#dev 2018-07-05

2018-07-05 UTC
eli_oat joined the channel
#
[miklb] what is monocle
#
Loqi Monocle is an open source IndieWeb-enabled reader developed by Aaron Parecki https://indieweb.org/Monocle
eli_oat, renem, [miklb] and [manton] joined the channel
#
ludovicchabant hey there! is there a "proper" way to differentiate (using Microformats) between a normal (long form) blog post and a microblogging update? So far I can only come up with: an h-entry that has no e-content or p-summary, or has some but are the same HTML element as the p-name
petermolnar and snarfed joined the channel
#
KartikPrabhu what is post-type-discovery
#
Loqi Post Type Discovery specifies an algorithm for determining the type of a post by what properties it has and potentially what value(s) they have, which helps avoid the need for explicit post types that are being abandoned by modern post creation UIs https://indieweb.org/post-type-discovery
#
KartikPrabhu ludovicchabant: ^
#
KartikPrabhu what is article
#
Loqi https://indieweb.org/article
#
KartikPrabhu what is note?
#
Loqi A note is a post that is typically short unstructured* plain text, written & posted quickly, that has its own permalink page https://indieweb.org/note
#
KartikPrabhu those too
#
KartikPrabhu mostly articles have a p-name and notes don't
#
ludovicchabant "note" is a Microblog/Mastodon/Twitter-type microblogging update I assume?
#
ludovicchabant on aaronpk's website, microposts have a p-name and e-content (both on the html element containing the status update)
#
ludovicchabant so I think in the post-type-discovery page, it would fall into bullet 15 -- where the p-name _is_ a prefix of the content (it's the same actually)
#
ludovicchabant thanks btw :)
#
ludovicchabant oh cool, I guess I can just use mf2util to run the post type discovery
#
stevestreza.com edited /chat-names (+109) "Add stevestreza.com" (view diff)
#
KartikPrabhu ludovicchabant: that p-name e-content on the same element is to stop the mf2 parsers from implying the name property. This has been corrected in the revised spec
#
tantek__ KartikPrabhu: no it is still useful in PTD
#
KartikPrabhu tantek__: I meant on aaronpk's posts
#
tantek__ hey devs posting to GitHub from their own site (or wanting to) - has anyone thought about, brainstormed, or prototyped how to comment and close an issue?
#
tantek.com edited /IWC_Berlin (+0) "2018!" (view diff)
#
dougbeal tantek__: comment on the original post?
#
KartikPrabhu dougbeal: I think tantek__ wanted to know about the "closing" part
#
ludovicchabant mmh mf2util is not doing PTD the way I expected... I can't get it to spit out a "note" type
#
KartikPrabhu ludovicchabant: I don't think mf2util does that
#
KartikPrabhu but it has functions to do it
#
tantek__ dougbeal, commenting on issues is well understood and implemented. it's the *and close an issue* part that hasn't been figured out AFAIK
#
ludovicchabant KartikPrabhu: ah yeah you're right even its unit tests expect "entry" for notes
#
ludovicchabant oh right, there's a different function for PTD (`mf2util.post_type_discovery` and not `mf2util.interpret_entry`)
#
KartikPrabhu yes, it distinguishes between events and entry type but doe not do full PTD
#
KartikPrabhu or maybe I am wrong :P
#
ludovicchabant yeah it feels like it implemented a bit of ad-hoc PTD before it was formalized, and then the other function was added when the spec started being drafted or something like that
#
ludovicchabant oh well at least now it spits out the info I want :)
#
KartikPrabhu cool
#
ludovicchabant now I have to figure out how to extract the images in the note... feels like I might have to match parsed entries with their raw html element counterpart and fish for `img` tags
#
KartikPrabhu also consider the photo property
#
sknebel oh wow, did twitter un-break their rel=me?!
#
ludovicchabant KartikPrabhu: you mean the `u-photo` class?
#
KartikPrabhu or photo property in the parsed mf2
#
ludovicchabant looks like adding this on the note's pictures makes PTD (or at least mf2util's implementation) change the interpreted type to "photo" post instead of "note"... which isn't what I want (and possibly isn't correct?)
#
ludovicchabant errr what do you mean photo property in the parsed mf2?
#
KartikPrabhu if there is a u-photo in the h-entry, it will show up a as "photo" property for the "h-entry"
#
KartikPrabhu and yes PTD will mean its type is a "photo" post
#
KartikPrabhu https://www.w3.org/TR/post-type-discovery/#algorithm
#
ludovicchabant ah I see, thanks
#
tantek.com edited /GitHub (+245) "/* Features */ reply on an issue can also at the same time close open issue or re-open closed issue, lock/unlock" (view diff)
#
tantek.com edited /issue (+725) "/* Brainstorming */ Close and re-open issues when commenting, Lock and unlock issues" (view diff)
#
tantek__ Looking for additional input / ideas! https://indieweb.org/issue#Close_and_re-open_issues_when_commenting
#
ludovicchabant tantek__ : regarding "Do we need an additional property for a reply?", it's interesting because I'm looking at some of my articles on my website, and I have some long-form articles that comment on/reply to other articles, and those are technically "reply" posts, but there's also note replies, i.e. twitter-like replies to other twitter-like (note) posts
#
KartikPrabhu ludovicchabant: those all count as replies
#
ludovicchabant still, it would be nice to be able to separate from "replies that are notes" and "replies that are articles"
#
ludovicchabant especially for POSSEing (which is what I'm implementing)
#
KartikPrabhu maybe, but that is completely different from tantek__'s close and comment stuff
#
ludovicchabant the former would be posted in their entirety to twitter/mastodon/etc while the latter would just be "{title} {link}"
#
ludovicchabant sure, but there's just that comment about "additional property for a reply"
#
ludovicchabant which might help with what I'm talking about
#
ludovicchabant although I guess in the short term I'll just remove the "in-reply-to" stuff from most of my posts
#
KartikPrabhu I don't see why. You could run PTD to find that it is a reply, and then if it has a p-name it is an article reply else it is a note reply
#
ludovicchabant mmmh good point...
#
KartikPrabhu no need to complicated this for note vs article. But tantek__'s use case is very different
#
tantek__ ludovicchabant my questions about needing an additional property for a reply is for a reply that is also closing an issue
#
tantek__ I'll clarify
#
ludovicchabant yes ignore me -- I was just thinking out loud :)
#
tantek.com edited /issue (+63) "/* Close and re-open issues when commenting */ clarify question about additional reply - specifically for closing/re-opening" (view diff)
cweiske and leg joined the channel
#
tantek.com edited /issue (+407) "/* Close and re-open issues when commenting */ Plain text close re-open thoughts" (view diff)
#
tantek.com edited /issue (+162) "/* Close and re-open issues when commenting */ perhaps model issue closing/re-opening orthogonally, then allow for it to be incuded in a reply" (view diff)
#
tantek.com edited /issue (+140) "/* Plain text close re-open */ example of "Closing issue."" (view diff)
#
tantek.com edited /untag (+48) "/* How to mark up? */ see also micropub remove" (view diff)
leg, iasai and barpthewire joined the channel
#
@nhoizey ↩️ @fvsch Should render as italics in my WebMentions… Bold is more **like this** in Kramdown. (twitter.com/_/status/1014818698610257921)
#
sknebel I noticed earlier that indiewebify.me for some reason could verify rel=me for twitter. do I misremember and that was never broken, or did they do something so that sometimes they do deliver the useful version of their pages (e.g. user-agent depenendent?)
#
Zegnat They did a thing where pages would not load at all without the cookies from their initial JS-only redirect page (or something of that kind) making it impossible for tools like curl to get at the HTML of a given URL
#
Zegnat Possibly they have come to their senses and are actually serving HTML on URLs again
#
sknebel I can't quite figure it out, but something has gotten better then
#
sknebel browser devtools give conflicting info, but curl finds a rel="me" link
eli_oat, [wiobyrne] and [jgarber] joined the channel
#
skippy good morning.
#
Loqi rise and shine!
[jgmac1106] joined the channel
#
[jgarber] !tell aaronpk The whitespace/linebreak conversation from yesterday reminded me of this issue on the microformats-ruby repo: https://github.com/indieweb/microformats-ruby/issues/83
#
Loqi Ok, I'll tell them that when I see them next
#
Loqi [aaronpk] #83 <br> tags are not interpreted as whitespace when converting HTML to plaintext
#
Zegnat [jgarber], we’d really like to solve this in the mf2 spec, so all parsers can do the same. Stripping <br> tags completely is currently correct behaviour per spec.
#
Zegnat See https://github.com/microformats/microformats2-parsing/issues/15
#
Loqi [Zegnat] #15 What should mf2 textContent parsing result in? User expectation vs. DOM specification.
#
Zegnat I took a shot with https://wiki.zegnat.net/media/textparsing.html but that needs revising now as the Python parser has been iterated further. On my todo list for today to get things in sync again
#
[jgarber] zegnat: Ah, yeah. There’s definitely some inconsistencies between the spec and the various parsers.
#
[jgarber] zegnat: I’ll give that issue a read! Thanks for the link.
#
Zegnat I think the inconsistencies stem from the fact that everyone understands textContent to not be sufficient. Sadly HTML’s innerText implementation is very much dependent on CSS rendering stuff. So we have to try and find some middle road to standardise on
#
[jgarber] I was about to suggest Markdown’s whitespace handling as a source of inspiration, then I dug into the docs and found:
#
[jgarber] ```
#
[jgarber] ```
#
[jgarber] Forcing a line-break\s\s
#
[jgarber] ```
#
[jgarber] Next line in the list
#
[jgarber] > To force a line return, place two empty spaces at the end of a line.
#
[jgarber] 🤔
#
Zegnat You probably want to steer clear of that one. Also, you’d need reverse-markdown-whitespace-handling. Where do you break between paragraphs? How do you lay out lists?
#
Zegnat I keep thinking someone must have figured this out already, and I keep coming up blank. (Well, there are HTML-to-Markdown scripts, but I find little documentation on whitespace. And they often only support subsets of HTML.)
#
[jgarber] Right, right. There are n Markdown specs out there and the most official one is John Gruber’s, but that hasn’t stopped the community from creating forks and adding/changing functionality.
#
[jgarber] So yeah, it’s a crapshoot on the HTML => Markdown conversion.
#
[jgarber] Because it’s early and I’m just starting in on coffee: how useful is the plaintext serialization of HTML content?
#
sknebel which one? innerText, which we suggest right now, is horrible to implement, since it depends on CSS etc
#
Zegnat I think the idea is that HTML filtering is hard, so if we can get serialisation for plain text (mf2 p- properties) people are given a standard way to include comments and other content from third parties without worrying about XSS and other things
#
Zegnat Sometimes you just want to know “what does X say” rather than “what HTML is used by X to express it” (which would be mf2 e- properties)
#
[jgarber] Ah, yeah, good points all around.
#
Zegnat There is also the use-case for POSSE: you want to copy the contents of a thing to a place (e.g. Twitter) where rich HTML content is not allowed.
#
[jgarber] In all of these use cases, it seems likely that whitespace _could_ be significant and the intention of the author for clarity, style, etc.
#
[jgarber] e.g. I’m posting a haiku to Twitter
#
skippy replace all non-newline whitespace characters with "&nbsp;" ?
#
Zegnat whitespace is kept the way you author it in your HTML (default by textContent). The problem is that `<p>Wow</p><p>Yes</p>` doesn’t include any whitespace and thus turns into one word: `WowYes`.
#
sixtwothree.org edited /Netlify (+92) "/* Netlify Users */ Add self to Netlify Users" (view diff)
#
Zegnat What I’d actually love to do is sit down with someone who understands the dense CSS spec, and rewrite HTML’s innerText to not require CSS knowledge. Right now I have trouble grogging some of the stuff, example step 3’s explanation of select dropdowns: https://html.spec.whatwg.org/multipage/dom.html#inner-text-collection-steps
#
sixtwothree.org edited /Netlify (+360) "Add Related Tools section" (view diff)
#
Zegnat [jgarber], if you ever feel like implementing innerText and going through the CSS spec for it, hit me up ;)
#
[jgarber] * whistles through teeth *
#
[jgarber] That’s a tall order. 😉
#
[jgarber] Unrelated question: Is it helpful for wiki pages for services like /Heroku to list pricing tiers? That seems like a moving target and I’d imagine the wiki would get out-of-sync fairly regularly.
eli_oat joined the channel
#
sixtwothree.org edited /Heroku (+370) "Add HTTPS Support section and add SpaceHolder to IndieWeb Examples" (view diff)
#
Zegnat I think we have done some price information about different TLDs, to help people getting started on a domain of their own. But I wouldn’t do per-service or per-page pricing info unless you sign-up for keeping it updated
#
sknebel I personally like having a note about free tiers, but other than that agreed
#
[jgarber] Okay, so a generalized “Pricing” section on pages like Heroku, etc. would be useful with links off to the service’s website for details on paid tiers?
#
sknebel not necessary imho, pricing is generally easy to discover from their homepages. I guess maybe if there's oddities to how they bill stuff that need explaining
[wiobyrne] and [snarfed] joined the channel
#
[snarfed] re whitespace, if it helps, bridgy publish currently converts html to text for twitter/flickr, and to markdown for github, and works hard to get the whitespace right. feel free to try it out use it as an example
#
[snarfed] also https://brid.gy/about#ignore-formatting
eli_oat and snarfed joined the channel
#
@jgmac1106 @bradenslen I still think we need to figure out why your webmentions are not displaying on https://ramblinggit.com/. Stop by the WordPress chat room anytime. Also on: https://is.gd/64EWBR (twitter.com/_/status/1014877148157501446)
#
Zegnat [snarfed], is there some sort of spec you follow for the html-to-text conversion?
#
snarfed Zegnat: maybe? not sure
#
snarfed it's mostly https://github.com/Alir3z4/html2text/ with a couple different configurations
#
Zegnat I’ll have a look :) I don’t like a spec addition that says “do what application Y does”. That’s not helpful for encouraging plurality and multiple implementations
#
snarfed oh of course, i'd never propose that. just mentioned it as an example existing implementation
TripFandango, [jgmac1106] and [kevinmarks] joined the channel
#
[kevinmarks] Html2text was originally aaronsw's contribution to markdown
eli_oat joined the channel
#
@rubygems webmention-endpoint (0.3.0): Discover a URL’s Webmention endpoint. https://rubygems.org/gems/webmention-endpoint (twitter.com/_/status/1014898834701520896)
jackjamieson, KartikPrabhu, snarfed, eli_oat and [manton] joined the channel
#
[manton] Any best practices on HTML/JSON response from a Micropub media endpoint? Currently I just return the Location header, but there are some tools (such as Workflow on iOS) where it would be useful to also have the URL in the JSON body.
#
aaronpk hm yeah a recommended response body would be useful
#
Loqi aaronpk: [jgarber] left you a message 4 hours, 23 minutes ago: The whitespace/linebreak conversation from yesterday reminded me of this issue on the microformats-ruby repo: https://github.com/indieweb/microformats-ruby/issues/83
#
aaronpk let me see if there's any notes on that so far
#
aaronpk ah there's the experimental "query the last thing that was uploaded to the media endpoint" feature
#
[manton] Thanks. The spec says "The response body is left undefined" but if other tools are already returning some JSON or HTML, I think I'll copy that as a convenience.
#
aaronpk ah looks like I do two things
[jgmac1106] joined the channel
#
aaronpk if the Accept is text then it returns just the URL
#
aaronpk if the Accept header is json then it returns {"url":"..."}
#
aaronpk documents on the wiki
gRegorLove joined the channel
#
Zegnat conneg++ ;)
#
Loqi conneg has 1 karma in this channel (3 overall)
#
Zegnat sees himself out
#
snarfed conneg--
#
Loqi conneg has 0 karma in this channel (2 overall)
#
snarfed waves
#
myfreeweb [manton]: for photos, my media endpoint returns a big blob of json that should be inserted into the photos/whatever property. it contains... a lot :) links to both jpeg/png and webp versions + blurry tiny webp preview to display while loading + metadata parsed from exif + a color palette :D
#
aaronpk whoa
#
aaronpk can you document that in this new section i'm creating?
#
myfreeweb heh, i'm currently thinking of getting rid of this... and moving the image processing to the client side
#
aaronparecki.com edited /micropub_media_endpoint (+1000) "add brainstorming for standardized response body" (view diff)
#
aaronpk nevermind then :)
#
aaronpk [manton]: https://indieweb.org/micropub_media_endpoint#Response_Body
#
myfreeweb though i just realized that it would result in 3rd party clients not being as good for picture uploading
#
aaronparecki.com edited /micropub_media_endpoint (+6) "/* Response Body */" (view diff)
#
myfreeweb woah matrix is messing up the message order... loqi
#
myfreeweb 's snippet arrived two messages before the link it's for
#
myfreeweb oh it's not a snippet it's a wiki edit
#
myfreeweb sorry, haven't been here in a while, i'm forgetting things haha
#
aaronparecki.com edited /micropub_media_endpoint (+226) "/* Response Body */" (view diff)
#
snarfed so aaronpk i was thinking through some auth details last night for third party services that aren't part of a site itself, like aperture and maybe eventually baffle
#
snarfed i get the auth flow in general, and i get that the service would use an individual site's token endpoint to verify its Bearer token(s)
#
snarfed ...*but* the service needs to know which site a given request/token is for, so it can discover the token endpoint in the first place, right?
#
snarfed so external microsub/micropub/etc services would generally need unique endpoints per site, with the site's domain or user id (like aperture does) or something baked in, so they know which site each request is for, right?
#
aaronpk yeah that's why I have a unique microsub endpoint per user in aperture
#
aaronpk aperture goes and queries your own token endpoint to find out if the token is valid, since aperture wasn't the one that issued those tokens
#
snarfed right, ok. i figured. just checking.
#
snarfed i assume you don't verify tokens on every request though...? do you cache verifications?
#
aaronpk correct
#
snarfed cool thx
#
snarfed we should maybe eventually add guidance on that kind of caching to https://indieweb.org/token-endpoint
#
snarfed i'm guessing short lived, maybe a few hrs or so
#
Loqi totally
#
aaronparecki.com edited /micropub_media_endpoint (+110) (view diff)
#
snarfed and i'm guessing you used user id instead of domain so that people could switch domains and keep their subscriptions? or was that not intentional?
#
aaronpk indieauth IDs aren't limited to domain names, some of them are URLs with paths, and I didn't want to deal with escaping those to use in the microsub endpoint URL
#
snarfed aha right
barpthewire, wagle and [eddie] joined the channel
#
aaronparecki.com edited /videos_about_the_indieweb (+137) "/* 2018 */" (view diff)
#
[manton] [aaronpk] That wiki example for the Micropub response is perfect. Thanks.
#
aaronpk 👍
#
aaronpk I was going to just do the JSON one but then I didn't know how to parse JSON from Workflow so I made the text response too
#
snarfed hey GWG when you get a chance, would you mind pushing out a new version of the bridgy wordpress plugin to remove facebook? people are still (trying to) sign up for bridgy facebook with it
#
snarfed i know you're doing a bigger redesign, but i expect that will take a while
jackjamieson and [jgmac1106] joined the channel
#
skippy Aperture can't find any feeds at https://protonmail.com/blog/ but https://protonmail.com/blog/feed/ is listed in the meta, and that link is a feed (though invalid, according to the feed validator)
#
aaronpk it looks like that site is returning HTTP 403 when XRay fetches it
[kevinmarks] joined the channel
#
skippy weird
leg, [cleverdevil] and snarfed joined the channel
#
KartikPrabhu is it UA sniffing?
#
KartikPrabhu feed looks fine on FF
#
aaronpk I thought I set the UA to avoid this
#
aaronpk there are some pretty aggressive Wordpress plugins tho
#
dgold oh wow.
#
dgold I just realised that when manton said he was setting up m.b. for indieauth he meant as a _provider_ of token/auth
#
[cleverdevil] Yup! Huge influx of Indieauth install base now 🙂
#
skippy aaronpk: reading "quoted rewteets" in Monocle is weird, because the original tweet is not displayed. Is it possible to fetch the quoted tweets for display in Monocle? And is this ultimately a Granary, Aperture, or Monocle issue?
#
KartikPrabhu or XRay
#
snarfed skippy: is this granary's html output or atom? i know quote tweets render ok (fully) in atom at least. example in newsblur: https://snarfed.org/quote.png
#
snarfed if html is worse, i can definitely look
#
snarfed runs to a mtg
#
skippy compare https://skippy.net/images/2018/07/Screen Shot 2018-07-05 at 2-126490.png to https://twitter.com/lmorchard/status/1014942939657768960
#
@lmorchard OH YEAH https://twitter.com/BotEquippedWith/status/1014863986381160449 (twitter.com/_/status/1014942939657768960)
#
@skye_mackinnon @cw1985 thanks for the tweet. Just published w/ link to you on https://www.twinesocial.com/skyemackinnon/fantasy-reverseharem-indieauth/222527078 (twitter.com/_/status/1014948172303622144)
#
skippy snarfed: sorry, I dont understand the question.
#
aaronpk oh yeah I have a pending issue about that in XRay
#
aaronpk but if this is coming from granary then it's a slightly different story
#
aaronpk I assume this is a granary->jsonfeed thing again?
[eddie] joined the channel
#
skippy yes
#
aaronpk can you paste the granary JSON somewhere I can look at?
#
aaronpk hm jsonfeed doesn't have the concept of reposts so there probably isn't much I can do there
#
aaronpk might have better luck with granary via the HTML output
#
aaronpk IIRC all my granary feeds I've added are the HTML format
#
skippy ah. interesting. thanks.
eli_oat joined the channel
#
aaronparecki.com edited /Dropshare (+509) "add links to tweets about adding micropub support" (view diff)
#
loqi.me edited /Indiepaper (+67) "[cleverdevil] added "https://github.com/cleverdevil/Indiepaper-macOS" to "See Also"" (view diff)
eli_oat, [jgarber] and [manton] joined the channel
#
@rubygems webmention-endpoint (0.3.1): Discover a URL’s Webmention endpoint. https://rubygems.org/gems/webmention-endpoint (twitter.com/_/status/1014981278515462144)
#
@DavidWolfpaw ↩️ Similarly, if I had the proper skillset I'd love to push for more methods to be included like OpenID and IndieAuth. (twitter.com/_/status/1014985419237314561)
#
@aaronpk ↩️ Check out the work done on the IndieAuth plugin: https://wordpress.org/plugins/indieauth/ It also works as authentication for the REST API. IndieAuth has the benefit of not needing to pre-register clients, so it is actually useful in a distributed setting like how WordPress works. (twitter.com/_/status/1014985856069693441)
[jgarber] joined the channel
#
GWG snarfed, will do. I am plugging along at Micropub, but that is a worthy quick fix
#
snarfed GWG++ thanks!
#
Loqi gwg has 39 karma in this channel (379 overall)
#
GWG snarfed, I pushed a version of a drastically changed Micropub server to my repo if you are curious.
#
snarfed ok!
#
GWG It is not ready for a PR
#
GWG But I think I might make a Media endpoint a stretch goal
#
snarfed as long as it's a separate PR, go for it!
[jgmac1106] joined the channel
#
GWG It will be, but the rest API is a gigantic change that I can't do in smaller pieces
[cjwillcock] joined the channel
#
snarfed sure! that makes sense. but it sounded like you were doing other changes and refactoring along with that (besides media endpoint). hopefully those can at least be pulled out
#
GWG snarfed, the refactoring was necessary as part of the REST API stuff. I had to change how it worked.
#
GWG So, the authentication had to be separate from the endpoint code.
#
snarfed huh ok
#
snarfed so who do we know who's familiar with wordpress rest api, auth, etc who could review all this effectively? miklb? kraftbj? ...?
#
GWG snarfed, pfefferle reviewed the REST API conversion for Webmention. I think he understands it now.
#
snarfed ah, great! ok
#
GWG snarfed, the big code work is done. The unit tests are taking me a lot more time. Wondering if I should just write new ones
#
GWG Rather than trying to edit the existing ones
#
snarfed hah, wow. if that would be faster than porting the old ones, sure! up to you
#
snarfed as long as you don't lose any, i'm fine with whatever
#
GWG But it is a great lesson for me.
#
snarfed testing++
#
Loqi testing has 10 karma in this channel (29 overall)
#
GWG Either way
#
GWG I am finding it a challenge that will improve my skills
#
[cleverdevil] Wowza - https://twitter.com/viticci/status/1014975736367734784
#
@viticci You can run arbitrary JavaScript in Safari with Shortcuts and I'm going to lose my goddamn mind with this app https://pbs.twimg.com/media/DhXrDEOWAAAjCmb.jpg (twitter.com/_/status/1014975736367734784)
#
[cleverdevil] This is a big deal for iOS automation and the web.
#
GWG snarfed, the hardest part is monitoring what goes through the filters
#
GWG It never would have occurred to me to check that
#
snarfed [cleverdevil]: hah, so ios finally gets android's tasker!
#
snarfed GWG++
#
GWG aaronpk, can you query a media endpoint?
#
aaronpk experimentally yes
#
aaronpk https://indieweb.org/micropub_media_endpoint#Query_the_last_thing_uploaded
#
GWG I see, but what other things are you think?
#
aaronpk maybe list the last several uploads?
#
GWG How about media metadata?
#
aaronpk what sort of metadata?
#
aaronparecki.com edited /micropub_media_endpoint (+122) "/* Query the last thing uploaded */" (view diff)
#
GWG I am looking at the WordPress media api endpoint
#
aaronpk feel free to document it on that page
#
GWG I may.
#
GWG But I know petermolnar displays things like camera and focal length on his site
#
snarfed o_O
#
KartikPrabhu baby steps. no need to handle everything at once
#
unrelenting.technology edited /micropub_media_endpoint (+862) "add sweetroll" (view diff)
#
GWG I am just brainstorming, I haven't built anything yet, and I am not going to so soon
#
GWG But if I do, I might return the WordPress metadata in the response
#
myfreeweb heh I also display camera and focal length and stuff e.g. https://unrelenting.technology/photos/mac-mini
#
myfreeweb just documented the media endpoint stuff related to that on the wiki
#
GWG So, pretty much what I thought
KartikPrabhu and snarfed joined the channel
#
aaronpk trying out bridgy github for the first time
#
aaronpk it's finding stuff but getting errors actually sending the webmentions
#
snarfed aaronpk++ fun!
#
Loqi aaronpk has 151 karma in this channel (1666 overall)
#
snarfed ruh roh ok