- 1 hour
I totally recommend just clicking through random pages on there for a good 10 minutes or so. You’ll be surprised at how many random, cool, interesting things you find that you’d probably never see otherwise in a million years. (also an AMAZING place to find small blogs to add to your RSS reader of choice)
They feed a lot of this into their main search index if you pay for Kagi search too, so a lot of these sites will appear higher than they would on Google, Bing, DuckDuckGo, etc, if their content is relevant. Especially fediverse sources.
- brendansimms@lemmy.worldEnglish3 hours
Sounds interesting. I currently pay the 5$/month for the Kagi search engine and that works great so I’m inclined to trust them on their other ventures.
- FauxLiving@lemmy.worldEnglish3 hours
Their ‘Research’ AI is pretty decent too. I can get to the same ends eventually, but I can’t read and digest 25 different pages and then launch another 10 follow-up searches in 15 seconds. It’s summarizing and not simply inferencing so the hallucination rate is acceptably low and it also cites the sources so you can click a footnote to fact-check any important details.
I self-host most things, but I can’t self host a search engine. I much prefer paying monthly than having to fight the eternal battle against tracking and ads.
- boonhet@sopuli.xyzEnglish2 hours
You CAN self-host a search engine, look up searxng.
Yes it’s a metasearch engine, but so is Kagi for the most part, their own indexes don’t cover everything.
- FauxLiving@lemmy.worldEnglish21 minutes
searxng
I have this on my projects lists, I’ve seen it recommended in a few places.
If I’m understanding the purpose (privacy filtering for searches), Kagi is a service fitting a similar role. Kind of like Headscale vs Tailscale. Is that right?
- brendansimms@lemmy.worldEnglish2 hours
saving this comment for when i dive deeper down the rabbit hole later. at the moment i think im only self-hosting jellyfin and owntracks, and I have a couple llm models in LM studio but haven’t messed with them in a while.
- chunes@lemmy.worldEnglish3 hours
Thus including the type of people who screwed everything up in the first place
- teohhanhui@lemmy.worldEnglish10 hours
This Kagi?
Between the absolute blase attitude towards privacy, the 100% dedication to AI being the future of search, and the completely misguided use of the company’s limited funds, I honestly can’t see Kagi as something I could ever recommend to people.
- grandma@sh.itjust.worksEnglish3 hours
Use of funds is misguided sure, but all the AI stuff is completely optional and has never ever gotten in my way
- OhVenus_Baby@lemmy.mlEnglish4 hours
6 months to a year ago it was all people were recommending, Kagi and searxng. Maybe another paid one I forget the name. Now Kagi is bad?
- Ludicrous0251@piefed.zipEnglish3 hours
That article is 2 years old. In the last two years Kagi hasn’t collapsed on itself, it’s not overrun with AI, the world hasn’t ended. They’ve implemented Privacy Pass, extended their browser support to Linux, introduced SlopStop for reporting AI websites, and generally continued to improve their main product.
Kagi is a business, run by people, who make decisions to the best of their ability based on their understanding of what’s going to best serve their needs/priorities.
Like any other product, the owners are guaranteed to make decisions that are not aligned with a fraction of their prospective customers needs/views. That’s what it’s like trying to serve a broad market like “internet search users”. Some of those users are inevitably going to get fired up enough to write a 20,000 word opinion piece on the subject.
For any service, you have to choose if the value proposition makes sense for you and your needs. For me, the value of most free search services has gone down the drain, and the value of spending monthly for Kagi is better than having to think about/maintain a SearXNG instance. YMMV.
- OhVenus_Baby@lemmy.mlEnglish30 minutes
Makes sense yeah. I use free search engines but your right, they almost never give the result I’m looking for.
14 hoursSo, like a baby Yahoo! directory?
I wonder just how long it will stay relevant and how they determine if the content comes from a human. So far I’ve been accused of being a bot several times, clearly reliably detecting humans is beyond the capability of … humans.
- Feathercrown@lemmy.worldEnglish3 hours
It’s always funny rewatching iron man 2 or home alone and being like “oh look, goebbels and hitler!”
or like lemmy? kagi is also very pro activity pub. this feature is kinda old actually. give it a try! also give kagi search a try
- Ludicrous0251@piefed.zipEnglish14 hours
Like a small curated stumbleupon. I gave it a few clicks a little while back and as you’d expect theres a pretty wide range of good to junk content on there, but it all felt distinctly human.
Since none of the pages are ad ridden its hard to imagine the AI crazies wasting tokens on something they can’t really monetize.
- masterofn001@lemmy.caEnglish13 hours
Whereby they can detect anomalous usage of language. Moreover, using whereby and moreover seem to be key indicators that something is written by ai .
- tirateimas@lemmy.ptEnglish11 hours
This is actually pretty nice. I guess we will see similar projects popping up in the near future.
- ExLisper@lemmy.curiana.netEnglish10 hours
This sounds great. You still need a subscription to use it? I may be the thing that finally convinces me to get one.
- FauxLiving@lemmy.worldEnglish3 hours
I don’t think you need a subscription. I just tried from another browser where I’m not logged in and it works.
They have a small web ‘Lens’ in their search (you can limit search to specific categories), so you’d need to subscribe for that. Although, with an account, you get 100 searches/month for free.
They also have a Fediverse lens:
- vacuumflower@lemmy.sdf.orgEnglish14 hours
“Small web” is how usually Gopher, Gemini and such protocols and resources accessible via them are described.
I knew I was right to suspect Kagi, trying to hijack an established name from an important phenomenon is one of the most certain red flags.
In general hijacking of names is one of the dangers very specific to our modern era and not really a problem before the Internet.
And yes, they are doing just that.
- ImgurRefugee114@reddthat.comEnglish13 hours
They talk about the name in the initial announcement back in 2023 where they link to many blogs discussing the topic.
https://blog.kagi.com/small-web
The term is a bit broader than those protocols and Kagi is far from the first to use it; it certainly isn’t a “hijack” as if it was the name if another project or something. ‘Small Web’ isn’t them claiming to own the concept of the small web, or that it’s somehow only accessible through them… It’s just a feature; search, as a curated product they offer and maintain.
It’s just what they named the lens, because it’s a lens for the ‘small web’ as they defined it; like the other lenses. They aren’t hijacking the word ‘academia’ by having an academia lens…
https://help.kagi.com/kagi/features/lenses.html#default-lenses
Sure, maybe they could slap ‘Kagi’ in front of ‘Small Web’ just to be sure, but I doubt anyone will confuse the concept of noncommercial small websites with a paid service…
Also
In general hijacking of names is one of the dangers very specific to our modern era and not really a problem before the Internet.
Neat fact: you can trace the roots of trademark law back like 7000 years …
Because I thought you were obviously wrong about the 7000 years thing, here’s a history of trademarks by some guy named Olivier Pierre:
Since ancient times, merchants have been using signs or marks in trade to distinguish their products. Registrations came much later, in the 18th century with the establishment of Intellectual Property Offices.
[…]
The use of trademarks dates back thousands of years, however we can’t date their origins with precision. Some of the earliest forms of identification of marks date from Prehistory. For instance, the Lascaux cave paintings in France show bulls drawings with marks on them. Experts believe that people were using personal marks to claim ownership of livestock, long before literate societies. That was about 15.000 years ago.
The Egyptian masonry from some 6,000 years ago shows distinguishable quarry marks and stonecutters signs, to identify the source of the stone and the laborer who carried out the work to claim their wages. There were creative entrepreneurs who marketed their goods beyond their localities and sometimes over long distances. Wine amphorae marked with seals were found inside the Tomb of the pharaoh Tutankhamun who reigned between 1336 a.c. to 1327 a.c. over ancient Egypt.
I’ve gotten so used to think of trademarks as registered trademarks, but it makes sense that it has existed much longer in the literal sense. The earliest known law however dates back little more than 4000 years, and there’s nothing about trademarks there, so I think it’s fair to say trademark law is a lot more modern. :)
Sorry for being entirely off-topic.
I’ve seen “small web” used to describe personal sites and blogs in HTTP-land for a while, like how Kagi is using the term. I think it just so happens, due to the nature of protocols like Gopher and Gemini, all of the content there is very personal, so it gets described as small web.
- mbirth 🇬🇧@lemmy.mlEnglish11 hours
Yes, besides Gopher and Gemini, “small web” usually includes traditionally coded HTML+simple CSS+handmade/small JS websites. I.e. websites that are not using a big JS framework or huge amounts of CSS.
- j4yc33@piefed.socialEnglish14 hours
Also, monetizing the ability to find searches that are not AI slop is, for lack of better phrasing, fucking bullshit.
- Ludicrous0251@piefed.zipEnglish14 hours
Literally any free search engine could stop serving searches that are slop - Kagi isn’t stopping them.
It’s almost like “free” search engines have ulterior motives for how they prioritize search results.
- ImgurRefugee114@reddthat.comEnglish13 hours
Worth noting that when it was initially launched, it didn’t have such an emphasis on AI.
https://blog.kagi.com/small-web
It just so happens that noncommercial small sites tend to be slop-free, and there’s a big demand for that, so now that’s how they’re marketing it.
