You aren't going to get decentralized free open search that's any good because search lends itself to being a business because of the absurd costs involved. The bare minimum table stakes are 1. Index the entire internet and 2. Have a good method of efficiently selecting a subset of that index relevant to a query. Both of those steps are extremely expensive and work far better in approaches that accumulate revenue to offset costs.
People are still saying that Google's index (and server clusters) are far and above the competition and that might be still true but what does it matter when they are throwing out indexed data from older pages?
Also if most of the webs content is consolidated into a few sites like reddit, stack exchange and Wikipedia, IMDb etc. then doesn't the search index only need to be about as big as those sites?
I just don't see results I'm getting from today's Google meaningfully far and above the others and increasingly more and more I have to use either another search engine or search the site directly to find what I'm looking for.
