I was thinking about how we might do search on 
when I remembered a decentralized torrent search app Iād tried years ago.
It uses itās own tor-fork and communicates with peers who can share torrent files. No websites, no centralized servers.
It occurred to me that this could be forked and modded to use instead of (or in addition to) torrents and maybe wouldnāt even be that much work to get it done.
What do you think?
Iām not sure how it would integrate with Autonomi because nodes donāt know anything about files, only chunks. So you would need to figure out a way for people to publish the file addresses which an app could then access.
At the moment this is difficult because there is no name-space, and the way registers are implemented prevents the creation of universal name-spaces in the way that we had earlier.
Right now a list of files or similar (like website addresses) need to be shared explicitly, rather than discovered or looked up.
Would you elaborate on this please ā as perhaps Iām missing something.
It appears to me, that such information could simply be packaged as a torrent and then added to a Tribler fork with tags and description ā¦ then itās shared.
Iām not supposing that this is a perfect integrated solution. Merely a means to submit, search for, and retrieve content on
My mistake. I thought you were looking at using Autonomi API / nodes rather than an independent network handling torrents. Of course separate indexes can be built, but as you go outside the network you are going to dilute or break the benefits of Autonomi, so I donāt consider those approaches consistent with our goals (cf. the fundamentals).
Iād be very happy if there was a proposed way to build this integrated into the network. Until then any way we can enable a searchable way to share the data we upload is a win I think.
I think Tribler already has a method of detecting the file type - as it can play videoās when they are in the torrent. So it should be possible to create a ā.antā file and put the info in there needed by the client ā¦ then we can pass the file based on extension to a script to use the client to get the file or files and after getting those, it could pipe it to whatever app is needed based on extension, media player or even a browser perhaps.
You only need an APP that you send your siteās indexing to. Then the search engine owner or collectively the users, will read the message and add the that Siteās indexing record to the indexes of the search engine database
ChatGPT
You can find this clip on YouTube. Based on the screenshot, the clip is from the YouTube channel āAnimatedā under the video titled āTrevorWallace Fondled a Fruit While on Shrooms - Tales From the Trip.ā The video is part of a series where people recount humorous or surreal experiences, often animated for comedic effect.
To find this specific clip:
- Go to YouTube.
- Search for āTrevorWallace Fondled a Fruit While on Shrooms - Tales From the Tripā.
- Look for the video from the channel āAnimatedā.
Alternatively, you can directly visit the channel āAnimatedā and find the video in their uploads or playlists.
Wouldnāt it be more fun if I didnāt have to go to Youtube and search, but Chatgpt could just show the clip as the 1ās and 0ās that they really are after receiving the image above? What is search actually in the age of AI? If AI experienced everything on the internet already, isnāt search/accessing data a regurgitation of data in which ever format you like?
AI has to get itās information from somewhere ā¦ so we still need a system to aggregate links and info on the link.
Sure, but once it got it that can actually be stored in memory, the Network is in a sense already that memory and it wouldnāt have to go to a third party anymore. The fun part would be that the video above could even be replayed in south park style if you so choose to, with a language of your choice etcā¦
It doesnāt actually. I would be nice if this was built in, but not part of any plans Iāve seen. Thatās why Iāve proposed a forked Tribbler app as a way to have a method for discovery.
