@Craigp @darius on Duck Duck Go it was the 5th result for me, pretty good.
Strange that you see a stack of summaries, wonder if thatโs location based? Iโve got the UK filter on. I also run Pi-hole + AdGuard in Safari.
Googleโs results in comparison are horrible for content and layout. PG is very far down the list.
Screenshots of both, down to the PG result, attached.
"moby dick a thon" is still the correct top result. also the best way to read Moby Dick
ME: So, you know what the quote marks mean? GOOGLE: Yes. ME: And you're going to search for all of these words? GOOGLE: Yes. ME: I don't want results on anything else. GOOGLE: Ok. ME: Good. What did you find? GOOGLE: lol i ignored all your words here's a fucking shower curtain lmao
@StrangeNoises @darius @anildash yeah, the serious answer these days is go to Standard Ebooks first.
The unserious answer is that obviously the best way to read Moby Dick is http://clickhole.com/the-time-i-spent-on-a-commercial-whaling-ship-totally-c-1825124286/
@darius @StrangeNoises @anildash Itโs interesting (and depressing) how Googleโs pivot from โwe prioritize the oldest and bestโ to โwe prioritize the newest and shiniestโ completely failed to spot some future web-treasures like Standard.
If I may ask, what is your use case these days for raw Gutenberg text?
@darius two ads, a wikipedia summary, a couple more sites and then Gutenberg. IDK why anyone still uses google search in this decade
https://duckduckgo.com/?q=%22Moby-Dick+by+Herman+Melville&ia=web
@darius On Google after Wikipedia I get online booksellers, for the next thirty-forty results at least.
On ddg Gutenberg is in the top ten, after some different language Wikipedia entries.
@darius There's a nicely formatted version here derived from Project Gutenberg:
@darius Fair enough but I'll note there's the source XHTML files at:
https://github.com/standardebooks/herman-melville_moby-dick/tree/master/src/epub/text
@darius
https://kagi.com/search?q=Moby-Dick+by+Herman+Melville&r=gb&sh=NKDthgNg4aBf0LjlX4RLAA
It's at #3, after Wikipedia and Britannica, which seems reasonable.
@darius
this is all planned.
create corporate owned AI
train it on all available knowledge
then destroy the public knowledge to prevent competitors
advertise to all ppl to use AI, because it gives better results
have a networo effect, critical mass of AI user, to get exclusive access to all the new knowledge, to cement your position forever.
it is playing out in front of our eyes slowly, but also quite fast.