this post was submitted on 04 Dec 2023
119 points (92.8% liked)

Asklemmy

43945 readers
21 users here now

A loosely moderated place to ask open-ended questions

Search asklemmy ๐Ÿ”

If your post meets the following criteria, it's welcome here!

  1. Open-ended question
  2. 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.
  3. Not regarding using or support for Lemmy: context, see the list of support communities and tools for finding communities below
  4. Not ad nauseam inducing: please make sure it is a question that would be new to most members
  5. An actual topic of discussion

Looking for support?

Looking for a community?

~Icon~ ~by~ ~@Double_[email protected]~

founded 5 years ago
MODERATORS
 

Assume that the future can change based on your actions, so any historical information that you bring along with you from the intervening 25 years may quickly drift out of the new realities history.

Edit: also assume that you can be given a healthy 21-year-old body if you want or take your previous self's place.

Further, identification will be provided for you if you were not born at that time.

you are viewing a single comment's thread
view the rest of the comments
[โ€“] [email protected] 4 points 11 months ago (1 children)

I have a theory about NFTs.

One of the issues with running any system on the internet where users can upload files is that you have to store those files and make them accessible to people.

Which is fine. But often times users will upload the same file repeatedly or with only minor trivial alterations and that is how you end up with a multi petabyte server farm completely full of duplicate memes and reposted video content.

If there were a system that could uniquely identify each and every single file that is uploaded to it and then on the fly cross compare that file with previous uploads and identify when the file is a duplicate, that would be very handy.

If such a system like that existed, then system would only need a single copy of the file to share out to every single person that wants to view it, saving possibly hundreds of gigabytes of storage space.

Something like NFTs could be used to accomplish that, if it were fully fleshed out and developed.

It would probably need one of the new neural network systems overlaid on top of it to analyze the content and cross compare it, and to single out minor differences and create overlays to account for them.

For instance, if you have a base image on knowyourmeme.com, then all the AI would have to do is separate a layer that contains the text in text format with the information of the font that is displayed on top of the image, and it could share up millions of variations on a base level meme for only a few megabytes of storage space for the text information.

If your computer system already has the base meme file on it rather than sharing the entire file to you it could just share the text overlay to your computer and decrease transmission file sizes and time as well.

[โ€“] meekah 4 points 11 months ago* (last edited 11 months ago) (1 children)

If there were a system that could uniquely identify each and every single file that is uploaded to it and then on the fly cross compare that file with previous uploads and identify when the file is a duplicate, that would be very handy.

That exists. And we could probably even make something that recognizes small changes like added text on the image and save the data accordingly, probably using some kind of neural network as you later mentioned. The issue is just that that would be extremely computationally expensive, so companies usually choose to just buy a shitload of HDDs. Maybe something like this might be more feasible with quantum computing at some point.

Something like NFTs could be used to accomplish that, if it were fully fleshed out and developed.

What makes you think that? An NFT is basically just an ID card with extra steps. I don't see how NFTs relate to the problem you described at all. Just because most NFTs the public heard about were somehow tied to images or videos, doesn't mean that NFTs are a technology for detecting duplicate images/data in general.

[โ€“] [email protected] 1 points 11 months ago (1 children)

I mean nft's basically serve as a identifier for a file. So if you can identify the file then you can set it as a base image and then cross compare with that identifier. Idiots are using it to like have some sort of weird digital ownership of something but that's not its ideal use case

[โ€“] meekah 5 points 11 months ago* (last edited 11 months ago)

NFTs are non-fungible tokens. They are literally made for claiming some sort of weird digital ownership of something. There isn't really another feasible usecase for that technology, even if I agree that that hasen't been done in a meaningful way yet. If you want to identify the files in that anti-duplication system of yours you are better off using just a regular incremental integer or GUID or something. No need for all the extra steps to the ID card