#GWG!tell kylewm, snarfed The Bridgy literature says it returns a 201. It seems to return a 200. 200 is correct according to the webmention specification. The literature seems wrong.
#GWGsnarfed: Also, the WordPress recommendation is for the Indieweb plugin...not sure if you want to add to that as it doesn't include the webmention or semantic linkback pieces.
#GWGThe change to class names necessitated a Bridgy Publish plugin bugfix, so I was looking in the about page for anything else I might do
#snarfedi understand the 200 argument based on the spec
#snarfedbridgy publish is a pretty unusual use of wm, though, and it definitely does create something as a side effect, so 201 is technically more correct for that
#snarfedi don't really care either way though. happy to change it to 200, or we can see what aaronpk thinks
#GWGsnarfed: I know it is an unusual use case. Although if you are still returning 201...it might break people's sites.
#snarfedeh. i doubt that in practice - by that argument changing it to 200 would more likely break *existing* users - but i hear you. happy to change it if the community agrees.
#GWGsnarfed: I also put in a proposal to allow query variables to be sent with a filter in the webmention plugin to support your extra query parameters.
#GWGsnarfed: I want to improve the Bridgy Publish plugin. When the query variable filter is stable, I'll likely switch to using it. But always a question of how I can do better.
#snarfedlove it! i guess i'd just encourage you to start from itches instead of from tools :P
#GWGsnarfed: I like to know what is possible. I usually make note of it in an issue for future use.
#GWGsnarfed: Same with the webmention plugin. There, the itch was somewhat intangible.
KevinMarks and gluytium joined the channel
#ben_thatmustbeme!tell singpolyma have you make a PR back to gnu social for you webmention fixes yet?
#kylewmsnarfed: your talk was really impressive. Lots of information conveyed in a concise and approachable way
#Loqikylewm: GWG left you a message 37 minutes ago: The Bridgy literature says it returns a 201. It seems to return a 200. 200 is correct according to the webmention specification. The literature seems wrong. http://indiewebcamp.com/irc/2016-06-11/line/1465689739846
#kylewmI hope you had a good time at the conference!
#GWGkylewm: I wrote that poorly. It should return a 200. snarfed and I just discussed it.
#ben_thatmustbemei don't get many webmentions honestly other than bridgy
#GWGben_thatmustbeme: Some stats on how much usage Vouch gets would be interesting.
#ben_thatmustbemeit looks like i only have 8 where i stored the vouch url, but i'm not sure I store them if they are whitelisted, have to check the code
tantek joined the channel
#ben_thatmustbemeyeah, confirmed, i don't keep the vouch if its someone i whitelisted
#ben_thatmustbemeso i actually don't know how many i really receieved
#GWGgRegorLove: I was curious to see a webmention implementation
#gRegorLoveLike for a public webmention form? It would be the same as any other form.
#GWGgRegorLove: Agreed. Just I like to see how people implemented things
#gRegorLoveNot sure I see much benefit, since by design a webmention endpoint can receive posts from anywhere
#GWG"if a Webmention endpoint does accept requests with additional headers, it SHOULD protect itself against Cross-Site Request Forgery (CSRF) attacks. One way to prevent CSRF attacks is by including a CSRF token in a query string parameter of the Webmention endpoint, so that a Webmention sender finds the token when discovering the endpoint."
#gRegorLoveThat's odd. I don't get it; must be missing something.
#gRegorLoveOk, it's in the context of authentication headers or cookies.
#gRegorLoveNot aware of any implementations doing that
#gRegorLoveMicropub question: on /token-endpoint the JWT example uses 'date_issued' then later under verification it uses 'issued_at'. Is one of those incorrect, or is it just whatever is set in the JWT initially?
#tantekgRegorLove: cool. I know copy/pasting each month is a bit of a pain, but I do think the lower barrier to experimenting with the format of the page (without risking changing past pages) is helping with iteration to keep improving them
#j4y_funabashiaaronpk: Just going through webmention.rocks and wondering if test 12 should have 2 more not-webmentions, one in the Link header and an extra anchor before the correct one?
frzn, Erkan_Yilmaz, friedcell, kerozene, Rev_Illo, arthurspooner, Kopfstein and tantek joined the channel
tantek, KevinMarks_, KevinMarks__, KevinMarks and gRegorLove joined the channel
#gRegorLove!tell aaronpk Micropub question: on /token-endpoint the JWT example uses 'date_issued' then later under verification it uses 'issued_at'. Is one of those incorrect? Is it OK to use whatever's in the JWT initially, or is 'issued_at' expected?
#Loqiaaronpk: gRegorLove left you a message 1 hour, 20 minutes ago: Micropub question: on /token-endpoint the JWT example uses 'date_issued' then later under verification it uses 'issued_at'. Is one of those incorrect? Is it OK to use whatever's in the JWT initially, or is 'issued_at' expected? http://indiewebcamp.com/irc/2016-06-12/line/1465770133222
#aaronpkGWG: snarfed: what is the issue with thhe 200/201 for bridgy? the spec only talks about creating "status page" resources and returning http 201, but it makes sense that if something else is created it would also return 201. it should only return 200 if there is no location header returned
#aaronpkben_thatmustbeme: interesting about the http 449 response. wonder if we should add some leniency to the spec to support extensions like that then
#ben_thatmustbemeI figure the extension would specify to override that
#aaronpkgRegorLove: the example that has "date_issued" is a self-contained example of one way to create tokens so doesnt relate to the other example with "issued_at" below
#gRegorLoveaaronpk: So the token verification step, issued_at is optional?
#aaronpkj4y_funabashi: did you encounter a specific case where your suggestion would have had different results with your code?
#emmakgRegorLove: i didn't bother including the issued date in my tokens
#emmaki think its only needed if you want them to expire after a certain time