#JKingThere's another fork <https://github.com/nicolus/picoFeed> which has fixed at least some of the tests. It has some pretty big changes (and pretty severe bugs as a consequence) besides test fixes, though.
#Loqi[nicolus] picoFeed: PHP library to parse and write RSS/Atom feeds
#aaronpkheh, that nicolus one is actually a fork of mine, but then he changed it to make it not look like a fork
#[jgmac1106]well this looks kinda like the way I want, the audiopoem was supposed to span 1/3 nut it just goes one, nit sure why so I moved everythignelse over one: https://jgregorymcverry.com/mypoetry
#aaronpkstarting in php 7.2 there's an error with phpunit and in order to fix it i'm going to have to upgrade to a new phpunit but that's going to break eeeeverything
#Loqi[snarfed]: jamietanna[m] left you a message 4 hours, 6 minutes ago: I think we're pretty much done now 🙌 I've got it working from no-authZ to publishing, the main thing left is the doc updates and the ability to disable it - unless there's anything else you think that needs to be done
#jacky[snarfed]: no problem! thankfully, I've gotten better at this lol
beko and sivy joined the channel
#[snarfed]so jacky evidently when you use bridgy publish, elixir sends four identical webmentions to bridgy for it, each just 100ms or so after the last, and bridgy publish's transactionality is a bit brittle. it doesn't do anything wrong, it only POSSEs once, but occasionally the contention makes one of the wms 500
#[snarfed]i'll work on that, but if you can de-dupe down to one wm, that will help
#@swyx↩️ i tried setting up webmentions today. one thing that felt broken was the crawling only turned up like 40 things and it didnt feel like it was going to scale to 1000's of webmentions (more likely the real number)
is there a trick to get bridgy/webmentionsio to crawl everything? (twitter.com/_/status/1226742112626057216)
[jeremycherfas] and [chrisaldrich] joined the channel
#@schnarfed↩️ right. bridgy doesn't send webmentions for everything in history. it's mainly designed to monitor and send webmentions for *new* tweets and reactions, as they happen. (twitter.com/_/status/1226752663854862336)
#LoqiThe Open Graph protocol (OGP) is an open* standard developed and controlled by Facebook for expressing the primary subject of an HTML page in custom <meta> tags for the purpose of Facebook showing link previews; in practice only a couple are even sometimes necessary for that use-case, and you can use existing open standards instead https://indieweb.org/ogp
#swentelhow does the offline reading work for you?
#GWGswentel: I only did a minor test. I spent hack day enhancing my Parse This library (my xray equivalent) to support turning the WordPress rest API into a feed and importing opml
#GWGI have work to do to get Microsub displaying correctly for all my feeds
ankit, swentel, NeatManatee, petermolnar, gxt, TGiske, wink, crazed, simons and [jgmac1106] joined the channel
#[jgmac1106]don't use the experimnetal x in a u-read-of post, there are enough people playing it isn't necessary, gRegorLove doesn't do that with indiebookclub
#[jgmac1106]talking to inoreader dev he did say he looked into h-feed when adding support to micro.blog for inoreader but the h-feeds were so malformed he had to drop...which is true the mf2 can be quite messy in micro.blog themes, so went with json feeds instead
sivy joined the channel
#GWGInteresting... I wonder what we could talk them into
#sivywondering: those who of you who are POSSing tweets/notes, how do you handle replies? I'm POSSEing new Notes, but on't really have a way to "reply"
#swentelGWG, when microsub is empty? at least 'items', but it's an empty array then
#swentelbut I actually don't know that the spec says about that
#LoqiZegnat: jeremycherfas left you a message 21 hours, 53 minutes ago: I like the way Pinboard unobtrusively tells me I have saved something before and asks whether I want to edit or replace.
JKing, NeatManatee, simons, gxt, KartikPrabhu, [Michael_Beckwit, [LewisCowles], aaronpk, [manton], [tantek] and [snarfed] joined the channel
#[snarfed]jacky i think i've fixed the bridgy publish 500s you were seeing, you should get 429s now instead. feel free to leave elixir as is and see if that change helps.
#[snarfed]sivy re replies, are you asking about UX? most of us either use social readers like indigenous, which let you reply via micropub within their UIs, or use bookmarklets or browser extensions to open a reply authoring page on your site with in-reply-to the current page
#sknebelsince it's only the teaser texts, I'd use e-summary instead of e-content
#sknebelthe author is only a child, not an author property
#sknebel(you could also move it to feed-level, assuming you are the only author on that site, instead of repeating it every time
#sknebelthose should be fairly simple tweaks though, good job :)
[KevinMarks], [Katherine] and CreamyNebula joined the channel
#[tantek][snarfed] good response re: Bridgy and webmentions. But that gives me an idea
#[tantek]would it be too bold to suggest "Bridgy Backfill" as a service that went through and gradually sent webmentions for *past* responses (obeying various limits etc.)?
#[tantek]could help with migration away from social media
#[tantek]anyway if it's even remotely interesting to you (as opposed to way out of scope 🙂 ) then I'll gladly file an issue to describe, brainstorm etc.
[snarfed] joined the channel
#[snarfed]not at all! maybe not the exact name, since that would confuse users as to ownership/support, but the idea, totally
#[tantek]totally, I'll suggest it as a placeholder and happy to accept whatever you bikeshed it to 🙂
#[snarfed]oh, as a feature request for bridgy, sure. substantial one though
#[tantek]feels like a new top level Bridgy service, that's my point
#@seaotta↩️ Thanks for the feedback A.J., adding @AaronGustafson who works with me on the project and setup the voting.
The site uses webmentions so if you share out on Twitter, for example, that counts as a vote. (twitter.com/_/status/1227018667537268736)