[lunarcave]

In the "choose a default search engine" page, it has a slightly amusing summary for each.

> Google

> Your personal data fuels its monopoly. Market-dominant due to anti-competitive and anti-consumer practices.

> Qwant

> Based in Europe. Uses Bing results. Sends tracking data to Microsoft.

> DuckDuckGo

> Privacy-focused. Relies on Bing results but never tracks or profiles you.

> Ecosia

> May plant trees for clicking ads. Relies on Bing and Google. Sends tracking data to Microsoft and Google.

> Microsoft Bing

> Collects extensive personal data. Privacy controls are buried and limited. Subjectively overwhelming UI.

> Kagi

> Privacy-focused. Customizable results without ads or tracking. Requires a paid account.

[firejake308]

Slightly amusing, perhaps, but accurate and concise? Definitely.

[hopelite]

I wish we could just add our own default search with a search string template like when the Internet was still alive.

That being said, I like using the slightly more obscure presearch.com and Swisscows.com, for what it’s worth.

[hdjrudni]

> I wish we could just add our own default search with a search string template like when the Internet was still alive.

Can't we? The %s thing works in Vivaldi. Worked in Chrome last time I checked.

[lpln3452]

Firefox still lets you do this.

You can add any URL as a custom search engine by providing a string template for the query.

It doesn't have to be a formal "search provider". Any URL that accepts a query string will work.

[a022311]

IMO the problem with Firefox is that custom search engines in Firefox can't use POST requests, even though it's supported. You may want to check Mycroft Project [1] out for that.

[1]: https://mycroftproject.com/

[int_19h]

The only major browser that I can think of that doesn't support custom search URLs, including making one the default, is Safari.

[nunobrito]

Please review your opinion about Qwant, the overwhelming majority of search results are produced internally and they are very clear about what isn't: https://betterweb.qwant.com/en/2023/09/18/web-indexing-where...

In Europe they are still IMHO the best option for an independent search engine.

[godelski]

The irony is it is a Chromium browser...

[keyle]

Imagine reading that list in 1995. Sigh.

[TiredOfLife]

> Kagi

should be changed to

> Openly and proudly collaborates with russian government

[klibertp]

Kagi uses Yandex to improve search results for relevant queries. That's all they do.

As a company providing the service of web search, Kagi should do whatever it takes to improve search results. I imagine Yandex is the biggest and most complete index of Russian-language content - not using it would make the search results worse. The fact that Kagi still cross-references other indexes and allows users to downgrade specific results provides a check on propaganda content.

It's OK to have an opinion, and it's OK to dislike Kagi because it doesn't have the same opinion. It's wrong to mischaracterize what Kagi does, using wording that strongly suggests actions way more nefarious than giving a few dollars to a Russian company in exchange for some (anonymized) API calls.

[n4bz0r]

What's the context? Is there a proof of sorts?

[TiredOfLife]

https://kagifeedback.org/d/5445-reconsider-yandex-integratio...

[n4bz0r]

There is nothing about collaborating with Russian government in the post. They merely make use of a Russian-based search engine to provide better search results. I won't argue if this is good or bad, but from your statement it sounds like they collect all the users' data and sell it directly to Federal Security Service.

[TiredOfLife]

Yandex is russian government

[n4bz0r]

That's as shallow as "<anything Russian> is Russian government". Russian government certainly established strong leverage over Yandex (and every other major business for that matter), but they don't exactly own the company.

Such blanket statements really don't bring anything to the table.

PS: I think you might be confusing Yandex with VK. VK are known to be loyal to government and provide users' data to law enforcement at a whim, without proper procedures.

[flexagoon]

Ok, so they say Yandex queries are 2% of their costs. Kagi currently has 57341 paying members. Even if you assume that every single user has a $25/month Ultimate membership and that 100% of that money goes towards search API (which is both obviously not the case), you'll get that Kagi pays about $29000/month to Yandex. According to World Bank, the corporate tax burden in Russia is about 46% of profit, so if you assume all of this money as profit, Kagi pays $13000/month to the Russian government. In reality, Kagi spends MUCH less than $25/user on search API costs, and Yandex doesn't claim all of that money as profit, so the real figure is closer to $1000-5000, maybe even less. And most of that money goes to local governments or pension funds anyway. So yeah, Kagi may pay likeabout 1-2 cents a month from each user's subscription to the Russian government in exchange to massively improving the quality of their service. That's not nearly enough to call it "proudly collaborating with the Russian government", and I can guarantee you that MUCH more of your money goes to Russia every month in other ways. In fact, if you live in Europe, you probably pay more to Russia in your own taxes.