💾 Archived View for station.martinrue.com › haze › d07812590b7549e98d1530bce91856af captured on 2024-07-09 at 01:06:28. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2024-05-12)
-=-=-=-=-=-=-
Thanks to @krixano 's post on search engines. I found -motivation to finish my old post too!
gemini://gemini.clehaxze.tw/gemlog/2022/07-22-search-engine-dilemma-bias-vs-accuracy.gmi
I want more conversation about the topic. What kinds of search engine do we (as a community) want. There are tradeoffs that we make. But who's to juded the tradeoffs?
2 years ago · 👍 acidus, freezr, mfoo2
gemini://gemini.clehaxze.tw/gemlog/2022/07-22-search-engine-dilemma-bias-vs-accuracy.gmi
Your article is really good! But I do have some thoughts on the NLP portion, because it seems you only handle Machine-Learning NLP. I might be writing a response soon :D · 2 years ago
Oops... I fixed typos hehe · 2 years ago
Nice articles even though a lot of typos are affecting the text... 😅
My point here is this topic is addressed in a complicated way since the beginning, with too much overthinking.
Before to translate into a computational paradigm I would try to solve the issue in a pragmatic way.
For instance: how would you do a research in a library without an informatic archive?
We have the great benefit that scraping GMI files is pretty trivial since them are essentially linear text organized per lines.
Good job then! 👍 · 2 years ago