Timestamps are in UTC.
[[hlisting-challenges]] http://microformats.org/wiki?title=hlisting-challenges&diff=0&oldid=4889 * Rohit * (-12) Structured CFP -
[[hlisting-challenges]] http://microformats.org/wiki?title=hlisting-challenges&diff=0&oldid=4890 * Rohit * (+9) Minimal Classified Ad -
hi mark --
our javascript mf parser bookmarklet just passed 7000 lines :) :( !
how's yours, from what I hear?
rohit: that's a big bookmarklet!
what does it do once parsed, offer to save the item in the corresponding macro-format?
almost
it's called miffy, a microformats editor/inspector that finds what's on the page and extracts it to a search engine at tpd.angstro.net:19988.
if you want, you can click on search, (get no results), drag out hte bookmarklet, then go back to the home page and click on the logo to see the credits -- running miffy then will find the rel-license and hCard.
The mysterious part is that clicking on the Å logo in the miffy window will repost the found bits to angstro for search
the search also supports xquery, but that's another story...
interesting, i'll check that out
mlinksva is Mike Linksvayer and from Creative Commons
KevinMarks is a writer of code, limericks, weblogs & syllepses & his blog is at http://epeus.blogspot.com & he explains how to get Creators paid at http://mediagora.com & originally from London, UK & living in Willow Glen, San Jose, CA & working at Technorati & is WikipediaWorthy & part of http://microformats.org & PST (UTC-8) & also in every time zone ever created, for KevinMarks is everpresent & the Podfather & the spidermeist
tantek is Tantek <http://tantek.com> and works on Technorati and develops microformats <http://microformats.org>
KevinMarks is a writer of code, limericks, weblogs & syllepses & his blog is at http://epeus.blogspot.com & he explains how to get Creators paid at http://mediagora.com & originally from London, UK & living in Willow Glen, San Jose, CA & working at Technorati & is WikipediaWorthy & part of http://microformats.org & PST (UTC-8) & also in every time zone ever created, for KevinMarks is everpresent & the Podfather & the spidermeist
mlinksva is Mike Linksvayer and from Creative Commons
bergie is lives in Finland and blogs at http://bergie.iki.fi/blog/ and Midgard CMS developer
KevinMarks is a writer of code, limericks, weblogs & syllepses & his blog is at http://epeus.blogspot.com & he explains how to get Creators paid at http://mediagora.com & originally from London, UK & living in Willow Glen, San Jose, CA & working at Technorati & is WikipediaWorthy & part of http://microformats.org & PST (UTC-8) & also in every time zone ever created, for KevinMarks is everpresent & the Podfather & the spidermeist
trovster is a web developer from the UK who writes on http://www.trovster.com and runs www.csslounge.co.uk
karlUshi is karlcow
blake is allegedly human. Blake, also known as Cortland M. Setlow, studies at swarthmore.edu and enjoys building things, exploring buildings, and physics. He currently sleeps during the day.
Is there are link to a map in vcards?
bergie is lives in Finland and blogs at http://bergie.iki.fi/blog/ and Midgard CMS developer
bkdelong is B.K. DeLong, Head Research Analyst for HALO Worldwide - http://www.haloworldwide.com. Web: http://www.brain-stream.com. Email: bkdelong@pobox.com and lives in Salem, MA, USA (-5:00 GMT)
Morning
evening
that too. ;)
[Insert your time-of-day greeting here]
hehe.. it was just sitting there... I had to take it
dglazkov is Dimitri Glazkov (http://glazkov.com) and lives in Birmingham, AL, USA (-6:00 GMT)
bergie is lives in Finland and blogs at http://bergie.iki.fi/blog/ and Midgard CMS developer
bkdelong is B.K. DeLong, Head Research Analyst for HALO Worldwide - http://www.haloworldwide.com. Web: http://www.brain-stream.com. Email: bkdelong@pobox.com and lives in Salem, MA, USA (-5:00 GMT)
tantek is Tantek <http://tantek.com> and works on Technorati and develops microformats <http://microformats.org>
mlinksva is Mike Linksvayer and from Creative Commons
I've put a hcard into our sites promoting badge :d
trovster - url?
[[hcard]] http://microformats.org/wiki?title=hcard&diff=0&oldid=4891 * Trovster * (+177) New Examples -
I convinced churchzip.com to use hcard in it's gmap interface
[[hcalendar]] http://microformats.org/wiki?title=hcalendar&diff=0&oldid=4892 * Trovster * (+91) New Examples -
eventually there'll be a refactor in the traditional interface, in which mf's will be used
TantekC: Well, they're the public viewable microformats, at the moment.
But the promote page - http://www.multipack.co.uk/promote/ - has a vcard for the badge.
ah, very cool
nice overylaying of the hCard and hCalendar event on the home page: http://www.multipack.co.uk/
Nice spam title, "Former President Bill Klinton uses Voagra!"
Lets see, how to mark that up...
TantekC: I think you suggested that hCard/hCalender event overlap1
The site also has rel="home" on the header.
is there any way to designate the page authors vcard besides the address element?
[[hcalendar]] http://microformats.org/wiki?title=hcalendar&diff=0&oldid=4893 * Ragdoll * (+237) New Examples -
Do we have any examples of using a single hcard property in a blog post? ie if I want to just point out that a name is a name ....? do I just do <span class="fn"></span> ? Or do I need to create a div with the vcard in there?
[[hcard]] http://microformats.org/wiki?title=hcard&diff=0&oldid=4894 * Ragdoll * (+260) Examples -
[[hcard]] http://microformats.org/wiki?title=hcard&diff=0&oldid=4895 * Ragdoll * (-260) Examples -
[[hcard]] http://microformats.org/wiki?title=hcard&diff=0&oldid=4896 * Ragdoll * (+260) New Examples -
bkdelong: AFAIK, vcard needs to be defined. But if it's blog post, add it to the container
hrm...ok Here's the example:
<p class="vcard">I think this Post article by <span class="n"><span class="given-name">Dan</span> <span class="family-name">Froomkin</span></span> sums up a lot of the issues right now. You have <span class="n"><span class="given-name">Lewis</span> <span class="nickname">Scooter</span> <span class="family-name">Libby</span></span> - <span class="role">Cheney's ex-<span class="title">Chief of Staff</span></span> testifying that
I mean, is it even appropriate for me to be marking up other people's names with vcard?
Yes, it's appropriate!
Just checking. ;)
oops. Plame should be family-name
any thoughts on properly marking up Cheney's ex-Chief of Staff? I'm not sure if I should include the "ex" in there as that isn't an official title. Not to mention that Cheney is a family-name: <span class="role">Cheney's ex-<span class="title">Chief of Staff</span></span>
hober is Edward O'Connor and works for EVDB on http://eventful.com/ and lives in San Diego, CA (-08:00)
tantek is Tantek <http://tantek.com> and works on Technorati and develops microformats <http://microformats.org>
Hey tantek.
hey bk
Actually, you know trovster, I think I have to add a vcard class before every case of a different name.
Yes, you do.
messy.
I have it on here -> http://www.multipack.co.uk/members/
gotcha ok.
My god....what a nasty set of nested HTML. :)
<p>I think <a href="http://www.washingtonpost.com/wp-dyn/content/blog/2006/02/10/BL2006021000760.html">this Post article</a> by <span class="vcard"><span class="n"><span class="given-name">Dan</span> <span class="family-name">Froomkin</span></span></span> sums up a lot of the issues right now. You have <span class="vcard"><span class="n"><span class="given-name">Lewis</span> "<span class="nickname">Scooter</span>" <span class=
bk, you can use the fn/n shortcut in those
?
hmm... and actually, they are missing the "fn" so they are invalid
Wait....I thought n was appropriate
http://microformats.org/wiki/hcard#Implied_.22N.22_Optimization
http://microformats.org/wiki/hcard-example1-steps#hCard_example_iteration_2:_element_conservation
ah ok...so "fn n"
right, that will make it valid
What about the title/role for Libby?
but in this case, since all they have is given and family names, you can use the fn shortcut
oh right, with the nickname in the middle like that, you need the full "n" markup
where is the title/role for Libby?
;)
your markup got clipped at: ...Scooter</span>" <span class=
<span class="vcard"><span class="n"><span class="given-name">Lewis</span> "<span class="nickname">Scooter</span>" <span class="family-name">Libby</span></span> - <span class="role"><span class="vcard"><span class="n"><span class="family-name">Cheney</span></span></span>'s ex-<span class="title">Chief of Staff</span></span></span>
I dont think I can markup Cheney's last name. Not enough contextuality to mark it as a vcard
Unless I simply add his first name. Damn my laziness!
right - it's an adjective there, not a noun
So it is assumed with fn that if there are just two words, they're given-name and family-name.....
right, that's the shortcut
Can a title be nested in a role?
there is another proposed fn shortcut - if there is only one word, it is a nickname
yes
in general, anything can be nested in anything
So what if I said VP Dick Cheney's Chief of Staff? Would that warrant vcard fn/titleing Cheney?
I think when the markup:visible content ratio hits 20:1 or whatever that sample has, it's getting a tad ridiculous
you certainly could
essentially, hCard provides the closest thing to a <person> element in (X)HTML
that's the way to think of it
I can't imagine any normal human being putting that many span tags in
doesn't have to bu human
Eventually it could be automated.
qid, ideally you would just have a little "person" button (just like WYSIWYG UIs have "link') buttons, that could either insert the markup for you, or give you a drop down list of people to choose from
Or I could cut and paste Dan Froomkin's name from the WashPost byline and my hCard plugin can pull the revant markup from WashPost, (if they used it), and auto-include it in my document.
yes
But yes, it's ridiculous to a degree...but we have to start somewhere.
or if you started typing a person's name in your address book, the editor could auto-complete a minimal hCard (name and URL)
*from* your address book
Right. or a dynamic wikipedia-like lookup
similar to what Trillian uses
right, for names in general
and of course, this auto-complete could also benefit from all the people's names that your browser has seen while parsing hCards during your browsing
so they wouldn't even have to be in your address book
.. or a more contained case... a CMS that knows previous names that have been entered
right
a dynamic hCard Dreamweaver extension
yes
I put my hCards through a site to give me the vCards, and it only gave me the first one on the page.
hmm... could be a markup problem. which URL?
That members one
what is the URL for the site to "get" the vCards?
http://suda.co.uk/projects/X2V/get-vcard.php?uri=http://www.multipack.co.uk/members/
hmm.... I'm not seeing *any*
what happens when you remove the "fli" from "fli vcard"?
also, I noticed the "org fn" outside any hCards for "Multipack"
Yes, they're whereever Multipack is mentioned.
That's not the issue. And it's valid HTML too. They appear using Tails.
right, that should be fine
did you try removing "fli" from Simon's hCard?
No need. When running that script posted above, Simon's vCard appears.
All I'm saying is, that that script only gives the vCard of the first hCard it finds.
Anyhoo. Gotta go home. Be back in 15m
cool
Does jibot take messages for people?
not AFAIK
briansuda is brian suda of X2V fame
no need to save a message i guess :P
actually, I meant for trovster. ;)
brian, you had trouble reproducing trovster's problem?
trovster is a web developer from the UK who writes on http://www.trovster.com and runs www.csslounge.co.uk
yes, when i extract the vCards, i get all 16 just fine
trovster, that's in reference to your page
Extract with?
http://suda.co.uk/projects/X2V/get-vcard.php?beta=yes&uri=http%3A//www.multipack.co.uk/members/
brian, what if you try without the "beta=yes"?
it doesn't matter, with or without, i have done, both
odd, same here brian. both work for me now.
trovster, did you change something in the markup?
Ragdoll is learning to use jibot
Nope. I literally drove home. Heh!
Hello, folks.
glad to be of service! i like when things fix themselves
The last guy to use the Ragdoll handle registered a year ago and left a day later. Think this can be cleared so I can use it?
Is there an application out right now that collects hCalendar data? Like a place where I could look up events in Lincoln, Nebraska and it aggregates all the appropriate data.
bk, see your email ;)
Ragdoll: we can't help you with Freenode network issues, FYI
ragdoll, i believe EVDB is working on aggregating hCalendar
Cool, I will check it out.
Wow, how does this know my location without me explicitly telling it?
Thanks, Tantek. Very helpful.
am i missing something at eventful? whys it look like the site thinks its september
tantek: unfortunately, we're not working on that, no
Ragdoll: IP geolocation
pnhChris: how do you mean?
non registered user, go to the site for the first time in who knows how long... see "upcoming events in USA.. and they're sep3+ sep 12+ sep 17+.. etc
pnhChris: thanks. looking into it
ah.. its all year long "events"
sep 3 2005 - sep 30 2006
right.
[[press-faq]] N http://microformats.org/wiki/press-faq * Tantek * (+687) created. might as well answer press questions on the wiki.
You might call what it's doing a feature, maybe
:/
i guess i don't expect ongoing perfomances of a comedy troupe to be an "upcoming" event vs. an "ongoing" thing
[[faq]] M http://microformats.org/wiki?title=faq&diff=0&oldid=4897 * Tantek * (+154)
well, the particular performances that haven't happened yet are upcoming, yes?
or the upcoming thing be specific to the night this week they're playing so that it doesn't get stuck in a search query as sept of last year
Typically, the IP geolocation stuff will get a better location for you, not something generic like USA.
and in more specific locations, this is much less of an issue
but it is an issue, which has been raised with the relevant people
and less if i was signed up i'm sure
just an odd first impression
I agree
yes, note that upcoming has *also* ditched recurring events
as well as EVDB
We ditched recurring events?
that's what Brian keeps saying
(your Brian)
That's, err, interesting.
because it caused too much noise in results
they caused
We support recurring events from non-user data sources, but don't have a recurrence UI in place for users when adding/editing events.
ah that's it
odd that you trust non-user data sources more than users
Recurrence UI is hard; let's go shopping!
Well, parsing RRULEs isn't hard.
uh
i guess if they could be handled more like recurring events in the listing.. with perhaps a limit on how many times they could repeat or for how long it would cut down on the noise (have to relist monthly for example).. but i can see that.. 10 museums posting their daily repeating schedules and you don't have much room left
(well {R,D}{RULE,DATE})
hober, I *dare* you to repeat that on the ietf-calsify list
heh.
Obviously, I'm not the one who has to write that part of the code.
But hey, the code got written, and it works, so there you go.
i'm just saying many others have tried, and only gotten like 90% or so to work
What I mean is that parsing iCal is at least reasonably algorithmic. Creating great, usable UI isn't.
including across DST changes etc.
hober, that's certainly true
hober: hmm, what happens if someone visits your site from a multihomed IP address?
qid: dunno
I would like to solve the recurring events UI problem...
does the ical format have a way to exclude certain dates or repetitions in the cycle? I'm not very familiar with it unfortunately
pnhChris: yes, DDATE and DRULE
err, sorry, XDATE and XRULE
something like that anyway
the exclusion rules and dates are exactly the same as the recurrence rules/dates
hober: What service do you use for the geolocation?
Atamido is Paul Bryson, http://orangeman.commo.de/
Ragdoll: We bought a geolocation db, but I don't know which one offhand.
http://homepage.mac.com/kevinmarks/nv.mov
mlinksva is Mike Linksvayer and from Creative Commons
hey Kevin, join #barcamp
BenjaminCarlyle is http://soundadvice.id.au/blog/, GMT 1000
they're talking about tagging at MooseCamp: http://homepage.mac.com/kevinmarks/nv.mov
TantekC: Are you there at the session?
I am
I can't quite hear well - but what is the alternative to tagging they are discussing?
tantek is Tantek <http://tantek.com> and works on Technorati and develops microformats <http://microformats.org>
dotBen is Ben Metcalfe & blogging at http://benmetcalfe.com/blog/ & civil to mena http://flickr.com/photos/factoryjoe/70863244/ & not at BarCampNYC & the guy who likes to say bullshit & prone to shouting "Bullshit" in IRC Backchannels & does not like Apple computers & the man behind http://backstage.bbc.co.uk - the BBC's developer network
welcome ben!
hey
comments microformat
to set the context a bit
there are currently three fairly active, fairly well proceeding blog-related microformat efforts
(perhaps _conversation_ microformat, acutaly... hmmm not sure)
I think comments on a blog are a different enough case to justify studying as such
First there is hAtom for marking up posts in a blog
http://microformats.org/wiki/hatom
yup
Second, we have the blog-description work which is all about marking up everything else in a blog, see here for starters: http://microformats.org/wiki/blog-description-examples
yes, that's sort of the "vs structured blogging" stuff
And Third, we have the blog-info work for documenting how people describe and reference blogs from other sites. http://microformats.org/wiki/blog-info-examples
<ducks>
:)
heh, not quite
the portions of a blog besides the posts
I see
e.g. the header with description, author, contact info, blogrolls, footer with license etc.
well, for me it's a bout solving a problem
that's what blog-description is for
good!
the problem is how do you tack comments -- you've made, on your blog, on blogs your interested in
part of the microformats process is that we research existing behavior *first* before proposing any solutions BTW
track*
quite
ah yes
basically, how do we do a distributed CoComment?
it's the problem cocomment is trying to solve, but onto that in moment
well
have you seen http://benmetcalfe.com/blog/index.php/2006/02/05/cocomment-semantically-forked-conversation/
that's the other reason for it -- cocomment fundamentally doesn't work
and will never work in the longtail where the chances of the audience using cocomment are next to 0
<lets you have a read of that blog post>
brb
yikes!
the semantically forking thing is really stupid
ugh
i had no idea
yeah
to be honest
it's shit
so is asking *every* person who comments to sign up for the service
I have to say at leat they got _something_ out to solve the problem
yes
like you said, that won't work
right
because they can't screen scrape every different template to get the other non-cocomment submitted comments out
it's a nice try, but no cigar
right
and what's worse is you actually get mis-representation
as I said, what about spam
or comments you choose to remove/edit (not that you should edit, but if you legally need to etc)
Gahh.. too much email to catch up on. I'll try to look into recent hAtom happenings today, or tomorrow.
yeah, the fact that there is more than one "version" of a comment is a big problem
also, as the blog owner, I loose control of MY representation of MY comments
(on MY post)
ok, so it's shit
so what can we do?
well for a start there are a number of behaviors here
1) people like me, probably you, who's exposure in the blogosphere is mainly via newsreader and agregators
what you're asking for is really a comment aggregator
well that's the outcome from the microformat
something that aggregates *comments* on blog posts in addition to the posts themselves
yes
and then does intelligent things with them in the UI
but of course you can then do lots of other cool things too
i want the blog services that authenticate comments to create 'blogs' of them
(insert UI cleverness here ;) )
at the moment, I'm at the level where I want to output the XML
heh, don't assume XML ;)
what you then do with it is up to you
well
# inculde 'xhtml is xml'
that's part of the point of microformats. you don't actually need to create yet another version of the data.
like I said - the behavior I'm thinking about is people like me who use agregators - not browsers - to access the blogosphere
jakedahn is Jake Dahn he blogs at http://jakedahn.com . He also runs http://looce.com and is a volunteer @ Flock and BarCamp's Official Kid
sure, comment aggregators can offer RSS feeds of aggregated comments
we should discuss the xhtml/xml thing later/another time -- that's an interesting debate for various reasons
Comment aggregation could work in with distributed-conversation. You could treat them as individual blog entries hosted with the entry being commented on. They aren't that different to h
:)
Ooops.. to hentry
hmmm but from an object perspective blogs posts are different to comments
As I understand things the atom committee deliberatly wrote the comment problem off as too hard to solve in a feed format. It could still be solved externally, though.
there are many other properties a blog post can have that a comment doesnt
correct
ok ok
it is a different problem
can I outline an _idea_ I have
similar, but different
I'm kind of struggling to think of any... do you have an example?
ben, how about researching and documenting it first
benjamin, it might be a strict subset
I'm thinking of a comment as being pretty much the same as a blog entry I write in distributed-converstation style about the original blog entry...
Ben: technorati tags, enclosures
catagoris
Ok, that makes sens.
e.
it may be that a comment is a strict subset of an entry
but we should research it first to determine it
well, I disagree
ben, hence *may*
I think comments are very different
lol
ok
:)
ben, i'm not saying i know either
this is why step 1 is research and document examples
well to go back to behaviors
the intersting thing I would like to mini-brainstorm
is what other 'problems' are there around comment
comments
tracking is one (a la cocomments)
agregating is another
searching (a la technorati for conversation) is another
but there may be others
????
I guess the interesting question from the aggregation question is whether you would try to aggregate blog entries about something and comments about that some thing together...
i want the Flickr "Comments you've made" feature
yes
but across the whole web
exactly
well, if the microformat was adopted, there's a whole business in there for a technorati/pubsub/icerocket of converastion
ok, that's a good problem statement
so now we head to the wiki
dotBen, take a moment to read http://microformats.org/wiki/process
sure
hmm
looks like i missed a good discussion
comments, in my thinking, are a browser's "sent mail"
chris, it has just begun, don't worry
reread from archives to catch up
can I ask a question Taktek
which maybe you can't answer but would be nice...
yes of course
is this a problem technorati is already working on solving?
because if it is then it kind of would be good to know
it is a problem we would *like* to solve, but are not solving currently
ok
that's cool
it is a *difficult* problem
well, insofar as blogposts are comments we do
and just out of curiosity. from your experience...
what are the difficult aspects?
Kevin, we're not talking about that
aother than adoption
this is an attention issue too
for example
we're specifically scoping the discussion to *comments* on *blogs*
if i add an event to upcoming
it should list my comments next to events created...
attribution/authentication of commenter is hard
Kevin, that might be necessary for all applications
Kevin - that's an area I definitely DON'T want to try and solve
not for this anyway,
otherwise it will never happen
right, let's keep this simple folks
for me it's about a programtic representation of the comments list I see on a given blog post
yes
at this point in time I don't care whether the "factoryjoe" who has posted on a gien blog is really chris m or not
given*
agreed
i think it's about watching a discussion
and seeing threads emerge
bingo
hmm.. maybe i should stop commenting as "chris"
gmail for the blogosphere
irc is still quite useful, even though you don't always know that a nickname is always the same person
what we MIGHT need is a unique identifier for commentor
you mean like a URL?
if you want to traverse comment representations in a search/agregator environment
:)
yes
heh
yes
damn
i need rhyzo
like maybe their *BLOG* url ? :)
perhaps an md5 of the email, generated by the blog engine
yuck
yeah yuck
i don't always use a url.. particularly when they're blogs of eople i know
I said maybe
ew
I just thoutht of it in my head
dotBen: FOOL!
so its just "chris" .. and thus i figure the comment is lost
INFIDEL!
but look
you have two options on that:
dotBen, so first things first
1) acess a central service
been smokin' the hash tables again eh?
let's document the examples
2) geneate unique identigiers in the blog engine and hope everyone ends up with the same
and both are messy and beyond what I want to do with this format
right, punt on the whole uniqueids thing
they are not necessary, and a huge rathole
and then you get into all the identity arguments etc.
use urls, move on
agreed
so look
do all your microformats end up being xhtml
cos that's sucky too
sorry but it's true
no, that's what people actually use on the web
eh?
[[distributed-conversation-examples]] http://microformats.org/wiki?title=distributed-conversation-examples&diff=0&oldid=4898 * BenjaminCarlyle * (+207) href and quote example
dotBen: whoah man, you're in the microformats channel now
show some civility
These logs were automatically created by mflogbot on chat.freenode.net using a modified version of the Java IRC LogBot.
See http://microformats.org/wiki/mflogbot for more information.