924
submitted 1 month ago* (last edited 1 month ago) by [email protected] to c/[email protected]

Update: https://www.bleepingcomputer.com/news/microsoft/microsoft-outage-affects-bing-copilot-duckduckgo-and-chatgpt-internet-search/

It's also important to note that ChatGPT internet search and DuckDuckGo are experiencing similar issues because they use the Bing API.

UPDATE 2

20240523_210619

you are viewing a single comment's thread
view the rest of the comments
[-] [email protected] 77 points 1 month ago* (last edited 1 month ago)

Okay.

Now that we already know we are pretty much at the hands of one pupeteer, what options are there?

I already read about Kagi (apologies if mispelled) but I like to write as a hobby and 300 searches per month go fast.

What other options are there?

Edit:

For those who may be arriving now:

  • Kagi seems to be a good option for an alternative search engine; it is a paid service, for which I don't have the €€€ right now. Many speak very well about it.
  • SearXNG is a thing as well, to my understanding a decentralized search system. Worth the try, in my opinion. If it's something that is decentralized, it is worthy to support and divulge.
  • There is mojeek.com, supposedly not very good but any option that goes against the monopoly is worth the try! I'm going to try this one.
  • Brave.com is an option but is a bit shady.
  • You should try Ecosia if you want to support reforestation efforts. Read somewhere in the thread it is part of the Bing ecosystem.
  • Yep is a thing as well. Somewhere in the thread, a lemmy points they use the search results for AI trainning. So... That is that.
  • And it seems there is a search engine by the name of dogpile.com.
  • Startpage is another search engine (portal?) suggested by another user. I've used it before and like it. Read somewhere it somehow piped a standard google search but removed tracking and ads.
  • and I just remembered Presearch.com. This is a really shady one (crypto warning!) that I suspect is a fork/collab with Brave Browser. I've used it, they have reward-per-search reward system (or had) where they give you crypto for every search. Good results, some that don't come up neither on DDG nor Google.

p.p.s Should I start categorizing these from "shady" to "worthy"?

p.p.s 2 Does anyone remember StumbleUpon? I know it was never a search engine to begin with but it was the best source of good internet content I ever got acquainted to.

Can we get something like that back?

[-] [email protected] 33 points 1 month ago
[-] [email protected] 12 points 1 month ago

This is the single most obnoxious name I've ever seen. But, the service could be good (I've never used it)

[-] [email protected] 9 points 1 month ago

Basically the mother of all search engines. Merges Google, DuckDuckGo, Bing, etc. into one.

[-] [email protected] 8 points 1 month ago* (last edited 1 month ago)

it's a fork (Next Generation) of deprecated SearX project.

[-] LambdaRX 15 points 1 month ago

Mojeek is far from perfect, but also is 100% independent.

[-] [email protected] 3 points 1 month ago

Thank you I tried Mojeek a few days ago but I forgot what it was called. the site for mojeek is mojeek.com btw.

[-] [email protected] 2 points 1 month ago

Sometimes it's a hard one to remember 😅

[-] [email protected] 9 points 1 month ago

That’s a nice hobby

I would suggest you to install a local instance of a LLM (mistral or llama3 for example) to widen your source of information. Go straight to Wikipedia instead of “googling” it if you don’t already.

Anyway, I didn’t know about kagi so I might take my own advice and give it a try.

[-] [email protected] 8 points 1 month ago

How big are they, and what do I need to use them well?

[-] vale 4 points 1 month ago* (last edited 1 month ago)

Take a look at Ollama.ai, just follow the installation instructions. A decent GPU is recommended, and the models are around 10GB iirc.

[-] [email protected] 2 points 1 month ago* (last edited 1 month ago)

Most of 7b-8b models run just fine in 4bits quant and won’t use more than 4 or 5 GB of VRAM.

The only important metric is the amount of VRAM as the model must be loaded in VRAM for fast inference.

You could use CPU and RAM but it is really painfully slow.

If you got an Apple Silicon Mac it could be even simpler.

[-] [email protected] 2 points 1 month ago

I have an Intel Celeron Mobile laptop with iGPU and, I think, 256MB VRAM. How many bs does that get me for the LLM?

~~Only half-joking. That's my still functional old daily driver now serving as homelab~~

[-] [email protected] 2 points 1 month ago

Well, I got a good news and a bad news.

The bad news is you won't do shit with that my dear friend.

The good news is that you won't need it because the duck is back.

[-] [email protected] 5 points 1 month ago

There's also mojeek.com running their own index. Not perfect, but sometimes usable.

[-] [email protected] 2 points 1 month ago

you can also send us in queries where we're not perfect, or let us know elsewhere, and we're keen on fixing them

[-] [email protected] 1 points 1 month ago

That was unexpected, but welcome message. Thanks for caring.

[-] [email protected] 4 points 1 month ago

There's a new search engine called Yep, made by the team at Ahrefs, a SEO tool SaaS.

It looks promising because they have their own index, but it's a bit slow sometimes.

[-] [email protected] 13 points 1 month ago

We may collect aggregated, non-personal search data to improve search algorithms, train AI models

ugh

[-] [email protected] 4 points 1 month ago* (last edited 1 month ago)

I only tried one example, so the sample size is pretty small, but that search engine seems pretty bad. I tried looking up "rust bevy points" in both Google and Yep. The first Google result is a library to draw points in Bevy and the rest are pretty relevant. Yep simply doesn't have that result at all and all of their results are just generic results about Bevy.

I tried DDG for the sake of comparison and it's somewhere in-between. The results are mostly relevant and the "correct" result is still on the first page.

[-] [email protected] 3 points 1 month ago

I jumped back on ecosia. It's nice there, but I'm sure there are issues that a non tech person like me may not know about

load more comments (2 replies)
[-] [email protected] 2 points 1 month ago
[-] [email protected] 2 points 1 month ago

You're welcome.

[-] [email protected] 2 points 1 month ago

Ecosia also uses Bing and was also down. That's the one I use.

[-] Evehn 2 points 1 month ago

Just letting you know kagi has a family plan with unlimited searches, so you can probably split it with family or friends! I have yet to see how searxng search holds up to kagi.

[-] [email protected] 1 points 1 month ago

I run a searxng instance on my pi server, use it from all of my devices, honestly forget its a thing running from my office 99% of the time until I'm rebooting the pi for one reason or another and spend more seconds than I'd like to admit wondering why I can't run a search 🤣😅

load more comments (35 replies)
this post was submitted on 23 May 2024
924 points (97.3% liked)

Technology

55690 readers
2732 users here now

This is a most excellent place for technology news and articles.


Our Rules


  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each another!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots


founded 1 year ago
MODERATORS