>Almost nobody objected to my characterization of Google and Bing search results, but people told me that I was wrong about Kagi. In some cases, people sent me their actual search results. In every such case, the search results did not contain a good result that I could see (e.g., for the seasonal forecast query, the search failed to return an up-to-date seasonal forecast) and was full of SEO spam. In one case, a person passed me both their list of Kagi filters as well as the search results they got without making claims that the results were good or bad, but people generally insisted the results were good even though the results both failed to link to a useful result and were full of spam except in cases where the user did something like pin GitHub to the top of their results, which worked for the queries where the goal was to download software that's hosted on GitHub, but of course completely fails for the other queries from the post.
This deserves to be a separate post. Kagi would be great if it were 10x less expensive. For the price that is paid, the very first click should deliver the answer or most relevant information to the search with five nines accuracy. Clicks that require a user to perform more digging or are simply rejected as spam should come with some SLA of 0.0001%
Unfortunately we won't get a chance to see whether kagi could make it work before SEO was a thing, so this isn't a fair comparison.
(Not defending kagi necessarily, I've never tried it, no opinion on how it fares on today's Internet but comparing it to how 1999 Google did on the 1999 internet is absurd)
This deserves to be a separate post. Kagi would be great if it were 10x less expensive. For the price that is paid, the very first click should deliver the answer or most relevant information to the search with five nines accuracy. Clicks that require a user to perform more digging or are simply rejected as spam should come with some SLA of 0.0001%