3 ms·
Thinking you can design a better search engine by yourself is either egotism or ignorance. Assuming you based it on state of the art search engine research, and
by 0xbadcafebee 6y ago
Thinking you can design a better search engine by yourself is either egotism or ignorance. Assuming you based it on state of the art search engine research, and could somehow avoid patent encumbrance, it'd still take you 5 years to match Google's results (and even then not likely) sans all the SEO bullshit.
Most people still believe that it's possible for one search engine to help anyone find anything without it knowing anything about them, which is just ridiculous. To get good search results you practically have to read someone's mind. Google basically does this (along with their e-mails, and voicemails, and texts, and web searches, and AMP links, and PageRanked crawls, and context-aware filters) and they still don't always get it right.
There is no magic algorithm that replaces statistical analysis of a large corpus along with a massive database of customized rulesets.
> We should also prepare the software to boldly lead the way on new internet standards. Crawling and indexing non-HTTP data sources (Gemini? Man pages? Linux distribution repositories?), supporting non-traditional network stacks (Tor? Yggdrasil? cjdns?) and third-party name systems (OpenNIC?), and anything else we could leverage our influence to give a leg up on.
Oh, great, so become the Devil himself, then. Count me out.