#microformats 2012-01-11
2012-01-11 UTC
Morphous_ and BobJonkman joined the channel
# 
tantek edited /Special:Log/block () "blocked [[User:Young1]] with an expiry time of infinite (account creation disabled): Spamming links to external sites" (view diff)
# tantek deleted /User:Young1 "Vandalism: content was spam"
# tantek deleted /User_talk:Luffy091 "Vandalism: content was spam"
# tantek edited /Main_Page () "(-1222) Reverted edits by [[Special:Contributions/Jamesalex112233|Jamesalex112233]] ([[User talk:Jamesalex112233|Talk]]) to last version by [[User:ChiefRA|ChiefRA]]" (view diff)
# tantek edited /Special:Log/block () "blocked [[User:Jamesalex112233]] with an expiry time of infinite (account creation disabled): Spamming links to external sites" (view diff)
tantek_ joined the channel
# 
tantek_ also - FYI - I've posted the code for h2vx.com to https://github.com/microformats/h2vx.com
Phae and NimeshNeema joined the channel
# 
Loqi [http://twitter.com/_affordableseo] Create your hCard and get local listing quick.. http://t.co/3d3kO9lx
Phae joined the channel
# Loqi [http://twitter.com/jemyke] ì˜í™”ê°ìƒê¸° - ë°€ë ˆë‹ˆì—„ : ì—¬ìžë“¤ì„ ì¦ì˜¤í•œ 남ìžë“¤ .. ì–´ë¥¸ë“¤ì„ ìœ„í•œ 해리í¬í„° 시리즈ë¼ëŠ” í‰ http://t.co/IkShn8wD @extmovie ì—ì„œ
danbri, quiron, tantek and voxpelli joined the channel
# Loqi [http://twitter.com/gabrieljaquemet] @4sqSupport / url with button here : http://t.co/Hguw0nfo / URL working in your hcard tester is : http://t.co/fL0GI08c (sso = no redirect)
nonge_, quiron, tlr, NimeshNeema, lmorchard, krijnh, MacTed, Soopaman, andr3 and chiui joined the channel
# Loqi [http://twitter.com/nmcmahonweb] Just implemented the hCard microformat onto our company's contact us page. Does anyone know any applications that can extract the vCard?
# Loqi [http://twitter.com/microformats] @nmcmahonweb check out @H2VX in particular http://t.co/AfxHNtOm to create a link to download a vCard from your hCard. ^@t
andr3 joined the channel
# mike created /User:Mike (+162) "Adding public domain license release so you guys don't have to delete all the stuff I wrote in 2006-07" (view diff)
patcito, mkowens and andr3 joined the channel
# Loqi [http://twitter.com/twiki] Need #microformat support? TWiki has a plugin for that http://t.co/DT8WHLev #e20 #microid #hcard #hevent
chiui, kennyluck_ and andr3 joined the channel
# Hixie ignored by whom?
# Hixie if you mean by me, i'm only up to e-mails from last june :-(
# Hixie autocompletetype is a reinvention of the RFC 3106 vocabulary, not the vCard vocabulary. :-)
# Hixie google toolbar and netscape 6
# Hixie can you name any implementations of vcard4 for autocompletion?
# Hixie docs for what?
# Hixie not off-hand
# Hixie search for "ecml google toolbar" and "ecml netscape"
# Hixie both give results
# Hixie i expect there are others, but i haven't looked into it
# mkowens There's even a test page for Google Toolbar for all the ECML Compliant Field Names: http://maettig.com/code/html/ecml_test_full.html
# Hixie i'm not convined authors care enough for it to be worth implementing anywhere, whatever the vocabulary
# Hixie whether it be one that reinvents ecml, or one the reinvents vcard that reinvents ecml :-)
# mkowens It would be difficult to convince most web developers to name their form fields semantically correct, but that's generally the case in a scenario where there has not yet been a really compelling reason to do so.
# Hixie tantek: 2426 is vcard3, you said vcard4
# Hixie <tantek> it's yet another reinvention of vcard4 vocabulary
# Hixie anyway my point was that the vocabulary that is most applicable here is ecml, not vcard
# Hixie vcard doesn't have shipping vs billing, credit card, etc, last i checked
# Hixie (ecml v1 is from 1999 or so
# Hixie vcard1 is from 1995 or so)
# Hixie i don't think it makes much of a difference in this case, since i think the entire feature is likely DOA
# mkowens It seems to me that developers might best be served by using the vcard format for the base "contact" data and extend it with the ability to categorize a given set of vcard data with the type "Shipping", "Billing", or some other "type" that could then be used by whatever agent is doing the autofill can use the correct type of vcard for the autofill.
# mkowens Just my two cents.
# Hixie how far from vcard is the proposed autocompletetype vocab?
# Hixie tantek: as in, just like with ecml, nobody will use it enough to make it worth it
# Hixie well it matters if what you care about is whether we have vocab forking, vs whether what you care about is berating people who are trying to improve the web :-)
# mkowens The tokens are somewhat similar, but the way it's structured is significantly different. There's no real structure to the token system.
# Hixie just the names, or the actual fields?
# mkowens Without any concept of inheritance via structural constraints, the specific names of tokens used to identify the items are more verbose in Autocompletetype than vCard.
# Hixie if it's just the names, then it's not a big deal, we can just define a mapping and then it's basically vcard
# Hixie if it's structured differently, that's a bigger problem
# Hixie (though only if anyone ever tries to use it)
# Hixie i recommend approaching chrome directly to get them to fix the vocab
# Hixie if they get adoption, then the fact that they shipped will bias the spec towards what shipped, regardless of how ugly the names are
# Hixie and if they don't get adoption, it won't get into the spec, so i won't care either way what the names are :-)
# Hixie it's not "oops we already shipped it", i told them to ship it because we need implementation experience.
# Hixie shipping experiments is how the web evolves
# Hixie you don't write standards before you ship
# Hixie mozilla does the same thing, and rightly so
# Hixie so does everyone else
# Hixie *shrug*
# Hixie it's all the same to me
# Hixie what we need is implementation experience
# Hixie it doesn't really matter if you post stuff on a project wiki before or not
# Hixie um...
# Hixie so MacIE was "shady"?
# Hixie fixes for mozilla security bugs are "shady"?
# Hixie come now
# Hixie standards work should be open, sure
# Hixie standards work comes after experimental work
# tantek and I wrote more about it here: http://tantek.com/2011/168/b1/practices-good-open-web-standards-development
# Hixie i've nothing against people proposing things in public, indeed that's all i do
# Hixie but i think it's disingenuous to suggest that anyone doing experimental work without public participation (or at least public documentation prior to shipping) is "shady" ("of doubtful honesty")
# Hixie it's especially so for something like chrome where all the code is in the public eye within days of being writted and within weeks of shipping
# mkowens Maybe the reluctance in implementing before open discussion is had here is more about how harshly it can prejudice future work on whatever it is that is being experimented upon.
# Hixie tantek: you are fully aware that the word "shady" has more baggage than that.
# Hixie mkowens: it's _supposed_ to prejudice future work. That's what it's _for_.
# mkowens Hixie: I'm not saying it's wrong to use experimental implementations to influence future work. I'm saying that the fact that there was no "open discussion" prior to the experimental implementation makes it more like "open* discussion" as it immediately creates an incumbent environment that is difficult, if found to be suboptimal, escape.
# mkowens to escape from*
# Hixie i don't see that as a problem
# Hixie we fix the big issues, and we move on
# Hixie q.v. canvas
# Hixie that is a canonical example of something developed not in the open
# Hixie which we then discussed, fixed (in a way that broke safari quite badly!) the one or two major problems, and adopted the rest.
# Hixie sure, that means certain things were decided in ways that are suboptimal
# Hixie but that's better overall because it means it was based on real implemented experience
# mkowens Right, but one of the things that we, as people who influence the standards, have a responsibility to do is don't break the web. Leaving potential places for future work to "[break] Safari quite badly" is not something we should be encouraging.
# Hixie honestly even if there was "open discussion" on a project wiki before hand it wouldn't make much of a difference if any
# mkowens In most cases, I imagine it wouldn't.
# Hixie it's not like everyone is going to participate in mozilla newsgroup / wiki discussions, mostly only mozilla people are
# mkowens But there might be a handful of times where it might get fixed.
# Hixie the whole point of experimental shipped stuff is that it is experimental and can be broken if necessary
# Hixie it wouldn't be experimental if that wasn't the case
# Hixie it would just be a proprietary tech
# Hixie sure it is
# Hixie it WAS made before shipping
# Hixie in code
# Hixie it's not like you would have noticed it if it was discussed in some public chrome dev list
# Hixie project wiki != discussion
# Hixie != open public discussion, anyway
# mkowens It's the closest we've got to an open public discussion.
# Hixie so if i put something on the wiki, and nobody notices, but you can google for it, you think that's good enough?
# Hixie you're making hair-splitting distinctions with no practical value here
# Hixie the checkins are googlable
# Hixie so by your argument, it was public
# Hixie and discussed
# Hixie o_O
# Hixie dude
# Hixie you made me do a search for this
# Hixie and i have now found that it WAS publicly discussed!
# Hixie In August last year!!!
# Hixie heck if i know
# mkowens Looks like Microsoft.
# Hixie i didn't know about it til about a day before they posted on the whatwg list
# Hixie but by your argument, that's public
# mkowens autocompletetype appears to have originated as a ASP.Net property
# mkowens as an*
# Hixie it was a project site, a searchable/discoverable place, months before shipping, which is completely open to all
# mkowens Eh
# mkowens scratch that. Just actually read the details of that and that's a very different use case.
# tantek hixie, what search on google.com led you to http://code.google.com/p/chromium/issues/detail?id=92121 on the first page of results?
# Hixie "and if the public wiki shows up" "it does = discussion"
# Hixie what search terms have you been randomly searching for over the past year?
# Hixie and you've been working on autocomplete?
# Hixie you're the one who claimed that if it could be found throught search that was good enough
# Hixie i'm saying that's silly
# Hixie because you're not searching for things you didn't know others were working on
# Hixie i wouldn't be surprised if at the time that was all the documentation that there was
# Hixie i doubt the thing on the wiki was written before it was checked in
# Hixie anyway
# Hixie this is all academic since i still think this is DOA :-)
# Hixie right
# Hixie looks like the order of events was bug, code, checkin, ship, document, ping me, post on public list.
# Hixie well the list is what matters
# Hixie the wiki is just a place to put notes
# tantek here is the work in progress on the related research that's been done on the microformats wiki: http://microformats.org/wiki/hcard-input
# mkowens Thanks
# mkowens I'll check it out when I've got a bit more time.