[chrisaldrich], Seirdy and gRegor joined the channel
#Seirdyjust added notes to my indieweb site https://seirdy.one/ and also changed all the URLs (re-directs are in place, no worries). and an atom feed to both posts and notes. i already had an rss feed for posts which i'm keeping since some people are already subbed to that, and i decided why not have an rss feed for notes too. so now i have four feeds, two for posts and two for notes. planning on
#Seirdyadding an rss+atom "combined" feed later, and then adding some activitystreams metadata to the atom feed before i set up WebSub (built from my atom feed). so that'll be three sets of three different feed formats in all. am i doing this right lol
#IWDiscordGateway<capjamesg> I mean is there an endpoint that would take my h-feed and turn it into an RSS feed on the fly?
#IWDiscordGateway<capjamesg> The use case is a translation layer for feeds on static sites so I don’t need to maintain the code for two separate feed types (and in reality it is three with JSON Feed included.)
arbitrandomuser, mro, tetov-irc, [Chris_Lott]1, lanodan, cybi, oodani, kloenk, Xe and [timothy_chambe] joined the channel
#vikanezrimayathinking about implementing a media endpoint that would automatically transcode files into efficient formats (because I love snapping pictures 5-10M big and I hate for them to load slowly... also thumbnails!)
#vikanezrimayapretty sure the idea is not new, but when I actually think about storing files, it might get a bit complicated
#vikanezrimayashould the files get linked somehow? say, by having the same filename or a json file that lists the original, the thumbnail and the efficient full-size copy? (e.g. webp) should it then get deleted all at once when a user requests deletion?
#vikanezrimayathankfully updating files isn't really a thing in the media endpoint, at least I don't have to deal with that
#vikanezrimayaOh, I know! I could store them in folders! then a folder will get a json file, a webp file and a jpg/png original file (for archival purposes)
#vikanezrimayaactually this makes the backing store completely transparent to the files I store inside... it doesn't need to concern itself with file formats, it just needs to reliably store byte arrays
cybi and lanodan joined the channel
#vikanezrimayaand only now I remember that allocating a lot of memory for a file upload is not a good thing and one should stream file uploads to backing storage if possible
mro, gRegorLove_ and cybi joined the channel
#[Chris_Lott]1[fluffy] Don't get me wrong, I think the simple CSS frameworks are great. If I had any clue how I could use one of them and still have the IndieWeb stuff work, particularly from a statically generated site, I would do it! Until then, I feel stuck with the monstrosity that is WordPress theming.
#Seirdy[Chris_Lott]1: you can actually get a lot of IndieWeb out of an HTML+CSS SSG if alongside the static site you have some API endpoints. I've got mf2, webmentions (API endpoint), Atom (will add ActivityStreams and WebSub to it soon), 3rd-party search, web sign-in, and client-side POSSE (my bookmark and note scripts add them to my static site and syndicate to Fedi and TinyGem)
#[Chris_Lott]1Can I see your site and/or are you writing up how you did it? This is ultimately where I want to go so I can, in the end, spend a lot less time working on the site and more time creating things for it!
#Seirdy[Chris_Lott]1: there's a "source code" link in the footer
#Seirdyjust gotta figure out how to make a "combined" Atom feed containing posts and notes (and maaaaybe bookmarks, but those are different and generated by extracting an export of my buku database: https://github.com/jarun/buku)
#Loqi[jarun] buku: :bookmark: Personal mini-web in text
#Seirdybasically as long as you use or host some separate services for webmentions, websub, search, etc. and make your "posting tools" generate content for both your site and POSSE silos, then you can get away with your actual website being 100% static. the site is only a part of the whole system.
#Seirdyalso searchmysite.net is super awesome and everyone who wants a quick and easy search tool should check it out. they also offer an API.