hi, this is a tiny experiment. I think that you will get this teal square with sha1sum 7659b733afa495b561acd7abc0f30201ad5eede0 in your server's media store, even though you don't follow my account and didn't solicit information exchange with my account whatsoever.
Context: https://kolektiva.social/@beka_valentine/109547485673333984
@jonn @moderation that media is hidden for me, and i follow both of you. like, i see a white square and a little hidden logo
@jonn @moderation in fact, i cant even open this in chromium
@beka_valentine @moderation that's peculiar! Perhaps, kolektiva.social is set up to hotlink images. I'm running an instance that is faithful to the media decentralisation philosophy, which says "if a toot is displayed on my instance, the server must be able to display it in its entirety".
There's also a possibility that there's some hybrid setup where some stuff is "half-federated"!
Either way, if this file doesn't end up in the storage of kolektiva, it's some really nice forward-thinking programming / protocol design!
@jonn @moderation thats not decentralization thats centralization through your instance. hotlinking is the essence of decentralization
@beka_valentine @moderation hotlinking dies to link rot, replication doesn't. Hotlinking dies to malicious actors too.
@jonn @moderation sure, i mean that's obviously an issue! but fixing it shouldn't be done by defaulting to centralizing mechanisms
hotlinking should be the default, with archival mechanisms as the fallback only
@beka_valentine perhaps. In my personal experience, I *wish* there was a bunch of servers who'd have backed up my posts and stuff I was hosting on bare metal. So I guess I'm too jaded by the experience of losing data.
@jonn i mean there are! the internet archive will do that for you. but it's not designed to backup dynamic content so you'd have to do something to deal with that. that's a design flaw in mastodon tho
@beka_valentine oh, it doesn't back up obscure pages though.
@jonn yes it does, you just have to explicitly request it. it tries to archive everything it can, but it might not *find* stuff, which means you have to explicitly tell it to archive it
@beka_valentine I'm older and smarter about backups now. I think it was an important experience anyway. Besides, there's a decadent pleasure in vaguely remembering the content that nobody will ever see. Death before death.
@jonn yeah. im sorry you've lost content. :(
but check it just in case! it might have captured it!
ALSO! you can just upload stuff to the archive, without even using the wayback machine. you can just archive things, as long as you're OK with them being public
also also, use github
there are so many options