this post was submitted on 02 Oct 2024
134 points (96.5% liked)
Asklemmy
43775 readers
1619 users here now
A loosely moderated place to ask open-ended questions
Search asklemmy ๐
If your post meets the following criteria, it's welcome here!
- Open-ended question
- Not offensive: at this point, we do not have the bandwidth to moderate overtly political discussions. Assume best intent and be excellent to each other.
- Not regarding using or support for Lemmy: context, see the list of support communities and tools for finding communities below
- Not ad nauseam inducing: please make sure it is a question that would be new to most members
- An actual topic of discussion
Looking for support?
Looking for a community?
- Lemmyverse: community search
- sub.rehab: maps old subreddits to fediverse options, marks official as such
- !lemmy411@lemmy.ca: a community for finding communities
~Icon~ ~by~ ~@Double_A@discuss.tchncs.de~
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Would use 1/2 on ebooks, 1/2 on passible quality music I'd give up some quality for size reduction. I'd probably save a little space for a sudoku game.
20 years ago, I used to have custom playlists that I'd load onto a 64mb SD card for my commute. A 64kbit wma file is surprisingly listenable. You can fit more than a standard album onto 64mb. I'm trying to remember now how much, probably 20-25 songs?
At that compression rate, I expect you can get a few hundred songs into half a GB. Of course, picking them would be its own challenge! It'd be the most curated playlist in history.
I'm going to willingly misread this as you wanting AI "compressed" ebooks lol
I have read somewhere that some text can be compressed incredibly efficiently in some AI models. The issue being that the compressed data is worthless without the model and power to recover it.
https://www.hendrik-erz.de/post/why-gzip-just-beat-a-large-language-model