sebbuyou could make a temp or new email, and make thoses accounts on big sites, and never check them except if you need a 2fa code or some other verification thing
GWGFor example, it doesn't say how you as a consumer might ask for a ticket... because asking is out of scope. I believe one implementation sent one to anyone who had an endpoint
GWGIn my opinion, it is the sending/redeeming of tickets and their usage. Requesting one, save any debugging solution we put in for testing, should be out of scope as well as how/what uses the token once redeemed... once we get the ticket flow tested we can discuss more around the periphery.
sknebelright now there is only one url given, does the token work for other urls? which ones? (bunch about that in questions section on the page already)
rubenwardyI can't seem to log into telegraph, just says " The authorization endpoint for the returned profile URL (https://rubenwardy.com/) did not match the authorization endpoint used to begin the login. "
[tantek]rubenwardy++ for a well-written and thoughtful post. appreciate the sharing of experiences good & bad and hope we can use that to improve docs & tools
[tantek]rubenwardy, I fear the answers to your authorship questions here (in chat) may have been given from a plumbing perspective rather than a publishing perspective, and I'd like to fix that.
[tantek]there's a very unfortunate webdev habit of providing plumbing-centric answers to publishing-centric questions. two very experienced members of this community just did that exact thing in #microformats channel recently.
[tantek][snarfed] literally all of my early POSSE work was figuring out how to take relatively "rich" post types and "downscale" them into plain text notes on Twitter, then limited to ~117 characters (leaving space for original post permalink in the 140)
LoqiText-first design refers to the practice of designing information and UIs so they are readable/usable/actionable at least as basic plain text https://indieweb.org/plain_text_design
[tantek]so in the specific example of: "eg the fediverse, Bluesky, and Nostr don't understand bookmarks (among others), we can't yet POSSE them there, but we can happily post them on our own sites" (as [snarfed] noted in #indieweb earlier), the answer is to allow cross-posting of bookmarks to those other destinations with the plain text equivalent of a bookmark post
[snarfed]I love that for people to choose on their own sites, but I was reluctant to lean far into it in Bridgy classic, since it has more potential to surprise users if they don't expect it
[manton]Hey ActivityPub nerds… 🙂 So Mastodon recently added “Translate” links to posts that (I assume) don’t have a language set. It adds a lot of clutter because I don’t need to translate from English to English. I’m finally looking into this so I can update http://Micro.blog to spit out whatever JSON Mastodon wants.
[manton]From what I can tell, the only language set is using “contentMap” with “en” for the content… But that means duplicate the content text in every single JSON. Seems bloated. Anyone have experience with this? Maybe I’m missing something?
[tantek][snarfed] agreed with keep Bridgy Publish simple! The special Bridgy markup makes sense. Another standards based way could be plain text p-summary markup.
[tantek]we have a long running thread about using p-summary for plain text fallback behavior in general, so I wonder if that's a good signal for Bridgy Publish to use on "unknown" post types as an explicit fallback
gRegorI've tried p-summary with BF before and I think opened a github issue. I think the main thing was Mastodon (or AS?) use of "summary" as the spoiler tag
[tantek]not suggesting any use of Bridgy Fed to send p-summary -> Mastodon AP "summary", because that would be a disaster due to the content warning overloading 😕