💾 Archived View for station.martinrue.com › krixano › 570d8cf2d4ae4a4f9242072f7d5b0a86 captured on 2023-09-08 at 18:02:59. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2023-07-22)
-=-=-=-=-=-=-
I don't know why I have to fight off people who want to be extremely defensive about their favorite search engine algorithm. I've made my choice. I've read the papers, I've seen the band-aid fixes that are required to prevent link spam and reduce TKC effect crap, even though these two things only exist *because* of the algorithm looking at links the way it does, which is extremely explicit in the papers themselves.
I don't give a crap that these papers call it "authority" or "influence". That's just a cover-up to make it sound fancier than it really is, which is literally popularity - statistics on how many links something has, and how many pages link to it.
1 year ago · 👍 superfxchip
@danrl Thanks :)
I'm just frustrated because instead of people respecting my opinions about how the algorithms inherently work being bad, they'd rather claim I'm just *not understanding* is properly, or dismissing it based on what words are used. If the only algorithm someone has compared these popularity ranking algorithms to is basic FTS and metadata searching, then obviously they are going to look "better". But that doesn't mean it's a good algorithm. · 1 year ago
u do u. thx for what u do! · 1 year ago