I use Kagi and uprank non-shitty sites. I think there’s also uBlacklist or uBO filters that can remote shitty sites.

Maybe try this https://www.scienceos.ai/ – not as a replacement but as an addition

I’d say you’re not using the right filtering, which honestly takes a lot of practice and sheer luck.

Usually if I get garbage results, I try neuroscience -“top 10”, which would remove the generic top 10 reasons for X results. If it’s still self help garbage, then try more keywords like lecture, presentation, or theory. Some searches I straight up can’t figure out though, and the query ends up removing being a mess. In those cases, it’s worth trying again later and seeing if you’ve phrased it won’t.

https://ahrefs.com/blog/google-advanced-search-operators/

Most of the operators in the working list also work in other search engines including YouTube’s internal search engine.

Some search engine do not index page on their visit rate or activity but on the contrary look for “forgotten” page, my favorite is https://search.marginalia.nu/.

Another way of searching for interesting page would be the use of internet link collection, one of my favorite being https://peelopaalu.neocities.org/. But while the links and pages are often interesting it’s hard to look for a specific subject.

What you’re missing is the semantic and context inherent a syllabus. If you find a mentor that helps develop a syllabus then it’ll become much easier to find learning resources.

Due to how shitty our world is… you just slowly accumulate trusted sources. Google has no motivation to give you a reliable news source that’d compete with their incredibly anticompetitive news tab.

As a tech person I hang around on HackerNews a lot and have found some good sources through that.

While not a direct answer to your question, I want to provide this

https://scholar.google.com/

I paid for Kagi for a while and many of my coworkers use it. It’s a solid and growing engine that’s getting a a lot right re: creating good UX and generating search results (which should be the goal of a search engine, *sigh).

That said, l use SearxNG daily nowadays because it’s decentralized and privacy-focused. You can use any of the public instances or host your own if you like.

Here’s an example of the search results for “Neuroscience” on the instance I use.

You can use the - (minus) sign,
which excludes pages that contain the word which has the - in front of them.

For example, currently I’m replaying GTA V single player, but when I search for content related to it, I’m often given articles about the online version.

To solve that I search:

GTA V <insert-topic-of-interest> -online

Which excludes all articles containing the word online.

I use SearXNG though,
but afaik this is implemented by most search engines.

Go to Wikipedia and look at the references. The Wikipedia editor credo is “show don’t tell”. Meaning don’t explain it. Summarize something that does, and link it.

Also, donate to Wikipedia.