Downvote me into oblivion but Kagi ain’t shit. It’s a glorified Google frontend. The author is right that the web is filled with AI generated articles and fake reviews and lists but Kagi is not immune to this enshittification.
I even tried the same query the author was bitching about.
Here is Kagi’s first two links for top 10 air purifiers. Notice how the first result is a BS website called top10.com and the second one is one of the “fake review” websites .
And here is Google’s. First result is Wirecutter, and this might be subjective but I trust Wirecutter reviews on most things.
Rest of the Google results are exactly what the author was mentioning. But Kagi was no different.
So $10/month to get the same shit? No thank you. I agree that Google turned to shit compared to what it was but it is still the best search engine out there. Now if the article was about privacy concerns then they would have a point. Which is what Kagi is all about anyway. So let’s stop the fucking act.
All the LLM-generated “top 10” listicles are grouped into one large block I can safely ignore. (I could hide them entirely but the visual grouping allows for easy mental filtering, so I haven’t bothered.) Your weird top10 fake site does not show up.
But yes, as the linked article says, Kagi is primarily a proxy for Google with some extra on top. This is, unfortunately, a feature as Google’s index still reigns supreme for general purpose search. It absolutely is bad and getting worse but sadly still the best you can get. Using only non-Google indices would just result in bad search results.
The Google-ness is somewhat mitigated by Kagi-exclusive features such as the LLM garbage grouping.
What Google also cannot do is highlighted in my screenshot: You can customise filtering and ranking.
The first search result is a Reddit thread with some decent discussion because I configured Kagi to prefer Reddit search results. In the case of household appliances, this doesn’t do a whole lot as I have not researched trusted/untrusted sources in this field yet but it’s very noticeable in fields like programming where I have manually ranked sites.
Kagi is not “all about” privacy. It’s a factor, sure but ultimately you still have to trust a U.S. company. Better than “trusting” a known abuser (Google, M$) but without an external audit, I wouldn’t put too much wight into this.
The index ain’t it either as it’s mostly Google though sometimes a bit better.
What really sets it apart is the features. Customised ranking aswell as blocking some sites outright (bye bye pinterest and userbenchmark) are immensely useful. So are filtering garbage results that Google still likes to return.
I searched both Cory Doctorow’s post and the linked 404media article in his post for “air purifier” and found nothing. What author are you referencing?
It’s not even comparable in quality. It’s like almost trolling to even suggest they are in the same league. If you don’t want to spend 10 dollars, fine, but maybe stop pretending that your instance is somehow the best quality search engine that exists… :)
Your argument clearly shows that you fail to see the benefits of doing it yourself. I get quality results from my local instance due to my persistence and work put in to adjust the settings necessary. I’ve balanced the privacy and functionality of the instance to fit my needs and it costs me nothing but a few minutes of my time each week to do so.
Kagi doing it for $10 a month sounds like they’re turning a neat profit off of you; and you’re refusing to accept that I have achieved levels of search competence that Kagi has without paying for Kagi or even using their free searches or service.
Whether or not it makes sense to you value-wise to pay or not pay for Kagi does not matter in this discussion. it only matters that none of the things Kagi can do that I find useful are things that cannot be done with SearXNG.
My understanding is that a locally hosted SearXNG instance doesn’t really give you any privacy, unless you “dilute” your searches by letting others do searches from your instance too.
The nice thing is that I can customize it however I like too; change weights, choose which engines to pull from always, or even from search to search; so I’m not getting cruft.
SearXNG always rearranges the crap most engines serve to the bottom without fail.
Dunno, I got entirely different results. Reddit, homeairguides, forbes, a bunch of listicles like consumerreports, wired, ny times, cnet and whatnot, and other websites.
I wouldn’t use “air purifier” as a metric, since it was already a big public story that surely any search engine that’s even half paying attention would have made sure the results for are good. Probably some other consumer good is better for an un-preannounced test run.
(Also I’m not sure that searching “top 10 air purifier” and complaining that you got a top result of top10.com/air-purifiers and that’s not what you wanted makes a ton of sense. FWIW, I did try “air purifier” just out of curiosity and saw a very clear result that DDG had the best results, Google second, and Kagi third.)
I repeated it for “good wireless router” and saw different results; for them, the outcomes were fairly similar with Kagi somewhat better in results (returning Wirecutter as the top result, and an obselete Stack Exchange answer as the 2nd, which okay it’s not right but I get where you’re coming from sir), and Google and DDG as secondary (returning PCMag and CNet at the top and Wirecutter only further down below).
Downvote me into oblivion but Kagi ain’t shit. It’s a glorified Google frontend. The author is right that the web is filled with AI generated articles and fake reviews and lists but Kagi is not immune to this enshittification.
I even tried the same query the author was bitching about. Here is Kagi’s first two links for top 10 air purifiers. Notice how the first result is a BS website called top10.com and the second one is one of the “fake review” websites .
And here is Google’s. First result is Wirecutter, and this might be subjective but I trust Wirecutter reviews on most things.
Rest of the Google results are exactly what the author was mentioning. But Kagi was no different.
So $10/month to get the same shit? No thank you. I agree that Google turned to shit compared to what it was but it is still the best search engine out there. Now if the article was about privacy concerns then they would have a point. Which is what Kagi is all about anyway. So let’s stop the fucking act.
Your search results look very different to mine:
Did you disable Grouped Results?
All the LLM-generated “top 10” listicles are grouped into one large block I can safely ignore. (I could hide them entirely but the visual grouping allows for easy mental filtering, so I haven’t bothered.) Your weird top10 fake site does not show up.
But yes, as the linked article says, Kagi is primarily a proxy for Google with some extra on top. This is, unfortunately, a feature as Google’s index still reigns supreme for general purpose search. It absolutely is bad and getting worse but sadly still the best you can get. Using only non-Google indices would just result in bad search results.
The Google-ness is somewhat mitigated by Kagi-exclusive features such as the LLM garbage grouping.
What Google also cannot do is highlighted in my screenshot: You can customise filtering and ranking.
The first search result is a Reddit thread with some decent discussion because I configured Kagi to prefer Reddit search results. In the case of household appliances, this doesn’t do a whole lot as I have not researched trusted/untrusted sources in this field yet but it’s very noticeable in fields like programming where I have manually ranked sites.
Kagi is not “all about” privacy. It’s a factor, sure but ultimately you still have to trust a U.S. company. Better than “trusting” a known abuser (Google, M$) but without an external audit, I wouldn’t put too much wight into this.
The index ain’t it either as it’s mostly Google though sometimes a bit better.
What really sets it apart is the features. Customised ranking aswell as blocking some sites outright (bye bye pinterest and userbenchmark) are immensely useful. So are filtering garbage results that Google still likes to return.
I didn’t change any settings. fresh account
Is “Grouped Results” disabled in settings?
I searched both Cory Doctorow’s post and the linked 404media article in his post for “air purifier” and found nothing. What author are you referencing?
https://pluralistic.net/2024/02/21/im-feeling-unlucky/#not-up-to-the-task
Thanks.
I pay nothing for running SearXNG locally on my machine.
It’s not even comparable in quality. It’s like almost trolling to even suggest they are in the same league. If you don’t want to spend 10 dollars, fine, but maybe stop pretending that your instance is somehow the best quality search engine that exists… :)
Your argument clearly shows that you fail to see the benefits of doing it yourself. I get quality results from my local instance due to my persistence and work put in to adjust the settings necessary. I’ve balanced the privacy and functionality of the instance to fit my needs and it costs me nothing but a few minutes of my time each week to do so.
Kagi doing it for $10 a month sounds like they’re turning a neat profit off of you; and you’re refusing to accept that I have achieved levels of search competence that Kagi has without paying for Kagi or even using their free searches or service.
Whether or not it makes sense to you value-wise to pay or not pay for Kagi does not matter in this discussion. it only matters that none of the things Kagi can do that I find useful are things that cannot be done with SearXNG.
On top of that they’re still paying using their time (and power).
My understanding is that a locally hosted SearXNG instance doesn’t really give you any privacy, unless you “dilute” your searches by letting others do searches from your instance too.
Route it through a vpn with gluetun and it does…
That is the way to go.
The nice thing is that I can customize it however I like too; change weights, choose which engines to pull from always, or even from search to search; so I’m not getting cruft.
SearXNG always rearranges the crap most engines serve to the bottom without fail.
TIL of this. Thanks.
Dunno, I got entirely different results. Reddit, homeairguides, forbes, a bunch of listicles like consumerreports, wired, ny times, cnet and whatnot, and other websites.
I wouldn’t use “air purifier” as a metric, since it was already a big public story that surely any search engine that’s even half paying attention would have made sure the results for are good. Probably some other consumer good is better for an un-preannounced test run.
(Also I’m not sure that searching “top 10 air purifier” and complaining that you got a top result of top10.com/air-purifiers and that’s not what you wanted makes a ton of sense. FWIW, I did try “air purifier” just out of curiosity and saw a very clear result that DDG had the best results, Google second, and Kagi third.)
I repeated it for “good wireless router” and saw different results; for them, the outcomes were fairly similar with Kagi somewhat better in results (returning Wirecutter as the top result, and an obselete Stack Exchange answer as the 2nd, which okay it’s not right but I get where you’re coming from sir), and Google and DDG as secondary (returning PCMag and CNet at the top and Wirecutter only further down below).