this post was submitted on 22 Feb 2024
363 points (100.0% liked)
tails: A Place for Mastodon Posts
328 readers
1 users here now
A virtual community
Posts from Mastodon users, featured natively in a community, so you can view them without the need for them to be re-hosted or screenshoted, and reply to the original author and Mastodon respondents if you wish.
Has so far included content from Warsandpeas, Mr. Lovenstein, SMBC, Loading Artist, Low Quality Facts, nixCraft, ElleGray, and other interesting or provocative stuff I've random'd across on Mastodon.
Supported:
Comments & Upvotes
Unsupported:
Posts, Downvotes, & PD's Automod
founded 9 months ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Things like this do a good job of showing why regular “dumb” search engines will continue to be relevant for finding referenced knowledge.
IDK, he did get the answer, and he didn’t have to scroll through Amazon ads for pipes.
Yeah, good point. I guess I’d rather scroll instead of try to convince an AI why it should give me what I asked for, but that’s probably just because to me scrolling is easier than putting effort into constructing a sentence.
Modern Google searches do a great job of returning results for people who want to buy things, but not a great job for people who want to learn things.
I think my ideal solution would be to have a custom search engine that only searches against wiki style sites or other websites dedicated to hosting reference material.
Google scholar still works well, however, it’s not very accessible to people that don’t have postgraduate level abilities in the subject areas they are exploring.
Or… Hear me out…
…Uncensored language models
I mean I want that, but for some reason they always turn into 4chaners
Other than the fact that they don’t hallucinate
They most certainly do, what else is every rd result in a search?