Ivy_Alpha[d], KartikPrabhu1 and KartikPrabhu joined the channel
#npd[m]I've often dreamed about creating a system for "scrobbling" my consumption of all kinds of media, and especially of the things that I read, whether it's pages in a book or short ephemeral social media posts
#npd[m]remember when music scrobbling was a thing, rather than just only using Spotify and then seeing its cute Wrapped year-end summary?
#aaronpkscrobbling is still a thing!! last.fm can even pull your listening data from spotify :)
edburns[d] and nertzy joined the channel
#npd[m]I'll try to reset my lastfm password and see how many years ago I stopped scrobbling music :)
KartikPrabhu joined the channel
#npd[m]there might be a 14 year gap, but I'm impressed the data is still there and maybe I'll try using it again
Seirdy, gRegor, Guest64873, [jeremyfelt], P1000[d], gerben, gRegorLove_ and KartikPrabhu joined the channel; Guest64873 left the channel
#Murray[d]Yep, definitely still a thing, there have even been some useful new tools out for Last.fm in the past year (though I see one of them seems to have disappeared again :() and plenty of people use other services for scrobbling video media as well 😉
#Murray[d]I still need to dig into iTunes libraries and see if I can pull out listen count, I used that for years to track music habits. It's definitely not date stamped, which is a shame, but I feel like I should be able to know how many listens I've had at least 😄
tetov-irc, IntriguedWow[d], KartikPrabhu, Seb[d] and kogepan joined the channel
#petermolnarTIL I've re-learnt tumblr has a 200 posts/day limit
Guest6 joined the channel
#petermolnaryou gotta be kiddin' me... it seems like tumblr hard rewrites (as in in HTML source) in-description HTML URLs to https://href.li/?[original url]
#petermolnarthere doesn't seem to be a way to turn this behaviour off - href.li in front of _any_ URL on tumblr - meaning I don't seem to be able to make an actual link to the canonical post
#voxpellipetermolnar: Twitter does the same, I believe its a thing where they want to be able to block bad URL:s at runtime without having to update all source links across all places
#voxpelliwebmention would allow for you to follow the redirect, no?
#petermolnarit doesn't get that far, because the parses can't find the link to the source url
#aaronpkthis is possible, but you have to think about it differently and also the receiver has to accept URLs on domains that are not their own
#aaronpke.g. for my twitter profile twitter.com/aaronpk it links to a t.co URL, but you could send a webmention from twitter.com/aaronpk to my webmention endpoint with a target URL of the t.co URL. the webmention is valid and my webmention endpoint can validate it, but it has to be willing to accept a webmention for the t.co domain
#[tantek]it wasn't (just) the network effect that stopped it, it was the RSS/Atom wars and the insistence that RSS was "done" and didn't need to evolve (e.g. to incorporate new social features that were emerging and popular)
#[tantek]combined with Atom being saddled with XML baggage and XML namespaces being so completely dev-unfriendly as to not be a reasonable path to extend/standardize any "social media" extensions in Atom, though people earnestly tried with Activity Streams 1.0, and IMO only failed there because extending XML is crap for actual standardization & interop
#[tantek]and don't get me started on the distraction that was Open Social and the Open Stack
#[tantek]neither of which was actually about making sites interoperate directly with social media features
#[tantek]or rather, they were both saddled with so much other unrelated/unnecessary baggage that they were hopeless for getting actual developer adoption (which they ignored in deference to attempting to get BigSilo adoption, which was never a path to success for anything anyway)
#[tantek]also AS 1.0 got absorbed into OStatus which you (voxpelli) already pointed out what happened with/about
#[tantek]aaronpk, re: "where a project's data has no escape route" criticism of GitHub is pretty true tbh
#[tantek]1. the API only provides a subset of access as we've discovered with various Bridgy features
#[tantek]2. it's still a *proprietary* API which presents a high barrier even to most devs
#[tantek]but yeah, that's worth acknowledging in that criticism subsection. "They have an API that covers most project data functionality, however, it is only 'most', and a proprietary API is still a big barrier to easy import/export or other interactions"
#[tantek]and we can cite specific Bridgy issues (feature requests IIRC) that are blocked because of GitHub API limitations
#[tantek]finding common ground here (desire to get data in/out of GitHub more easily) may be worth it
#[tantek]and maybe folks working on Mastodon will start "owning their own issues" or even "owning their own reacji" when interacting with GitHub
#voxpelli[tantek]: yeah, the format wars really contributed and web3 is yet another instance of that
#voxpelliJust like replacing XML with JSON didn’t fix social networks, replacing HTTP with something blockchain based won’t fix it either
#voxpelliIt’s just a distraction of needless plumbing that prevents any actual useful work to be done
#[snarfed]Bridgy's difficulties with GitHub's API have generally been about the _mode_ of access, not the availability of data. Their API(s) have pretty comprehensive to all types of data there
#[snarfed]what they don't have, that Bridgy would ideally want, is realtime interaction style queries like "give me the most recent comments on all issues I've participated in," or "give me the most recent emoji reactions to anything I posted," especially across repos
#[snarfed]er sorry. yes! they're great for feed readers, but not really for programmatic use. they don't provide queries like the ones we need, and the URLs are per user and afaik not discoverable via API at all
#capjamesg[d]Indeed! The RSS feeds are not ideal for programmatic use.
#capjamesg[d]And in many cases they require a secret token which I don’t think anyone except a GitHub user themselves should have access to.
#[snarfed]right. so for code to even start using them at all, a user has to go dig them out of the settings and copy and paste them into a text box. kind of a non starter