I Still have no idea what that that term means exactly, but I did use AI to build the program. I just rigorously test it and go into the code to figure out the issues

Lol I didn’t copy the hole prompt I deleted the bit at the end, but it was late and I was tired so I used an AI just to fix my original text

I want to encourage fedizens and Internet user to collect their own data and run their own algorithms since it’s worked well for corporations in general but to keep using their services. Seems as though people enjoy algo based on their data.

It’s neat that fediverse has a general principle of not collecting user data, so if more people used fediverse instances more often then less Data going to corporations. This browser extension is outline of how collecting your own data can affect your experience with fediverse. There’s so much you can do with your data and data from api

Extension are like plugins, the program monitors watch time and wanted that to be done by the user and stored by the user and shared consensually. Plus if this algorithm boost engagement it should benefit people that don’t use it because it will encourage people to watch and like things which contribute to other algorithms

its a browser extension for brave (or any chrome based browser) it’s in the github readme. recomendation algo was self explanatory. it’s meant to recommend you videos on peertube. i only screen shot the only ui that exists, the only things I can screenshot is variables stored in indexdedDb and local extension.

also the installations instructions are in the github readme

I did make a classier for for videos that inputs title, tags, description, and closed caption into an LLM. I got roughly 1000 entries classified that way, issue is that most of them were non-english videos and then new videos come from somewhere on peertube that don’t hace these classifiers.

Video processing is cool just expensive computationally. Also watchers could classify the videos themselves then use a cosine similarity (or whatever algo) on that. I did suggest to peertube to share the categories people say a video is with other people (like it’s a Mastodon post) eventually it morphed into an idea light weight peertube instance that only does api.

apparently not id have to convert the scripts, and i have a lot of issues with the current script

sweet i do need ui to display the recommendations once i get there

yeah no algorithm are a big thing. I’m trying to figure out a simple algo right now so it can light weight. one guy did suggest a slider for things you do and don’t want suggested. I did point this out in another post that there wont be 1 algorithm theres gonna be different kinda of algorythm (on the fediverse), i can see that theres gonna be user collected, user run algo then theres gonna be algo where people share their data and some central server will send suggestions to individual based on that data(because there are people that are okay with sharing their data).

I’m trying to push “collect your own data, run your own algorithms” as a tenant of the fediverse.

I’ve been developing for brave so it will work for chrome and it should work for firefox. for brave its easy you put the github files into a folder, put brave browser into development mode, and then manage your extensions by adding the folder, you press f12 to get the console and look at the variables in extension storage.

if you get it to load properly you should see 🔄 Starting fetch for template: https://dalek.zone/api/v1/videos?sort=-publishedAt&nsfw=both&count=10

in devTools on brave

MIT

i figured that some did it when elon banned that account on twitter i never got around to looking for it

Yeah collecting your data and running your own algorithms are going to be opt in by default. All peertube instance would have to do is send information described all the videos and you’re local machine will pick it out for you

lol we ignore real problems in the world to do this

i wouldn’t endorse violence if the rich werent actively doing violence

yeah i saw that stats when i looked at my videos, but i didn’t know if that would be worth collecting since its guarded not readily accessible. i guess an instance can use it for recommendations of local videos. right now im just trying to profile both the watcher(locally) and videos

really im thinking about what data is okay to share and what data should be kept to the user. basically I determined that description of the video is only thing that can be public and the people/bot describing it okay to share (like associating their channel to a description they make to specific video) and the watchers device can collect video meta data to find suggestions

I like that idea for a stupid simple algorithm. ironically I plan for there to be like a Varity of algorithms both that are user only and a aggregate. really im trying to pin down a standardized video vector that can describe any video to any level of detail

be better to store the video vector on an instance so that watchers can retrieve, just logistics. video vector (element) can be calculated anywhere just communicated to an instance, the idea is to be flexible. activityhub protocol has made the decisions easy the video vector has to be a .json element in a video json data.

it would be better to store the results of a calculation to avoid repeated calculations. im looking into music classifications, and like the entire video can be sent to parse to see if its music or not, the tempo, genra, id assume that would be fairly costly to calculate or instance can send the video vector that states all that information

im not at the aggregating data stage but you can just put a random id on a data set