This is great. I wish there was a way to block certain domains from ever appearing on search results.
EDIT: Looks like there's already an open issue:
EDIT: Looks like there's already an open issue:
The intent of SearXNG is to be stateless (with no sessions on the server) and to work without JavaScript.
However, this approach limits certain features because of the restricted size of cookies (and other forms of browser storage require JavaScript).
I'm picturing more of an instance-wide configuration of domain blocks for a private, single-user, self-hosted instance. But I understand this may not be the intended use of the project.
https://github.com/searxng/searxng/blob/f1a148f53e9fbd10e95b... https://github.com/searxng/searxng/discussions/970
Kagi has this feature built in and it is a good user experience.
You can also use the uBlacklist browser plugin. My problem with that is that is slows everything down. I am not certain but I think all the works is done after the search is complete. That it filter the actual result. The two above limit it from ever being part of the result.
I understand Kagi is generally reputable, but I like the idea of a self-hosted alternative where you're in full control.
My guess is that if they are found to do so, then they open themselves up to lawsuits. Not collecting data isn't merely a perk - it's practically the reason Kagi exists.
If you don't have any logs you can just always say the princess is in another castle, since you can't provide data that doesn't exist.
If on the other hand you do have the requested information, you need to determine the validity of the request, and then extract the data; or refuse to comply and possibly put yourself at legal risk. For a smaller business that's probably a can of worms you'd rather avoid opening.