Timestamps are in UTC.
pnhChris is Chris Casciano, blogs at http://placenamehere.com/ , and a member of the Web Standards Project.
[[work-of-art]] M http://microformats.org/wiki?title=work-of-art&diff=0&oldid=5582 * TimG * (+1) changed "a xhtml" to "an xhtml"
[[workofart-brainstorming]] http://microformats.org/wiki?title=workofart-brainstorming&diff=0&oldid=5583 * TimG * (+44)
[[workofart-examples]] http://microformats.org/wiki?title=workofart-examples&diff=0&oldid=5584 * TimG * (+68) Added project navigation in introduction
[[workofart-formats]] http://microformats.org/wiki?title=workofart-formats&diff=0&oldid=5585 * TimG * (+69) Added project navigation in introduction
[[workofart-brainstorming]] http://microformats.org/wiki?title=workofart-brainstorming&diff=0&oldid=5586 * TimG * (+63)
[[workofart-brainstorming]] M http://microformats.org/wiki?title=workofart-brainstorming&diff=0&oldid=5587 * TimG * (+4) editorial adjustment in introduction
[[workofart-examples]] http://microformats.org/wiki?title=workofart-examples&diff=0&oldid=5588 * TimG * (+98) Added UCLA Hammer Museum example, alphabetized examples by museum nameExamples -
http://benbarren.blogspot.com/2006/03/paula-abdul-to-start-microformats-idol.html
[[directions-examples]] MN http://microformats.org/wiki/directions-examples * Tantek * (+480)
karlUshi is karlcow
[[directions-examples]] http://microformats.org/wiki?title=directions-examples&diff=0&oldid=5589 * ZenPsycho * (+3113) examples -
[[directions-examples]] http://microformats.org/wiki?title=directions-examples&diff=0&oldid=5590 * ZenPsycho * (+919) examples -
wow, msn's source is cleaner than google's
I think google attempts to compress their source
fewer bytes on the line...
That does not seem to be the case in google maps
most of the code on the page is placed there by javascript
and itself includes much javascript
directions-examples
[[User:Izo]] http://microformats.org/wiki?title=User:Izo&diff=0&oldid=5591 * Izo * (+38)
[[hresume]] M http://microformats.org/wiki?title=hresume&diff=0&oldid=5592 * Izo * (+92) Examples in the wild -
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)
[[hcard]] http://microformats.org/wiki?title=hcard&diff=0&oldid=5593 * Article7 * (+167) Examples in the wild - Window on Woking
dglazkov is Dimitri Glazkov (http://glazkov.com) and lives in Birmingham, AL, USA (-6:00 GMT)
[[hatom]] M http://microformats.org/wiki?title=hatom&diff=0&oldid=5594 * Fil * (-207) Pre 0.1 hAtom implementations - sedna is hatom 0.1 compliant now
[[hatom]] M http://microformats.org/wiki?title=hatom&diff=0&oldid=5595 * Fil * (+207) 0.1 hAtom implementations - Sedna is hAtom 0.1 compliant now
dglazkov is Dimitri Glazkov (http://glazkov.com) and lives in Birmingham, AL, USA (-6:00 GMT)
[[hcard]] http://microformats.org/wiki?title=hcard&diff=0&oldid=5596 * Ciordia9 * (+174) Examples in the wild -
pnhChris is Chris Casciano, blogs at http://placenamehere.com/ , and a member of the Web Standards Project.
hober is Edward O'Connor and works for EVDB on http://eventful.com/ and lives in San Diego, CA (-08:00)
RobertBachmann is Robert Bachmann <http://rbach.priv.at/> and lives in Austria (Timezone: UTC+02)
[[hcalendar]] http://microformats.org/wiki?title=hcalendar&diff=0&oldid=5597 * Raster * (+90) Examples in the wild - added link to ocono.com
limbo_ is Eran and blogs at http://hellonline.com/blog/
anyone here run NetNewsWire 2.x (not lite) and want to try something?
(hatom subscriptions)
I don't
pnhChris: i do
factoryjoe: i haven't written up instructions yet, but want to try subscribing to a hatom document with my applescript at: http://placenamehere.com/mf/netnewswire/
(slightly modified from what I mailed the list a few days ago)
ok
(1) extract the zip... put the 3 files in ~/Library/Application Support/NetNewsWire/Scripts
(2) follow the pictures to subscribe to something that uses hatom ... like http://chunkysoup.net
.. and let me know if it works ;)
kk
it'd be nice to get a usability expert of some kind to help determine the best way for presenting an interface to editing/creating an hcard (and/or other MF's) would look like
the labels are abstract/ambiguous to some degree to support a wide range of uses...
and most of the attributes are probably commonly unused
bewest: i'm working on something like that
so how do you provide them in a way that isn't distracting and is easy/natural to use without the user being familiar with the special knowledge of what 'region' means in their locale
yeah?
I'm very interested
how are you working on it?
well
you're on the MF list?
erm
I thought I was
but I don't get emails
so maybe I'm not
maybe it's going into my spam
I'd think the interactions between apps are what really need the examination.. that's after all where the big win is.. not in adding am hcard to an individual page. .but grabbing data from Outlook and dropping it into an app.. or puliing data from a page generated by an existing db and resuing it
especailly things like hatom that get templated once and then all happens behidn the scenes
oh I have been recieving them
it's in my label "microformats" even
600+ messages
heh
pnhChris: yes, but the data has to come from somewhere
bewest: check out today's thread about behavior
pnhChris: and it's going to come from a human
.. or general CMS / blogging tools that remeber data you've already entered and give you good ways to reuse them isnide the app
pnhChris: i think moving off the desktop will be essential
as in the web clipboard demo
it's website to website now
not just app to app
it's also about pulling data from arbitrary sources at "run time"
I'm currently concerned with how to collect the most data from the human (for the initial capture of an hcard) with the most fields possible with the least amount of effort/angst on the users end
i.e. when you're visiting a webpage
bewest: not always from a human... infact i think most cases will remain to be extractiosn from existign data sets
once it's captured in an information system yeah, we can pass it around the web all we like
i dunno
well.. you may have a point
i think we need to look at data slurpage over time
not at one time
in MY app a lot is going to come from humans initially
... be it an amalgamation of data entered in your flicker account... or pulled from a claenderaing app that only uses hcalendar for presentation and instead collects other data
it's no longer about data entry
it's about data magnets
because there is no data to pull from another source
right
factoryjoe: yeah I like that
anyway, my specific problem atm is that I'm not storing the individual fields in the database... I'm storing an hcard (the markup) in the database in one field called "content"
so my content field has the entire hcard
yes, point was the process is bigger then entering, except in a few specific cases
which means to edit this record, I have to parse the hcard, and display a form with it all filled out
but I'm still left with the case of missing properties, and how do you provide addition/removal of extra properties
for example name prefixes and suffixes
right now I don't collect it at all because it'd clutter up the interface
but if someone really wanted it, what would be the best way to provide capturing it
a completely freeform hcard isn't easy
from a UI perspective
yeah, that's where I'm coming from right now
my form for the textpattern plugin was fairly limited as well
damn gmail is giving me some error 704 crap
and included a 2nd ui piece for adrs
I mostly copied tantek's hcard-o-matic
.. and i don't re-arse the info, just build the tag
re-parse*
but i could fairly easily, because the data is stored not in html, but in a custom tag (textpattern) that gets parsed after authoring
hmm\
that's a tough one
i mean, we really only care about FNs
yeah... I suppose the spirit of things here is to not worry to much about the fringe patterns?
80/20?
but at some point it'd be nice to have an editing solution that conforms to the whole standard, not most of the standard
there's 3 issues with hcard if i remember.. the variety of name formats based on what data you have, the 0-N adr issue, and the simple range of other datta you can collect at 0-N
(similar hcalendar editing issues)
I already use behaviour.js heavily
it's not that great for parsing hcards, afaik
well maybe for parsing
... and in none of those cases does a UI , even a great one, allow the ability to overlay microformats on freeform text well
but not for editing
... they're only good if you're dropping blocks of data into the middle of a document
yes
well i think that we'll see blogging tools get insertion tools first
and wikis second
(if i have any say, that is)
and hell, maybe a browser that knows blogging
well, UI is somethign i want to revisit on my textpattern tool
but i still don't see good ways of building, say, what will probably becoem the typical address/hcard/author build
on a page
hmm that window thin is neat
too bloated though
I was going to use something liek that
where its a line of text with some but not all visible content inside specific elements
pnhChris: what UI do you have?
can you flickr it?
except using http://particletree.com/features/lightbox-gone-wild/
just reusing the form builders that textpattern has for other tags
maybe there is just a basic form
briansuda is brian suda of X2V fame
with ghosted text like silver colored <enter name here>
when you click it turns into an input
they're fairly "dumb" now... with just a bunch of fields.. and the logic on the backend to handle some of the rules
when you blur it goes back to plain text, but with the actual input
and initially it's just basic information
but you can click on "advanced" and an advanced panel would slide out underneat
ends up similar to tantek's builders, with a bit more logic behidn them
yeah
in-place-transparent-editing
or something
would that be generally useful?
if so I'll start right away
gonna be a pain to code though
go for simple above all else
frankly if you can just embed a vcard span, we'll be in good shape
you could add a button to tiny mce
"insert person's name"
bootstrap it
well I've got a simple version http://nearwhere.com/guest/
and add to the data after the fact
basically
once you get jsut ONE hcard, you can expand it later
or, as you pick up hcards for the same person, merge the extended data
bewest: whoa...Warning: Division by zero in /home/nearwhere/www/libs/functions.php on line 36
Warning: Division by zero in /home/nearwhere/www/libs/functions.php on line 37
gah
NearWhere Lodging near -, -, -
Back to Map Navigation Link Without Marker
interesting
hrm
ip2geo must have failed on you
http://nearwhere.com/guest/40.3616,-74.263,6/?t=Map
the screenshot doesn't exactly tell the whole picture.. the tag that gets build can be put anywhere in a post... and the adr building comes from another window like that and gets placed into the content area of the hcard tag... http://placenamehere.com/temp/txp_hcard_builder.png
factoryjoe: did you get a chance to try the NNW thing?
.. but that textpattern plugin is "simple" like tanteks builders in that it doesn't provide a ton of flexibility
pnhChris: not just yet
btw, anyone want to meet up wed. or thur. night in San Fransisco?
I'm flying in for an interview
hey sure
off for a bit to get some work done
factoryjoe: ok neat
;)
good luck
thinking how to do inline editing... maybe something like <ul class="modal_edit"><li class="selected"><span class="fn init"><enter name></span></li><li class="inactive"><input id="vfn" value=""></li></ul>
then when you click on selected it switches around
no
that's ba
bad
don't create the hcard before it has a value
flock will be indexing MFs
and if it comes across empty hcards, it'll likely index them
I see :-(
yeah that's bad
hrm
so need a pseudo-hcard
maybe vvfn for display, vfn for input, then fn for the actual hcard
is flock going to detect/display rss feeds sometime?
?
it already does
pnhAway: your script rocks!
sorry i haven't tried since the first beta
can i blog it?
ah
you might try 0.5.13
I'll go 0.7
gah, problem with what I just described is that it breaks tabbing between fields
factoryjoe: was planning to write it up and give it some proper instructions as soon as i had time (today or tomorrow).. but if you want to link it be my guest
the zip isn't chaning.. since it seems to work
changing*
ok nice
how good does the hatom implementation need to be?
like, how robust is the script on bad xhtml?
not
kk
it doesn't run the page through tidy cause i don't know what the isntall base is by default.. know its not on this 10.3.9 box
.. but i think its on 10.4.x
ok
ooo
i'd love a script that would process NNW entries for hcards
and then add them to the address book
i guess that might be a bit premature
ah
ding ding ding
w/ the CSS layer
you could do something like .vcard fn:after {content: "add to address book";} and somehow linkify that post to the technorati contact parser
s/fn/.fn/
right
you could do an applescript to do it chris
right
well i'd have a hard time
but someone else could!
;)
ooo
ok
here's the challenge
pipe all incoming RSS into a plugin for spotlight that detects MFs
and when i do a search
it should be able to pull out contacts, events and so on
in fact
it should open them in address book and calendar
[[work-of-art]] http://microformats.org/wiki?title=work-of-art&diff=0&oldid=5598 * TimG * (+189) Reorganized index and clarified the contents of each page
incoming in what app though?
Can someone explain to me what problem hAtom solves?
I mean, we have atom/rss. Why do we need a syndication microformat?
representing a feed in HTML
conversely, I have a blog, why do I need another copy of the posts wrapped in XML instead of HTML?
Hmm.. K. That was my dumb question for today.
Thanks!
provides DOM access and XPATH access
hmmm I suppose someone has a list of XPATH things to parse an hcard I could just recycle
mm, imma php monkey.. gonna check that out..
jonnay: found it at phpclasses.org
looks good
actually it looks perfect
it must be new because I looked for something like this several months ago and came up with nothing
it claims complete DOM level 2 support
I'm going to attempt to copy the xpath stuff from xslt to a php script
The next time I need to do some dom parsing, I'll keep that in mind, it looks hella slick.
*[contains(concat(' ',normalize-space(@class),' '),' vcard ')] <--- is this a pure XPATH string or is it XSLT stuff in there too?
factoryjoe: btw, it's microid.org
not .com
doh!
thx
i clicked through and got very confused
ah
;)
the microid sounds liek the old tehcnorati blog claiming thing
which generated huge hash strings that were not much help
yeah
it doesn't seem very usable
and always got broken in 2 by the emailer
for humans
and it looks like rampant class abuse
yeah
it's not really a "class"
it's more like a marker
same bug as FOAF too
how so?
assumption that you have one hashable email address
email addresses aren't good shared secrets
ah right
[[hcard]] M http://microformats.org/wiki?title=hcard&diff=0&oldid=5599 * Tantek * (+93) Examples in the wild -
[[hcard]] M http://microformats.org/wiki?title=hcard&diff=0&oldid=5600 * Tantek * (+126) Examples in the wild -
Ugh.
I see what you guys mean bout microid.
What a great idea, but implanting the microid in the class is downright unuseable.
its neither human readable, or even very machine parseable.
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.
