@niclas @robby i toyed with the idea to build a small crawler to just index select sites, like programming library documentation, maybe select youtube channels via rss (or god forbid fedi users! :P) etc.
most often i don't search things i never have seen, but things i know to exist but i need more details on.
@bonifartius
Yeah... I haven't spent any time thinking about this until now, so I guess there are many aspects to this, but I think a combination of content filters and black/white domain lists are absolutely required.
Personally, I already have a problem with my own hard disk, the 2.3 million files are a bit "too unwieldy".
@robby