Hacker News Reader

Comment by msoad | Hacker News Reader

I remember when the more tech savvy folks were migrating from Altavista and Yahoo to Google because they understood it is better sooner than others. The same thing is happening. I consider myself tech savvy (still!) and I have to admit that I rarely use Google for all sort of information.

As a side note: I am using Safari and I noticed that Apple's search is also replacing my Google searches. In the past if I knew name of a company or organization but not their website I'd Google it. Now I put it in the address bar and Safari very often finds the website for me.

View on HN

Of course the savvy are using LLMs now, but they’re also reckoning with two things:

* you have to check an LLM result, especially if it cites something (because it may or may not exist)

* you can’t cite an LLM result

It’s a useful tool, but it lacks certain utility features that a useful web + effective search has. Or had.

100%. But gradually being solved as source links are being provided. If the source link is as good as what you’d find on Google anyway then you have a way better search experience and don’t really burn much extra time checking the LLM work

Even without using LLMs, Duckduckgo is good at providing little widgets that just give you a straight answer. Try searching "150 usd to eur" or "Weather Philadelphia".

> Of course the savvy are using LLMs now

I don't think this is true at all. It is people prone to hype, or who are naive, who are using LLMs. The savvy know that a tool which you have to verify every single time (because it isn't deterministic and makes shit up) isn't actually saving you any effort.

The point that there’s effort involved in verification is an important one and definitely part of any thorough inventory of liabilities and advantages of LLMs.

I am less convinced this means it’s always a wash (or worse). Sometimes it’s good to have something to start with. “You can’t edit a blank page” is a truism among fiction and non-fiction writers of all stripes for a reason. Of course, the quality of what you start with matters, and it matters more the less of an evaluatory mindset you come to your tools with. I know at least some earlier adopters have that mindset, but perhaps I’m too optimistic about how it generalizes.

> * you have to check an LLM result, especially if it cites something (because it may or may not exist)

You think that you don't have the check the link at the top of the page from a google search?

In fact, last I checked, for every google search I do, I have literally hit the page down key before starting to read.

> * you can’t cite an LLM result

And you think you can cite the top search result from a google search?

> And you think you can cite the top search result from a google search?

I think the issue with LLM answers is that I often have to go to Google or something to check if the answer is right. Maybe it shaves some time off, but we're still back to the original problem.

Of course you can. The top search result from a Google search yeilds (eventually) a URL —a uniform resource locator— which is something that can be cited.

Apple's search is Google search. Google pays Apple $20 billion a year for that.

He means the top part above Google suggestions. If you type a url or company name Apple sometimes guesses the url and provides a direct link. Google not involved there - only if you click a Google suggestion or hit search/enter or whatever

Reply on news.ycombinator.com