Selfhosted

I 2nd this, Codeberg is basically an open source community funded GitHub clone

Thank you so much!! Let me know if you run into any issue, it's fairly stable for a beta, but I'm sure there's a ton of quirks that still need sorting ❤

SearX has been dead for a long time. SearXNG i s brilliant and has no problems. If you're having problems, then it's your server getting blocked. Not a problem with the application.

4get.ca

Oh that's quite nice actually! Never seen it! Suppose it's very minimal tho, I'll do some market research to scope the competition lol I feel like more alternatives make a better internet, if you ask me :)

p.s. thank you! sorted out the readme, my bad

I would totally use a search engine called dogdog

Ow that's a big compliment, thank you! I think it's just different rather than better as it focuses on user experience more than niche features. It's meant to be an easy way to move out of using Google daily, without too much of a learning curve

Glad you are enjoying it ♥️

Yaaas! Let me know how you get on with it ♥️

Nice! Thank you! Let me know how you get on! :)

YES! Customisation is the main reason why I went out of my way to build it, and other stuff I listed in another comment in the thread :)

Thank you, if you ever end up making a store with your own plugins/themes and so on just lemme know and I'll add it to the repo, it'd be cool to have a list of available stores already in the beta version, so when I go out of beta it'll feel like a much more mature product <3

I keep getting results in different languages, a lot of my engines get rejected, it's quite slow at times and building extensions for it is fairly complex. They also recently restructured their own codebase, which means most of my customisation didn't work anymore, which is when I decided to build my own instead of re-making all the customisations from scratch.

This is again very personal issues and specific to my use case frankly, and it could be that I wasn't doing customisations right, but the way my brain works is to make something from scratch and learn the ins and out of it, so I just went for it.

I guess it also is styled more to my linking (plus theming is very extensive) and I wanted to create something that slots things in if wanted (see tmdb above results and rss on the homepage)

I found it oddly good at finding hidden pages containing the exact terms I was looking for but horrible at more general stuff

Echoing what @thagoat@lemmy.dbzer0.com said, I selfhost it and I only have one user...me.

turn on the limiter and link token, and that’ll stop in SearXNG. this is why most instances don’t work, is they don’t turn on the existing anti-bot measures.

https://docs.searxng.org/src/searx.botdetection.html

Hey! Well, aside from rotating multiple user agents, have a cheeky retry and having a "retry" button next to each engine I also heavily cache results so that for 12 hours the same query will give you the same results from cache. You can always invalidate cache from the settings.

Another thing you can also do is add multiple proxy that rotate on each request, traffic goes through proxy urls and you don't get hit with rate limiting (again, proxy settings are in the settings tab - you can find free ones around).

Searches happen asyncronously btw, which means if one engine hangs and fails the others will still return results quickly enough that clicking on the retry button shouldn't be that big of an issue. The biggest engine I have had issues with has been Ecosia, always hitting that pesky cloudflare anti bot page, but that's a third party extension engine, so it doesn't matter too much

So far I haven't been rate limited by google yet, not gonna claim it won't happen as it probably will, it's the nature of the tool, but finger crossed so far so good.

I host mine at home and only available on my tailnet. Haven't had any rate limiting since.

lol it's a play on the cron syntax, the * is a o in the name/logo