#techlifewebGWG: Well, that's something Im thinking about. At the moment, turning from template issues and moving to thinking about posse/pesos ease of use
#tantekYes I do. But it is so cumbersome for me to manually tag more than a few people that I gave up on person-tagging a photo with ~50 people
#tantekIt would probably take me more than an hour. So I reached a practical limitation in my current approach, and must design+write more code to improve the situation
colintedford joined the channel
#GWGDoes anyone have an interface that can tag so many easily?
#tantek.comedited /meta_http-equiv_status (+350) "/* Defined Values */ add explicit extension of meta http-equiv for "Status", for something citable and expandable beyond the spec summary" (view diff)
#tommorristantek: np. was a nice little "get back into writing Python" invitation
snarfed1 joined the channel
#tommorrissometimes, when in the middle of a large, fiddly mess, building a very small, simple thing and getting it out there is a win
#tantekindeed. and your implementation has helped inform me about how I can further improve the spec
#tommorristantek: it's kind of a proof of concept - if you are using Django, there's not really much of a need to be able to set status codes with a http-equiv. but one implementation is better than zero.
#tantekkylewm implemented it as part of Redwind's Webmention receiving processing
#tommorriseventually, shouldn't be too hard for someone to build, say, an nginx or apache plugin which is a more useful place to put such code for a static site
#aaronpkthis site actually has great examples of different card-style previews of posts https://hi.co/
wolftune joined the channel
#tantek.comedited /meta_http-equiv_status (+799) "/* Consuming Application Requirements */ split general consuming application and specific to webmention receiver" (view diff)
#tommorristantek: I think you misunderstand. the Django middleware isn't a consumer implementation. it's a proof-of-concept server implementation.
#tommorrisIt checks all responses and rewrites the status code to match the http-equiv if it is there
#tommorristantek: the only thing I'm slightly worried about is if it got widely adopted, it's something that would have to be scrubbed from any user input. if someone were to submit a 410 meta http-equiv, then Google were to index it...
#tommorrisas in, it were in the body. then the server picks that HTML up and starts serving that page as 410 Gone. SEO funtimes!
#aaronpkseems like it might be best to only look for the meta tag in the html head?
wolcen, KevinMarks and benward__ joined the channel
#tommorristhere's also some issues with how to handle streaming. will need to find or write a streaming HTML parser. if someone is sending a very large document, would need to handle that. (though if you are serving a 401, it probably ought to not be huge.)
#KevinMarksthe meta should be in the first few packets though
#KevinMarkswasn't there some limit on how close to the top meta should be? I suppose someone could have made a facebook meta clump wiht a data url for the photo, but that would be daft
#tantekaaronpk - correct, the spec says only in the head
#tantektommorris: I guess I didn't understand the default behavior aspect of the django implementation you made
#tanteke.g. does HttpEquivStatusMiddleware get automatically instantiated and process_response gets called?
#tommorristantek: yep, by Django/WSGI. You could use it to serve HTML files. (In practice, you generally don't serve static files from Django.)
#tommorrisDjango Middleware is used specifically for doing stuff like parsing responses and changing them. I've got some Django middleware to make it so that when you go to a .json URL (or send an Accept header asking for JSON), it parses the HTML with mf2py and returns the JSON
#tantek.comedited /meta_http-equiv_status (-56) "/* Implementations */ fix django implementation description, subsections for Server Requirements, Consuming Application Requirements" (view diff)
#tantekit doesn't drop punctuation well enough and escapes into ugly things
#tantekkevinmarks I often use a list at the top of my document to summarize and then have headings with the same text, so fragmentions would fail at that
#tantekwhere I want to link to the latter heading, not the first instance of the heading text
#KevinMarksah, yes that is an issue with the wiki too
#LoqiThat! (or "that ^" or "that ^^^") is a rarely seen reply often emphasizing agreement with a This post, but sometimes[1] merely emphasizing agreement with a previous reply http://indiewebcamp.com/that
#Loqi[Tantek Çelik] h-entry is a simple, open format for episodic or datestamped content on the web. h-entry is often used with content intended to be syndicated, e.g. blog posts. h-entry is one of several open microformat draft standards suitable for embedding data in ...
#aaronpkin this case there's no mf2 on that page. it's just a matter of getting the mf1 hentry to be parsed into something reasonable.
#tantekaaronpk: right, the implied p-name does not apply to mf1
#aaronpki don't actually understand where the … came from
#GWGI always run to my phone when I hear WordPress
#tantekgotta fix this overimplying of properties for compat roots
#tantekthat's clearly the root (so to speak) of so many problems
#aaronpkgood thing we have loqi being a canary for these things then
#gRegorLovephp-mf2 currently just converts legacy mf1 to mf2, then runs the mf2 parsing. Shouldn't be too difficult to keep track of the mf1 state. I'll need to read up on backcompat
cweiske joined the channel
#tantekgRegorLove: the mf2 parsing algorithm requires keeping track of whether a root (or property) class was from a backcompat conversion or not
#KevinMarkshah, because of the local caching by Peter Wilson, refreshing the page just flushes the previous tweet id, not anything else
#KevinMarksso I can create a section break that way
friedcell, wolcen, Leeky and loic_m joined the channel
#urantherwhat's the opinion about Google's Accelerated Mobile Pages (AMP)? Is this something that will disappear in a couple of years?
mlncn joined the channel
#KevinMarksthat's a tricky one - as it's open sourced, they can't sunset ti completely, but if they stopped favouring them in search then the point goes away
#KevinMarksI'm hoping they will take some of the criticisms of replacing native elements unnecessarily on board
#sknebeldoesn't it right now rely on their CDN to load components?
#sknebelthat's another weakpoint, depending on how strict implementations are with that
#urantherSo it may be worthwhile to implement on my website, then?
#urantherI just don't hear much about it - so it seems like the wider web community is not caring about it. But maybe I'm not listening in the right places.
#gRegorLoveuranther: I think larger publishers will definitely be adopting AMP if they haven't already. Hosted Wordpress.com supports it as well, I believe.
#gRegorLoveuranther: For me, it's not worth implementing on my site, but maybe it is for you.
#LoqiAMP is Accelerated Mobile Pages, a Google-led project that speeds up a subset of HTML through caching and dependencies on google: http://indiewebcamp.com/AMP
#uranthergRegorLove: I agree, it seems mostly helpful for content-heavy websites like NYTimes. I was thinking it would be a fun exercise to see how fast I can get my (content-light) website to load.
#gRegorLoveYeah, it's definitely a (kind of odd) response to JS-heavy, add-heavy sites.
#aaronpkthere's a conversation about linkedin happening in another chat room i'm in, (can I call Slack channels a chat room? is that allowed?) and I am wondering if there's some part of the appeal of linkedin that we could bring to indieweb resumes
#aaronpkone of the benefits for me of having my resume on linkedin is that they've gone to the trouble of figuring out how to lay things out and where to put things, so I just fill in the details
#aaronpkwhen i'm staring at a blank HTML page, there's a lot more cognitive stress about what to write first, that it makes it harder to actually put my resume info on my site
#aaronpkso I wonder if there was some sort of resume template I could copy, if that would make it easier to do that on my site
#[kevinmarks]We did have h-resume that was based on LinkedIn among others
#aaronpkright but that's not the same as UI/display guidelines
#[kevinmarks]True that. Playing with h-review and indiecards was interesting there.
Lancey, cweiske and danlyke joined the channel
#tantekaaronpk that analysis about cognitive stress is spot on
#tantekand it's true for resumes, for 'contact' (aboutme style) sites, etc.
#aaronpki guess that's why people like the little js widgets you can drop into a page too
#tantekok I'm starting to get annoyed at the FB practice of people "tagging you" in a post with a location masquerading as a checkin which FB then notifies you as "So and so said they were with you at location"
#aaronpkheh oops. foursquare fixed that already too.
#tantekwell FB has that distinction too - you can mention people in the prose of the checkin, instead of "tagging" them in it
#aaronpkhm so something about their interface encourages people to use the tagging feature "inappropriately"?
#tantekit's just that people (or their reaction to the UI? maybe mobile UI? mobile app UI?) seem to actively choose to use "tag someone in this post" as a FBism FYI
#Loqipage is short for webpage on the IndieWeb, and can by default be marked up with h-entry (like body class=h-entry) for link-preview purposes unless the page serves a more specific purpose like a homepage or an event https://indiewebcamp.com/page
#tantekA Facebook Page is a form of identity on Facebook that has its own profile like other identities, and a set of administrators who can take action on Facebook as the identity of that Facebook Page.
#tantekgRegorLove: still about 80-90% legit uses for me
#gRegorLoveI actually don't recall if that means they can't tag me in checkins at all, or if I have to approve it before it appears publicly. Might be the latter.