#microformats 2015-05-29

2015-05-29 UTC
KartikPrabhu, fuzzyhorns, KevinMarks and KevinMarks_ joined the channel
#
tantek KevinMarks: working on adding your rel-urls hash to microformats2-parsing
#
tantek in case you're up and want to review later
gRegorLove joined the channel
#
tantek edited /microformats2-parsing (+260) "[[microformats2-parsing-brainstorming#more_information_for_alternates]] part 1, add rel-urls key, update parsed rels example" (view diff)
#
tantek edited /microformats2-parsing-brainstorming (+196) "note changed opinion from textContent to text" (view diff)
#
tantek edited /microformats2-parsing (+79) "[[microformats2-parsing-brainstorming#more_information_for_alternates]] completed, pt 1 [[microformats2-parsing-brainstorming#more_information_for_rel-based_formats]] add rel-urls key, update rels ex." (view diff)
#
tantek edited /microformats2-parsing-brainstorming (+32) "more information for alternates - incorporated" (view diff)
#
tantek edited /microformats2-parsing (+487) "/* parse a hyperlink element for rel microformats */ part 2 [[microformats2-parsing-brainstorming#more_information_for_rel-based_formats]] rel parsing algorithm additions" (view diff)
KartikPrabhu joined the channel
#
tantek edited /microformats2-parsing-brainstorming (+56) "incorporated more information for rel-based formats" (view diff)
#
tantek KevinMarks - microformats2 rel parsing and example updated per your proposal (and I AFAIK implementation) - please review: http://microformats.org/wiki/microformats2-parsing#parse_a_hyperlink_element_for_rel_microformats
#
tantek !tell kylewm did you have a chance to try implementing this in mf2py? http://microformats.org/wiki/microformats2-parsing-brainstorming#Nested_h-.2A_objects.27_.22value.22_property I want to incorporate it into microformats2-parsing but double-check implementability first.
#
Loqi Ok, I'll tell them that when I see them next
#
tantek !tell aaronpk did you have a chance to try implementing this in phpmf2 http://microformats.org/wiki/microformats2-parsing-brainstorming#Add_meta_http-equiv_to_microformats2_parsing_model ?
#
Loqi Ok, I'll tell them that when I see them next
#
kylewm tantek: I haven't yet but will do
#
Loqi kylewm: tantek left you a message 2 minutes ago: did you have a chance to try implementing this in mf2py? http://microformats.org/wiki/microformats2-parsing-brainstorming#Nested_h-.2A_objects.27_.22value.22_property I want to incorporate it into microformats2-parsing but double-check implementability first.
#
tantek edited /microformats2-parsing-brainstorming (+191) "note status for a few issues" (view diff)
#
tantek thanks kylewm
#
tantek while you're in there, take a look at http://microformats.org/wiki/microformats2-parsing-brainstorming#Canonicalization_of_datetime_output also to see what if any changes you need to make to mf2py for that
#
tantek kylewm: and it looks like most of the resolved issues could use your implementation experience too, e.g. http://microformats.org/wiki/microformats2-parsing-issues#uf2_children_inside_a_classic_microformats_root_class_name
#
tantek I'll see if I can explicitly note the ones that need it (resolved, need implementation experience)
#
tantek edited /microformats2-parsing-issues (+384) "add status and need implementation experience for a few" (view diff)
#
tantek ok noted
#
tantek having KevinMarks try implementing his proposals to make sure they did what he needed for his use-cases was very handy
#
tantek that kind of implementation checking of resolutions something we should continue with - and it will also help show the priority of brainstorms/issues - if people are willing to try implementing the resolutions.
KevinMarks joined the channel
#
kylewm tantek: sanity check -- is "value" meant to be inside or outside of "properties": {}
KevinMarks_ joined the channel
#
kylewm nm got it, outside. i need to fix that example to reflect that
#
kylewm !tell tantek I implemented the "name" for p-* h-* and "url" for u-* h-* pretty trivially... supporting the second part (first dt-* or first e-*) would require some refactoring https://github.com/kylewm/mf2py/commit/63cd5882f53d8972c2b007fec93bf893c92acd1b
#
Loqi Ok, I'll tell him that when I see him next
#
tantek cool - thanks for fixing example
#
Loqi tantek: kylewm left you a message 2 minutes ago: I implemented the "name" for p-* h-* and "url" for u-* h-* pretty trivially... supporting the second part (first dt-* or first e-*) would require some refactoring https://github.com/kylewm/mf2py/commit/63cd5882f53d8972c2b007fec93bf893c92acd1b
#
tantek kylewm - we came up with the second part (for dt-* and e-*) so we at least had an idea what those *should* do
#
tantek however, I don't think we have any actual use-cases for them
#
tantek that is, for "dt-* h-*" markup or "e-* h-*" markup - do you know of any?
#
tantek perhaps we should have those fallback to "p-* h-*" treatment until there's a use-case for doing otherwise
#
tantek I'd rather avoid extra work for something without use-cases
kez joined the channel
#
tantek Loqi?
#
Loqi woot!
glennjones joined the channel
#
tantek !tell kylewm commented: http://microformats.org/wiki/index.php?title=microformats2-parsing-brainstorming&diff=64999&oldid=64998 - if that sounds good to you, please +1 it. And we'll see what tommorris, glennjones and other parser devs think.
#
Loqi Ok, I'll tell them that when I see them next
tantek joined the channel
#
tantek !tell aaronpk looks like Loqi stopped reporting wiki edits to this channel sometime after ~2015-10-28 22:30 PDT - compare IRC logs to http://microformats.org/wiki/Special:RecentChanges
#
Loqi Ok, I'll tell them that when I see them next
KartikPrabhu, Left_Turn, eschnou and KevinMarks joined the channel
#
glennjones For parser developers I have added a “textContent browser test” http://testrunner-47055.onmodulus.net/textcontent/ to use as a comparsion parsers output
#
glennjones !tell kylewm looked into textContent output issue from yesterday and your parse is correct and the test is wrong. I will go through the tests this morning correcting any other textContent errors I can find
#
Loqi Ok, I'll tell them that when I see them next
elf-pavlik, adactio, KevinMarks, kez_ and kez joined the channel
#
@AllTheTwits is proof-reading microformats book (twitter.com/_/status/604262894667194369)
glennjones, elf-pavlik, KartikPrabhu and TallTed joined the channel
#
@seldo2 Recipes are one of those use-cases where stuff like microformats seem to have really clear value. #jsconf (twitter.com/_/status/604284961483575296)
#
aaronpk hmm
#
Loqi aaronpk: tantek left you a message 9 hours, 39 minutes ago: did you have a chance to try implementing this in phpmf2 http://microformats.org/wiki/microformats2-parsing-brainstorming#Add_meta_http-equiv_to_microformats2_parsing_model ?
#
Loqi aaronpk: tantek left you a message 8 hours, 10 minutes ago: looks like Loqi stopped reporting wiki edits to this channel sometime after ~2015-10-28 22:30 PDT - compare IRC logs to http://microformats.org/wiki/Special:RecentChanges
#
aaronpk !tell tantek I've never touched the PHP parser internals... most I've done is added relative URL resolving
#
Loqi Ok, I'll tell him that when I see him next
globbot, gRegorLove, KevinMarks__, tantek and eschnou joined the channel
#
@deewu FD #2: I am way over my head. I don't code, have super basic HTML skills. Microformats what? I don't mind learning, it's just a lot. (twitter.com/_/status/604344192098254848)
#
tantek good morning #microformats!
#
Loqi tantek: aaronpk left you a message 3 hours, 7 minutes ago: I've never touched the PHP parser internals... most I've done is added relative URL resolving
#
tantek how goes the parsing issues / brainstorming discussions?
#
kylewm o/
#
Loqi kylewm: tantek left you a message 12 hours, 15 minutes ago: commented: http://microformats.org/wiki/index.php?title=microformats2-parsing-brainstorming&diff=64999&oldid=64998 - if that sounds good to you, please +1 it. And we'll see what tommorris, glennjones and other parser devs think.
#
Loqi kylewm: glennjones left you a message 9 hours, 35 minutes ago: looked into textContent output issue from yesterday and your parse is correct and the test is wrong. I will go through the tests this morning correcting any other textContent errors I can find
#
kylewm hi tantek, two things. I was doing some/most of the timezone normalization in mf2py, which caused lots of test suite failures, so i rolled it back for now. but it is very easy to add back in. i'd be interested in how easy or difficult it is for other parsers to change. in mf2py at least we had to do a lot of machinations to get dates to parse correctly at all, by far the trickiest part of the spec
#
tantek wait what that's a different issue
#
kylewm 2; totally agree with rationale to wait for use cases of e-* h-* and dt-* h-*, but why fallback to p-* h-* instead of the old behavior
#
tantek yeah you're right
#
tantek old behavior for e-*/dt-* h-* makes more sense
#
kylewm oh cool that was easy :)
#
tantek and especially for e-* - that's likely close to what's desired
#
tantek edited /microformats2-parsing-brainstorming (+248) "provide details of updated proposal, e-* h-* behaves as before, dt-* h-* just normal dt-* parsing" (view diff)
#
tantek kylewm: please review ^^^
#
kylewm tantek: not sure which indentation level I should have +1'ed
#
tantek not sure why wiki edits are not showing through
#
tantek sometimes
#
aaronpk oh yeah i was going to mention that, everything looks fine on my end
#
tantek checks the logs
#
tantek yup - logs agree with what I see
#
aaronpk it is a UDP packet from the microformats.org server to Loqi
#
aaronpk so it is entirely possibly that it just gets dropped sometimes
#
tantek aaronpk - so Loqi is not seeing the packet?
#
aaronpk I assume not, if it's not showing up her
#
aaronpk e
#
tantek or are you seeing the edits from Loqi and we're not?
#
aaronpk I don't have a lot of visibility other than what shows up in IRC
#
tantek ok got it
#
tantek kylewm: then we're in agreement - does your implementation agree also?
#
tantek KevinMarks: do you want to weigh-in on this? either "sounds good" "meh" or "no" ?
#
tantek this summary in particular: http://microformats.org/wiki/microformats2-parsing-brainstorming##Update%202015-05-29
#
kylewm yep, that's what my implementation does... not merged yet, but branch is here https://github.com/kylewm/mf2py/tree/nested-value
#
tantek I like this way of doing change control to microformats2 parsing
#
aaronpk is scared
#
aaronpk can someone add test cases for whatever is being changed and I can go poke at the PHP parser to update it?
#
tantek where 1) collect issues/brainstorms/proposals, 2) let people try implementing on their own incentive, 3) when there is consensus on a proposal AND proof of implementability, accept proposal
#
aaronpk i haven't really been following
#
tantek s/proof/proof & incentive for
#
Loqi tantek meant to say: where 1) collect issues/brainstorms/proposals, 2) let people try implementing on their own incentive, 3) when there is consensus on a proposal AND proof & incentive for of implementability, accept proposal
#
tantek aaronpk - I believe KevinMarks added test cases for the "rel-urls" change
#
tantek kylewm, Nested h-* objects' "value" property was your proposal - presumably you can make or have test cases that verify your implementation of it?
#
kylewm yep, I'd be happy to
#
kylewm I added one to mf2py's tests (copy/pasted from the wiki), will build out a more thorough test for microformats/tests
#
tantek also - what do you think of the 1,2,3 steps above for making changes/updates to microformats2-parsing ?
#
tantek (need to add use-case documentation in there somewhere - perhaps proposed resolutions that add things MUST document use-cases for each thing added?)
#
KevinMarks no rel-urls tests in the shared test repository only in mf2py now
#
kylewm the problem is rel-urls affects *every* test case
#
kylewm makes it hard to add to the test suite
KartikPrabhu joined the channel
#
tantek should we make the rel-urls collection create on first rel?
#
tantek as opposed to always there?
#
tantek e.g. "rels" is always there
#
tantek but "alternates" is create on first rel=alternate
#
tantek or does it not matter in practice because (nearly?) all pages have rel=stylesheet
#
KevinMarks i think always there is clearer
#
tantek ok that's the way I added it to the spec so we're fine
#
KevinMarks I was testing with the tests repository by deleting rel-urls from mf2py's output
#
KevinMarks any further thoughts on the whitespace collapsing for implied name?
#
KevinMarks aaronpk, kylewm have you looked at this form a consumer point of view
#
tantek KevinMarks - I'd say for more thoughts on that you have to look at what X2V does
#
tantek because that code contains years of collective experience / wisdom about in practice whitespace collapsing of microformats properties
#
@devseo Reviews: Author HReview Plugin - http://graywolfseo.com/reviews/author-hreview-plugin/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+Wolf-howl+%28Graywolfs+SEO+Blog%29 (twitter.com/_/status/604385326560317440)
#
tantek (and may very well justify your opinions on the matter)
#
aaronpk i have only found implied name useful for the minimal h-card example
#
kylewm "this from a consumer point of view"? the this refers to whitespace collapsing?
#
tantek it's useful for minimal things of any kind. events, venues, products
#
aaronpk pretty sure every other time i've seen the implied name property, it always has extra junk in it
#
tantek citations too
#
aaronpk either extra whitespace, or text that really shouldn't be in it
#
aaronpk i'm not saying it isn't useful, just that you have to be very deliberate about creating HTML to output a useful implied name
#
tantek I'm not sure about that
#
tantek I'd like to see examples we can reason from
#
aaronpk i'm sure i can dig some up
#
aaronpk lots of badly formatted comments on my site
#
tantek as the implied p-name behavior was modeled after simple hyperlinking of proper names in HTML
#
tantek which happens all the time in web pages
#
aaronpk i know, it works fine for the simple case
#
tantek aaronpk - we've discussed the h-entry exception for this before
#
tantek that's a known problem case
#
aaronpk i guess i don't have any input on changes to whitespace collapsing until that is fixed and i've had a chance to play with it
#
KevinMarks you get the same problem with events and locations too
#
tantek aaronpk: fixing "that" may be a matter of comment display for IWC - not necessarily a p-name problem
warehouse13, KartikPrabhu and KevinMarks_ joined the channel
#
tantek edited /h-review-aggregate (+462) "add summary to give implementers at least sufficient information to (start to) code" (view diff)
#
tantek edited /h-review (+727) "/* Parser Compatibility */ item handling" (view diff)
#
tantek edited /h-review-aggregate (+13) "p-votes too" (view diff)
#
@first__tweet @biz Do you know, you were the first person to tweet Microformats http://ctrlq.org/first/69178-microformats/ (twitter.com/_/status/604419848278540288)
#
tantek heh: https://twitter.com/biz/status/29925
#
@biz Listening to a talk about microformats (twitter.com/_/status/29925)
#
@kevinmarks http://ctrlq.org/first/69178-microformats/ @biz, @ev, @sacca first to mention #microformats on twitter (twitter.com/_/status/604420390639640576)
#
@first__tweet @ev Do you know, you were the first person to tweet hCalendar http://ctrlq.org/first/69281-hcalendar/ (twitter.com/_/status/604421255777783808)
#
tantek oops found a bug in first__tweet - it doesn't know to tell Twitter not to auto-correct the search term
#
tantek e.g. from hCalendar to calendar
#
tantek trying quoted term
#
tantek yes that worked
#
tantek https://twitter.com/t/status/646463
#
@t admires simplebits.com. Pulls back the covers, sees hCalendar! Subscribe to SimpleBits speaking: webcal://feeds.technorati.com/events/simplebits.com (twitter.com/_/status/646463)
#
@TinaHui RT @kevinmarks: http://ctrlq.org/first/69178-microformats/ @biz, @ev, @sacca first to mention #microformats on twitter (twitter.com/_/status/604423967865053184)
tantek, KevinMarks_ and KevinMarks__ joined the channel