Search isn't that business since staying small means you won't be able to create a good index of the world. And you won't have enough resources for your browser.
https://help.kagi.com/kagi/search-details/search-sources.htm...
It also uses other indices along Google.
> All results from external indexes.
The above is something I see all the time when using Kagi.
Unsurprisingly it's Pintrest that annoys people the most.
I haven't used Kagi much, but I don't think I've ever seen a single Pinterest result from any other search engine, I barely even know what the thing is for.
Is this a US thing? A Kagi quirk specifically?
And to be honest, I don't see as much Pintrest crap as I used to. It may just be that it has fallen out of favour because of its terrible design. But it was a site that basically let you create mood boards from images that you found as your browsed.
The big problem with it was that it simply copied the image so if you were looking for anything, Pintrest came up with the images for a product but then had removed all the source so it just flooded search results with uncited references and you couldn't find anything useful.
Reverse image searching to find sites thst sold a prpduct became useless.
If search works well, the result you wanted is in the first spot on page #1. This is a bad outcome for Google.
The more follow up searches you run and the more trash you wade through, the more ads you see and the more money Google makes.
On the index side I agree, I think they are using other people indexes so far, I don't know if they are thinking about building one themselves.
Same for LLM, but I think that there the problem is even worse.
Here's a list for the curious:
>A look at search engines with their own indexes
https://seirdy.one/posts/2021/03/10/search-engines-with-own-...
Obviously they don't have the ancillary services (Maps etc), but for just searching, Kagi is far more likely to surface useful results instead of just the highest bidder. Compare a search like "us esta" for a clear demonstration.
First search result for "us esta" in Google is "https://esta.cbp.dhs.gov/esta", same as in kagi, is that your objectivite fail at coming up with a simple metric?
Does ublock or ddg etc block sponsored results?
I also have my own WTFs with Google, I even started collecting them: https://imgur.com/a/bgFax59
First one on Kagi is the official US government ESTA website.
Let others handle rarer queries.
I'm not entirely sure what "objectively superior" is even supposed to mean in the context of a search engine, or how this follows from having 50k users, but that ceases to be an even remotely plausible statement if you've ever attempted to get good non-English search results.