sknebelAaronpk: where do you see HTML sanitization in the microsub stack? E.g. some posts seem to break monocles layout, and I wonder if that is CSS leakage of some kind?
LoqiZegnat: [jeremycherfas] left you a message 18 hours, 54 minutes ago: I will do that, but many of the errors were the result of my ignorance. My feeling is that copying the CLI instruction should come first, as that is the easiest option for people like me.
schmartyunless you specify jf2 when fetching, i think the data structures for wm.io data fetched via API are different than the data delivered by webhook that morris saves :/
LoqiIt looks like we don't have a page for "failure point" yet. Would you like to create it? (Or just say "failure point is ____", a sentence describing the term)
schmartydgold: i have moved some of my post's URLs in the past, so they have aliases. that line puts the canonical URL plus any aliases in an array so i can iterate over them looking for mentions that might have come to the old url
ZegnatToken endpoint test was the important part for me, so I wouldn't need aaronpk's hosted token endpoint anymore. Also easier for offline testing of Micropub.
tantekthat being said, I do think the right answer is to first focus on solving the "how do we be respectful to other humans we may not know" question, then second any nation/bureaucracy-specific mumbo jumbo
tantekI'm approaching it from yet another angle, which is the challenge of, I don't want "bad" (insert your own value judgment) "content" to show up on my site via Webmentions
tantekand since anything "federated" involves "other people's content", whether webmention or websub or feeds etc., there is dev-level rethinking to do
tantekjeremycherfas: there is sufficient diversity of discussions in dev that at this point I believe most people in dev may find most of dev incomprehensible
tantekjeremycherfas: that's not the point. the point is dev-like things are so disparate and specialized that it is inevitable that this channel in particular will sound "obscure" at times for everyone
ZegnatI might just rip some stuff out and move it into the config file. As for offline things, I want to disable as many external calls as possible. And I think it is currently impossible to disable things like notifying the WebSub hub
dgoldNO! You don't need a fecking Privacy Statement, and a Processing Register, and a Datalekregister (EXTERMINATE!) and a Processing Agreement to host a BLOG!!!!!eleven
ZegnatSo I have no idea to whom the advice is aimed. Whenever your website is your main business, *and* is handling personal data of visitors/commenters, I am going to assume all the GDPR boxes do need to be ticked?
ZegnatSome of the questions being asked by, who I assume are other bloggers following the #blogpraat hashtag, shed an interesting light on some things.
skippyWhew! Got silo-specific reply and repost support added to my micropub server. https://github.com/skpy/micropub as i only use twitter, this is actually overkill for me, but maybe helpful to others.
@frankmeeuwsen@voxpelli can you help me out? I use your Webmention endpoint on Heroku for my blog. How do I deal with the GDPR regulations that are coming in effect end of May with regards to your app/code? (twitter.com/_/status/993572662952767489)
Zegnat!tell dgold I think I may know why nanopub was failing with Mintoken: you only parse form encoded responses from the endpoint, while the spec only defines a JSON response
LoqiIt looks like we don't have a page for "mintoken" yet. Would you like to create it? (Or just say "mintoken is ____", a sentence describing the term)
Loqidgold: Zegnat left you a message 20 minutes ago: I think I may know why nanopub was failing with Mintoken: you only parse form encoded responses from the endpoint, while the spec only defines a JSON response
schmartysnarfed: oops, not that kind of domain fronting, haha. i am hosting the app on glitch.com and trying out a service called fly.io which lets me point a custom domain at a glitch.com project.
snarfedheh, i opened https://fly.io/ , and i immediately thought, hoo boy, how long had they been around when cloudflare launched workers, and will they survive :P
[tantek]Without research documenting existing publishing examples, analyzing them for patterns, documenting prior formats (attempts), there is almost zero chance of inventing something both actually useful (that someone would write a consuming application), and especially minimal
[tantek]Years of experience with lots of folks proposing various a priori microformats (like your post) that literally went nowhere because there was no evidence behind any of it
jgmac1106thx [tantek] right now there are no standards for marking up learnign events and any futre development is being done by member only profressional orgs where you can’t even see latest specs until becoming member
jgmac1106when I was researching today it said to try and not use any other classes for microfomramts to work, I might had read this incorrectly. I inferred it as I can’t do things in my style sheet if something had microformats if I wanted to ensure it worked
jgmac1106though it did mention in-line styling after the microformats class so that confused me, if I can add inline styles wouldn’t stuff in my stylesheet be better?
jgmac1106I spent a long time researching this and asked at the OER conference. Everyone said no work has been done since 2013. The closest would have been scorm packages