kirbyHm yes I'd like to install pleroma.
> postgresql
> elixir
> pre-built binaries on a web app?
what the fuck am I doing
lumpen 🍉 III@kirby lmao for real, i rember trying to test it out locally and turns out a valid SSL cert is required so i was just like fuck it
@scrum yea otherwise federation wont work
@kirby yeah it makes sense, they want safe defaults for average person on a VPS, i'm sure there's a technical way to do that, just wasn't a clear option in the install / build scrips to not use ssl, and, not familiar with erland a all .. it's nice to test out something on a local network, i mean it's just http right? i could host it over tor or i2p. and of course would use a reverse proxy to do that. I'm used to nginx handling all the ssl stuff, but i can see why having it built in is a good idea.
I am going to have to slap together a minimal php activitypub thingy when i get my yarvles
I am going to have to slap together a minimal php activitypub thingy when i get my yarvles
@scrum the pleroma internal webserver only listens on loopback iirc you MUST configure a reverse proxy or it will not work correctly also a dev setup lets you use over http on lan
Lamp@kirby @lamp
i appreciate the advice, you can moderate my lokinet based akkoma instance, i will pay you in poocoinne
poopadr=iu431iroh43peaijgfn;adshpiueqftr;oeai;jshf;oisahf;jsaf;lkashjfojash;flksafhsakjf'jsad'klfj;lksadjf;osaifuwenfkjandsgf;sahf;jhfrhwq;ojrhjwe;lkrjewqlkrjwqlkrjewqoihjosfjoiefjoiewjfoiewqfj monkey lasjfoasfoihewqoifh;osf;dsahfa;lkshf
>>> end of poop
save this in an encrypted JPEG
i appreciate the advice, you can moderate my lokinet based akkoma instance, i will pay you in poocoinne
poopadr=iu431iroh43peaijgfn;adshpiueqftr;oeai;jshf;oisahf;jsaf;lkashjfojash;flksafhsakjf'jsad'klfj;lksadjf;osaifuwenfkjandsgf;sahf;jhfrhwq;ojrhjwe;lkrjewqlkrjwqlkrjewqoihjosfjoiefjoiewjfoiewqfj monkey lasjfoasfoihewqoifh;osf;dsahfa;lkshf
>>> end of poop
save this in an encrypted JPEG
@lamp @kirby
yes, it is far beyond my skills, but with a good design far from impossible, but not something i expect to pull off for a long time. .
Obviously you would need to get different components working independently, but designed to eventually fit together they put them in one MVC, you you have the tracker bit, the booru bit, the microblog bit, activitypub bit, diaspora, etc. but all using the same database or cluster or databases. Then they all interface with the central Crud / console
It might be easier just to bridge a couple services together like bridge akkoma to opentracker or something. If the platform an addon system could work. Surprised you can't just entegrate discourse or flarum yet -- torrent trackers usually have built in forums. Might could just add activitypub support to a private tracker like Gazelle and boom.
yes, it is far beyond my skills, but with a good design far from impossible, but not something i expect to pull off for a long time. .
Obviously you would need to get different components working independently, but designed to eventually fit together they put them in one MVC, you you have the tracker bit, the booru bit, the microblog bit, activitypub bit, diaspora, etc. but all using the same database or cluster or databases. Then they all interface with the central Crud / console
It might be easier just to bridge a couple services together like bridge akkoma to opentracker or something. If the platform an addon system could work. Surprised you can't just entegrate discourse or flarum yet -- torrent trackers usually have built in forums. Might could just add activitypub support to a private tracker like Gazelle and boom.
@lamp @kirby
What i've been doing lately is studying all the established open xml schemas and microformats, and trying to ultimately synthetize them into one big xhtml standard, so one page of xhtml or atom might incorporate, xbel(bookmarks / webringts) , spff(playlist), newsml(industry news mark up). or recipexml(some thing i found) rss2 etc. and then i gotta map that to a database. At certain point can't avoid namespaces, but standardizing all that is horribly tedious. no one has to do this, why god
I'mma hit people with excessively semantic xhtml/atom feeds they won't know what to do with. and i hypothothies, will make it easy convert a post, to a blog, to a wiki entry, or news article, create image galleries, audio and videos etc.in as little javascript as possible, while opening up possibilities to 3rd party devs.
And starting with the booru mindset on the outset, is probably the best way to deal with media and gets the most out of SQL, basically media is uniquely identified via the hash, and you can add various tags and metadata to each unique file, so reduce redundancy of storage and bandwith by only having 1 of each image/video/audio and also make all media easy to organize, find and display. So you make a post of a torrent in your feed, it appeals as a 'toot' but then same info attaches torrent unique id of the image, post etc. So now that Movie title is associated with that image and torrent etc. So imagine you see a screenshot for movie, and there's a link under that to the title of movie, a torrent, a movie review etc.could also use activipub+media proxy to do illegal streaming, fun stuff,, Could also use the hash based media identification to deal with spam, harassment etc.
If you have multiple instances using this hash booru system, you can set up alternative sources, like file not available on this server, but hash matches on another trusted server, the source of media is distributed, very difficult to take down.
Why tho? The possibilities are endless, you could do a movie night, host Night of the Living Dead on your server, and anyone who has that movie can just upload at same time. Webtorrent is also a thing tho no sure how that can fit into private tracker scenario
What i've been doing lately is studying all the established open xml schemas and microformats, and trying to ultimately synthetize them into one big xhtml standard, so one page of xhtml or atom might incorporate, xbel(bookmarks / webringts) , spff(playlist), newsml(industry news mark up). or recipexml(some thing i found) rss2 etc. and then i gotta map that to a database. At certain point can't avoid namespaces, but standardizing all that is horribly tedious. no one has to do this, why god
I'mma hit people with excessively semantic xhtml/atom feeds they won't know what to do with. and i hypothothies, will make it easy convert a post, to a blog, to a wiki entry, or news article, create image galleries, audio and videos etc.in as little javascript as possible, while opening up possibilities to 3rd party devs.
And starting with the booru mindset on the outset, is probably the best way to deal with media and gets the most out of SQL, basically media is uniquely identified via the hash, and you can add various tags and metadata to each unique file, so reduce redundancy of storage and bandwith by only having 1 of each image/video/audio and also make all media easy to organize, find and display. So you make a post of a torrent in your feed, it appeals as a 'toot' but then same info attaches torrent unique id of the image, post etc. So now that Movie title is associated with that image and torrent etc. So imagine you see a screenshot for movie, and there's a link under that to the title of movie, a torrent, a movie review etc.could also use activipub+media proxy to do illegal streaming, fun stuff,, Could also use the hash based media identification to deal with spam, harassment etc.
If you have multiple instances using this hash booru system, you can set up alternative sources, like file not available on this server, but hash matches on another trusted server, the source of media is distributed, very difficult to take down.
Why tho? The possibilities are endless, you could do a movie night, host Night of the Living Dead on your server, and anyone who has that movie can just upload at same time. Webtorrent is also a thing tho no sure how that can fit into private tracker scenario