[schmarty]whew, slapped a token introspection endpoint onto this indieauth provider and ... i think i'm ready to switch? now to track down all the bookmarklets, ios shortcuts, scripts, micropub/sub clients, and more with long-lived tokens that will no longer work. 😭
@fluffy↩️ Step 2 is where I'm specifically proposing the use of IndieAuth and adding u-vr-avatar (or the like) as a property, and standardizing the way that VR avatars are represented (which, again, can be based on VRM or the like) (twitter.com/_/status/1552492708375064576)
jacky, alex11, ben_thatmustbeme, [Jacob_Hall], Javi, tetov-irc, nertzy and [campegg] joined the channel
@voxpelli↩️ Totally! I would add RSS to that list as well :)
When it comes to IndieWeb features the two most basic are Webmention and the h-entry microformat.
Combined those two makes it so one can easily interact/be social with the content of other people. (twitter.com/_/status/1552659163100184576)
[schmarty]whew! really slogged my way through getting an access token issued but got my media endpoint fixed up and working with my new indieauth endpoint using token introspection instead of the old "throw an Authorization header at the token endpoint".
[schmarty]in other news i think owncast's indieauth implementation is slightly broken? when starting an indieauth flow on their test server at https://watch.owncast.online/ the `client_id` they send does not include a path component. is it OK to assume `/` if path component is missing from `client_id` or should this be filed as an issue with owncast? or both??
[schmarty](taproot/indieauth requries a path in its `client_id` validation and i was about to loosen that restriction but i see mentions of `client_id` requiring a path component in other places like gimme-a-token.5eb.nl)
ZegnatYes to file an issue. But I cannot immediately come up with a reason why your endpoint could not add a / path by default. For http scheme that is almost always assumed
[schmarty]aaronpk: looks like Aperture doesn't know how to do token introspection (still does the old toss-a-bearer-token-at-the-token-endpoint method) which i am trying not to support because it's no longer in the spec!