#microformats 2012-09-01

2012-09-01 UTC
lgedeon, nonge, tantek, teleject, Cat4D and romainneutron joined the channel
#
arethaozailkfazwjaneiro created /Xiaxue_Slut2718286 (+507) "New page: [http://XiaxueSlut.blogspot.com Xiaxue Bitch] Slut or slattern is a term applied to an individual who is considered to have loose sexual morals or who is sexually promiscuous. The term is..." (view diff)
#
haroldqzsukfwdnlbieniek created /Xiaxue_Bitch9247127 (+501) "New page: [http://Xiaxueslut.blogspot.com Xiaxue] Slut or slattern is a term applied to an individual who is considered to have loose sexual morals or who is sexually promiscuous. The term is gener..." (view diff)
tantek joined the channel
#
tantek edited /Special:Log/block () "blocked [[User:ArethaozailkfazwJaneiro]] with an expiry time of infinite (account creation disabled): Spamming links to external sites" (view diff)
#
tantek edited /Special:Log/block () "blocked [[User:HaroldqzsukfwdnlBieniek]] with an expiry time of infinite (account creation disabled): Spamming links to external sites" (view diff)
#
tantek deleted /Xiaxue_Bitch9247127 "content was spam"
#
tantek deleted /Xiaxue_Slut2718286 "content was spam"
jkridner__, teleject, glennjones, chiui, nonge_, romainneutron, xtof_fr and barnabywalters joined the channel
#
@pretty_girl_014 omegash sana tlaga si papa hcard ka date ni mama dawn haayy...wag sanang hopia
NimeshNeema, tantek, ChanServ, tobyink and chiui joined the channel
#
@FreeWSOdownload Free download: WSO - WP Author hReview (Plugin) http://t.co/7QX9U90f
rasifiel, romainneutron, jkridner___, eighty4_, Phae, singpolyma, tobyink and chiui_ joined the channel
#
aaronpk Anybody know of a web service that takes a URL and outputs a JSON interpretation of the microformats present on the page?
barnabywalters joined the channel
#
barnabywalters tantek: I forked your hcard generator, added some mf2 support and output of minicard and microcard: http://test.waterpigs.co.uk/h-card.html
#
barnabywalters todo: more mf2 support, expose styling for minicard and microcard, finalise minicard/microcard markup
romainneutron joined the channel
#
tantek barnabywalters - that's awesome! looks great.
#
tantek looks like some of my out-of-date code made it in - for tags, we should stop referencing kitchen.technorati as it's been gone for a while :/
#
barnabywalters tantek: yep, not entirely sure what we should do for tags
#
barnabywalters apart from getting people to specify a base url for using…
#
tantek perhaps even just switch it to <span class="category">tag1</span>, <span class="category">tag2</span> etc.
#
barnabywalters e.g. for me it would be http://waterpigs.co.uk/tags/
#
barnabywalters yep, that's a simpler approach
#
tantek also - tag markup doesn't show up in the Code textarea ;)
#
tantek so something broke
#
tantek actually - the code areas are kinda broken in general
#
barnabywalters ah ha! weird. I'll have a look at that
#
tantek hmm - they're not actually showing anything I type in the fields
#
tantek except the mini/micro versions show *only* a p-name with the given-name inside
#
barnabywalters the main one isn't, no
#
barnabywalters that is very strange
#
tantek and the microcard should have h-card as the root class name - not p-name
#
barnabywalters I have barely changed your original code
#
tantek looks like you've got some more coding to do :)
#
barnabywalters tantek: as always :)
#
tantek also - I wouldn't bother adding the minicard / microcard class names
#
barnabywalters tantek: my plan was to give example css for the mini/microcards
#
tantek is everything inside the minicard a span?
#
tantek I'd just distinguish with root element
#
barnabywalters although I suppose the microcard could actually be styled just using the fact it's inside an <a>
#
barnabywalters ah, you got there before me :)
#
tantek e.g. full - <div>, mini - <p>, micro - <span>
#
tantek because those are likely to make sense in context for their different sizes / extents
#
barnabywalters yep. that is cleaner
#
barnabywalters is using <p> for mini semantically valid? it's not a paragraph of text
#
tantek yeah - the definition for <p> is actually quite broad in HTML5
#
tantek it would work for the minicard
#
barnabywalters I think I glossed over <p> last time I read the spec
#
barnabywalters okay, well that's a good todo list for that
#
barnabywalters got to go and eat now
#
tantek but yeah - in general, go through each field, and make sure changing it actually changes the produced code :)
#
tantek I would keep things like middle name in all three
#
tantek if someone enters a middle name, I think they intend for it to be in the "full" name
#
barnabywalters tantek: yes, that's true. I have come across people who do that
#
tantek consider doing only uf2 output for the minicard and microcard
#
tantek will keep them even more miniature
#
tantek the microcard makes sense to demonstrate the simple root-class-name-only case
#
tantek it should always just have class="h-card" at the top, and no other class names
#
barnabywalters pretty sure I'm already doing that — haven't added uf2 to the full one either
#
barnabywalters off to eat, bbiab
tantek_, barnabywalters and nonge joined the channel
#
aaronpk barnabywalters: where did your microcard markup go on your site! It was looking so good the other day!
#
tantek_ aaronpk - I see it on this page: http://test.waterpigs.co.uk/activity/
#
tantek_ is now working on how to incorporate that markup pattern into the implied properties algorithm of #uf2
#
barnabywalters aaronpk: yep, it's in the h1, showing off it's scaling ability
#
barnabywalters I'm way overcommitted at the mo, but when I get round to it (and the markup is finalised) I'll be rolling it out all over my site, and in indieweb @anywhere
#
barnabywalters which still needs a good name
#
aaronpk ahh I was looking at the "notes" page
#
barnabywalters not sure how best to expose it yet
#
barnabywalters activity is quite hidden
#
aaronpk tantek_: is this an appropriate h-card markup? I have v1 in here too: https://gist.github.com/3585643
#
barnabywalters aaronpk: I am no expert, but tantek_'s comment above suggests that "it should always just have class="h-card" at the top, and no other class names"
#
aaronpk where url is implied?
#
barnabywalters aaronpk: I think the idea is to wrap it in a span, although that seems a bit bloated
#
barnabywalters it does make sense from a styling point of view though
#
barnabywalters e.g. if we are creating minicards and microcards based on the root element of the h-card, a microcard might not have a url associated, and so the root element can't be an <a>
#
aaronpk true but if there is a url, then it should be ok for the root element to be an <a> with a url, right?
#
aaronpk <a href="" class="h-card"><img src="" alt="" /> Name</a> <-- minimal example
#
barnabywalters well, I want to be able to use and style the microcard markup universally
#
barnabywalters ah, yep, perhaps the implied syntax works out
#
aaronpk tantek_: How about this as the most minimal h-card? <a href="http://aaronparecki.com" class="h-card"><img src="http://aaronparecki.com/images/aaronpk.png" alt="" /> Aaron Parecki</a>
#
tantek edited /microformats-2-implied-properties (+1247) "/* additional markup patterns to consider */ provide finer grouping / explanation, and add root element with one child and text based on Barnaby Walters's example in the wild" (view diff)
#
tantek edited /microformats-2-implied-properties (+4) "cluster rejected with respective pattern" (view diff)
#
tantek_ ok I'm back - was deep in editing context :)
#
tantek_ barnabywalters - based on your markup pattern I came up with another implied property pattern to consider for uf2: http://microformats.org/wiki/microformats-2-implied-properties#root_element_with_one_child_and_text
#
rasifiel Oooh... More different semantic markup(
#
tantek_ rasifiel - yeah, from actual web publishing patterns
#
tantek_ aaronpk, checking your gist
#
tantek_ so you have two choices:
#
tantek_ 1. if you want backcompat with hCard v1, you need a separate root element (e.g. span) surrounding all the elements with properties.
#
Loqi http://loqi.me/4yV
#
tantek_ 2. or if you're ok with only h-card support
#
aaronpk i think i'll go with only h-card support on this one
#
tantek_ then you can drop class names vcard url u-url u-photo photo and the <span class="p-name fn">, assuming you like the markup pattern I *just* documented for barnabywalters
#
tantek_ this one: http://microformats.org/wiki/microformats-2-implied-properties#root_element_with_one_child_and_text
#
tantek_ OR - did you have a specific styling reason (or otherwise) for the child <span> ?
#
tantek_ (trying to understand if that's an actual use-case, or if you just put the span there as a hook for those properties)
#
barnabywalters tantek_: what if the person represented by the microcard has no URL?
#
barnabywalters (this is theoretical for h card generator btw)
#
aaronpk the <span> is there to separate the "@" from the name when I use nicknames
#
aaronpk I'll remove the <span> when I use a real name
#
tantek edited /microformats-2-implied-properties (+370) "/* root element with one child and text */ note gist from aaronpk as another possibility to consider" (view diff)
#
tantek_ barnabywalters - if the person represented by the microcard has no URL, then the root element is a span instead of an <a>
#
barnabywalters fab. thought as much
#
tantek_ aaronpk - do you have a publishing example of that "separate the "@" from the name when I use nicknames" ?
#
tantek_ I think I remember seeing it but I'd prefer to link directly to a publishing example in the wild
#
aaronpk tantek_: yea, let me push the code up, I just made some changes
#
tantek_ BTW re: h-entry p-name vs. entry-title, all I saw is more support and no objections, so I'm going to make that edit to h-entry in uf2.
#
tantek_ aaronpk - do you support h-entry in your markup yet? if so, you may want to wait a sec...
#
aaronpk it's ok, pushing is easy and I prefer to have smaller related changes in my commit history anyway
#
aaronpk tantek_: http://aaronparecki.com/2012/230/reply/1
#
barnabywalters okay, just pushed a few fixes to my h-card generator, including universal styling support (styling based on root elements)
#
tantek edited /microformats2 (+291) "/* h-entry */ drop p-entry-title (p-name is more reflective of actual use), switch to p-summary and e-content. "entry-" is no more." (view diff)
#
tantek_ ok, http://microformats.org/wiki/microformats2#h-entry updated per the discussion this week. h-entry markup just got simpler / shorter :)
#
barnabywalters awesome. most of my h-entries are already updated, but I forgot about e-content (think I used p- by mistake)
#
tantek_ barnabywalters - p-content is probably correct for simple *text* notes :)
#
barnabywalters true, not sure I even have any of those (even notes usually contain some markup)
#
tantek_ nah - usually text notes contain only *auto-generated* markup
#
tantek_ so I prefer to simply syndicate out the *text* because that's what I authored.
#
tantek_ any auto-generated markup is something my code is doing automatically on my server and something I may update (like we are right now on all our sites with the h-cards!)
#
barnabywalters good point
#
tantek_ aaronpk - in http://aaronparecki.com/2012/230/reply/1 I don't think you need class="u-photo photo" on h-card on brennannovak, assuming we like the implied markup pattern I documented: http://microformats.org/wiki/microformats-2-implied-properties#root_element_with_one_child_and_text
#
tantek_ for use cases where you only have the Twitter alias (often a nickname), consider indicating that explicitly in the markup as well: <a class="h-card" href="http://twitter.com/dreev">@<span class="p-name p-nickname">dreev</span></a>
#
tantek_ because then a smarter higher level application can reason better about it - and realize, oh, you just have a placeholder p-name because you only know their p-nickname which happens to be the same thing.
#
tantek edited /microformats-2-implied-properties (+138) "/* root element with one child and text */ note aaronpk's minimal example in IRC" (view diff)
#
tantek edited /microformats-2-implied-properties (+129) "/* root element with one child and text */ add aaronpk minimal markup example with real values" (view diff)
#
tantek_ has caught up with the discussion above :)
#
barnabywalters Pushed the suggested css to http://test.waterpigs.co.uk/h-card.html
#
barnabywalters it's working quite nicely now, producing much better markup
#
tantek_ is now updating Falcon per the h-entry spec updates.
#
tantek_ realizes we just started bim 5 of the year.
#
barnabywalters woo newcalendar
elf-pavlik joined the channel
#
@FreelanceWP WordPress › Author hReview « WordPress Plugins http://t.co/nkcYlRZ2
#
tantek_ banabywalters - I just checked my code and apparently I was using e-entry-content (now e-content) for my *text* notes as well.
#
tantek_ (in Falcon)
#
tantek_ What I don't know is if that was a deliberate design decision at the time I added that support, or if it was just an accident.
#
tantek_ maybe this is a larger indieweb POSSE question
#
barnabywalters or perhaps more of a 'how do we want potential h-feed parsers to use our content' question
#
tantek_ should we POSSE out markup (including potentially auto-generated markup) for text notes, or POSSE out plain text without markup for text notes?
#
barnabywalters tangent: http://test.waterpigs.co.uk/h-card.html updated to latest version, inc. uf2 support for full h-card
#
tantek_ it's a combination POSSE and ActivityStreams question
#
tantek_ I think I'd rather almost assume that feed parsers / feed readers are "dumb" and will do the wrong thing with auto-generated markup (links, embeds, people references)
#
barnabywalters tantek_: that would be the safest route, I think
#
tantek_ and thus err on the side of syndicating more "complete" markup for a note
#
tantek_ so that "dumb" feed parsers/readers can do more easily/quickly without having to try to be smart
#
barnabywalters it still allows intelligent feed parsers to do useful things with content — possibly even helping them provide more functionality
#
tantek_ yes that's an even better opint
#
tantek_ point
#
tantek_ because that richer note markup can also include h-card references to people with specific indieweb URLs etc.
#
barnabywalters I was thinking that and XFN data
#
tantek_ and therefore even *smarter* feed parsers/readers can do more things with more richly marked up notes.
#
tantek_ ok, so it sounds like richer marked up notes are better for both *dumb* and *smarter* feed parsers/readers
#
tantek_ I suppose we could provide plain text versions of notes in a p-summary
#
barnabywalters agreed
#
barnabywalters might be a bit redundant
#
barnabywalters I say we wait until someone actually develops an h-feed reader/parser
#
tantek_ well I've been thinking of providing plain text / abbreviated summaries
#
barnabywalters of notes?
#
tantek_ e.g. right now when Falcon POSSE's out to Twitter, it sends an automatically ellipsed version of the note
#
tantek_ and that has to be plain text
#
tantek_ and is often shorter / abbreviated
#
barnabywalters sure — similar to the titles (note presentational title semantic being used!) of my activities
#
tantek_ so I'm thinking a structure like <p class="e-content"><span class="p-name p-summary">sentence 1. sentence 2 up to 120 chars.</span> rest of the note content maybe with an h-card like <a class="h-card" href="http://aaron.pk">Aaron Parecki</a>. Even more content.</p>
#
barnabywalters does that not prevent useful markup from being present in the first 120 chars?
#
barnabywalters prohibiting, for example, "@aaron.pk bla blah blah"
#
tantek_ where the p-summary literally only has whole sentences up to ~120 characters, and that's both what gets syndicated to Twitter (due to their content limits), and likely works better as a working name for the post.
#
tantek_ here's how I'd do that
#
tantek_ <p class="e-content"><span class="p-name p-summary">sentence 1 refers to @<a class="h-card" href="http://aaron.pk"><abbr title="Aaron Parecki">aaronpk</abbr></a>. sentences 2…n up to 120 chars.</span> more text note content / markup… </p>
#
tantek_ that usage uses the 2nd markup pattern proposed here: http://microformats.org/wiki/microformats-2-implied-properties#root_element_and_one_child
#
tantek_ <a class="h-card"><abbr title>
#
tantek_ that pattern has the nice effect of allowing the transmission of the Twitter @-name reference in plain text, while linking to Aaron's indieweb site in both the indieweb original of the note, and in the h-entry syndicated copies.
#
barnabywalters ah, so p- just means 'strip tags' , not 'source'
#
tantek_ p- means "text content only"
#
tantek_ http://microformats.org/wiki/microformats-2#naming_conventions_for_generic_parsing
#
tantek_ e-* for "element tree", "embedded markup", or "encapsulated markup" - all of those make sense
#
barnabywalters yep, I see how it works now
#
tantek_ ok Falcon is now updated with the new h-entry markup as well.
#
barnabywalters is there a semantic difference between embedded and encapsulated mkup?
#
barnabywalters woo! now we just need a h-feed parser
#
barnabywalters getting late over here, I'm signing off
#
barnabywalters good night
#
Loqi see you in the morning!
#
barnabywalters cheers Loqi
#
Loqi dude
#
tantek edited /microformats2 (+397) "/* Examples in the wild */ added support for h-entry to tantek.com" (view diff)
#
tantek_ !tell barnabywalters when you've updated http://waterpigs.co.uk/ to support h-entry explicitly, update the entry about it explicitly here: http://microformats.org/wiki/microformats2#Examples_in_the_wild
#
Loqi Ok, I'll tell them that when I see them next
#
tantek_ aaronpk - so it looks like you're using h-card on your permalink URLs, but no h-entry (yet?)
#
tantek_ nor "hentry" it seems, though you have a class name / use of "entry-content" - e.g. on: http://aaronparecki.com/2012/244/note/1
#
@TasksMe Open Question: what is hcard integration in seo?: I want toi know about hcard integration. http://t.co/nkgM6I93
#
tantek edited /microformats2 (+170) "/* Examples in the wild */" (view diff)
#
tantek edited /hcard-examples-in-wild (+436) "more feedback on an hCard" (view diff)
brennannovak and tobyink joined the channel