It is an open-source project that doesn’t have any ads and works only on donations. Also, it doesn’t log IP addresses or searchers.
SearXNG is not a search engine. It is a Meta Search Engine. It is self hostable so you can choose any instance that you trust or host it on your own. It compiles and shows you results from other search engine. This way, when you use an instance of SearXNG, your IP address and other metadata is not leaked to any other platform.
Comrade Dessalines seems to like startpage https://github.com/dessalines/essays/blob/master/favorite_apps_and_services.md
Self hosted SearXNG still passes your queries to big search engines so there’s still a bit of data correlation they can do, but not nearly as much compared to you visiting their site and getting tracking cookies and all the rest. You can also set up additional proxies to reduce even that exposure further.
It’s just not the best self hosted app. I have an instance on my Raspberry Pi which is really slow and I often have to hit search twice to get results. It often gives me results that have a tenth-page-of-google feel. I tried running an instance on my more powerful desktop but the search engine integration doesn’t work so I have to open the searx landing page and then search. You can add more upstream search engines as well, but in doing so I’ve broken all possibility of getting pertinent results.
I use a public instance at work as well but I often find myself going to google or ddg anyway as only a handful of results on searx are relevant.
Tdlr: it’s a cool idea, but a bit immature.
I’m never sure about what privacy actually constitutes anymore. But since you can just run your own instance of SearX, I would guess yes. You should have control of everything you look up in the internet. Though sites you enter can still log info they get from you anyways, that can only change through other methods. I tried running my own instance of SearX for a while with moderate success, but it is something that requires more knowledge and time than I have, so I just reverted to use Brave Search. It was decent enough, I guess, but I feel that something, I don’t know what, was always lacking from it. I also was never able to run it in a reliable way for some reason, probably my fault. If you don’t host your own but rather use someone else’s instance, then you just have to trust said person/org I suppose.
Thanks for the answer! I use someone else’s instance because I figured that even if they do log things, at least is not Google. And I am supporting an open-source project. Also, I like the search results that I get and how much control you get to customize the page.
I never liked Brave Search because the results were always off and because the company’s founder and CEO is a homophobic right-wing Covid denier, reason which got him kicked out of Mozilla and made him create Brave.
