It’s better to have a pattern than having no pattern with ChatGPT to tell when it’s hallucinating or not.

I wish MLs were more useful than search engines, but they have still a long way to go to replace them (if they ever do).

Google still thinks I want to click on the sites I haven't clicked on in a decade even though they are first results. Search engines have a long way to go to catch up to GPT

You mean like prompt engineering?

What you’re describing as “clinging to an old way of things” is how every single thing has been, ever, new or old.

I don't know why you come here and say something so obviously untrue.

> The most extreme I can think of is when I want to find when a show comes out

Yeah, I find that queries which can be answered in a sentence are the worst to find answers from search engines because all the results lengthen the response to an entire article, even when there isn't an answer.

agreed on that antipattern although fwiw chatgpt is unlikely to know the answer for questions like these either.

Agreed except ChatGPT (3.5 at least, haven't tried 4) is unable to provide primary sources for its results. At least when I tried, it just provided hallucinated urls

Try it. There's a world of difference.

In general or for this specific application (linking primary sources)?

In general. I don't know whether it's better at providing sources.

Bing Chat for me, when mostly searching IT technical or programming stuff sometimes gives junk urls, sometimes gives some real valuable urls.

i love gpt-4 and i find chatgpt useless. so there is a big difference

GPT4All is capable of providing sources. This seems more to be a legal defense mechanism by ChatGPT than a technical obstacle.

It rather created new hybrid search engines, like perplexity and phind.

True. Although the training is on a snapshot of websites, including q&a like stackoverflow. If these were replaced too, where are we heading? We'll have to wait and see. One concern would be centralization/ lack of options and diversity. Stackoverflow started rolling AI on its own, despite the controversial way it did (dismissing long time contributors); it might be correctly following the trend.

Personally I prefer stackoverflow and such, because I can see different answer including wrong or non-applicable ones which don't solve my exact problem.

One site doesn't need to exclude the other.

Both have their uses.

who would have thought that few years ago, just goes to show that a Giant like Google is also susceptible when they stop innovating. The real battle is going to be fought between these two as Google's business is majorly dependent on search ads.