Timestamps are in UTC.
[[chat-strawman]] N http://microformats.org/wiki/chat-strawman * Ben Ward * (+10189) Create 'hChatLog' straw man proposal
BenWard++
OK, does anyone know why a Windows keyboard plugged into a MacBook Pro might intermittently spew duplicate/random characters into my text?
Other than it spiting me for writing a longish document.
It's sicked that you'd abuse it in such a manor
*manner
arg nevermind the typos in that sentence...
BenWard - please add your straw man to chat-brainstorming instead
tantek: I was going to link to it from there, with it being quite long?
oh is it?
nah, go ahead and put it inline
OK
much better to be able to compare/contrast different brainstorms/proposals
Yeah, that's a fair point.
I'd say try to always start with things inline
and then only move to separate pages when painfully necessary to do so
OK
with the exception of when we already have taxonomic patterns for new page creation such as *-examples, *-formats, *-brainstorming, *-issues, *-feedback, *-faq
thanks!
[[chat-brainstorming]] http://microformats.org/wiki?title=chat-brainstorming&diff=0&oldid=9874 * Ben Ward * (+10206) Example playground - Added strawman
Is it possible for me to remove the -strawman page? Or can only wiki admins do that?
do you see a "delete" link in the tabs at the top of the page?
Nope.
ok, must be admin only then
[[Special:Log/delete]] http://microformats.org/wiki?title=Special:Log/delete&diff=0&oldid=0 * Tantek * (+0) deleted "chat-strawman": contents moved to [[chat-brainstorming]]
Thanks tantek
np - thanks for your brainstorming
we were told at uni today not to call it brainstorming anymore
why, some sort of newspeak?
apparently that's offensive to dyslexic people
I heard that once too
politically correctness takes another victim
free speech > offensensitivity
Political Correctness is losing its touch. It gets in the newspapers and tells us we can't say things anymore, but no-one is listening. I'm pretty sure _more_ people say 'nitty gritty' now than before someone tried to take it away.
if not brainstorming than what?
brainjams?
;)
factoryjoe: mmmBrainjam.
Makes me hungry for toast, that'll never do.
Well, I shall take my jam on toast hunger and save it for breakfast. I'll have a crack at cleaning up that brainstorm from typos and the like in the morning. 'night all.
night
[[icons]] http://microformats.org/wiki?title=icons&diff=0&oldid=9875 * Chris Messina * (+2098) added web icons
this was on del.icio.us/popular : http://www.whymicroformats.com/articles/2006/10/04/introduction-to-microformats
tantek is Tantek <http://tantek.com> and works on Technorati and develops microformats <http://microformats.org>
cgriego is Chris Griego (-06:00) and a front-end architect with rd2inc.com
Hello
bergie is lives in Finland and blogs at http://bergie.iki.fi/blog/ and Midgard CMS developer
trel1023 is Terrell Russell of http://claimID.com and http://weblog.terrellrussell.com
bear is located near Philadelphia, PA and the build/release grunt for OSAF and an apprentice python hacker
bewest is Ben West and lives in San Francisco, CA. He daydreams about web style software, works at Alexa.com and blogs at http://bewest.wordpress.com/
tantek: on http://microformats.org/wiki/why-are-content-standards-hard you say "note: it was an *accident* that HTML was easy to learn for non-technical authors, designers, artists etc.: "
I'm not sure HTML was easy to learn, considering we are still combating abuses such as using tables for layout
and new people typically use tools to generate bad html because it is so hard to learn
s/it's so hard to learn/they re too lazy to think about learning/g
no, they make bad html because it is forgiving
which is a good thing
HTML is just text formating you know
many people new to html have a hard time learning css... especially when they know tables will make it /look/ like they want
that's not a laziness issue, it's satisificing at work
bergie is lives in Finland and blogs at http://bergie.iki.fi/blog/ and Midgard CMS developer
drewinthehead is the author of hKit and a developer for Yahoo! Europe
bengee is Benjamin Nowack (http://bnode.org/)
McNulty is Ciaran McNulty (http://ciaranmcnulty.com)
morning
trovster is a web developer from the UK who writes on http://www.trovster.com and helps with www.multipack.co.uk
bewest, HTML was easy to learn in comparison to anything else comparable simply evident by the raw numbers of people that did anything with it.
that aspect was an *accident* as it was not an explicit design objective of the creator
tantek - the fact user-agents have traditionally been very lenient about markup probably helped immensely.
indeed that is likely
it also helped that html only helped markup initially very simple data
literally building blocks like paragraphs and headings
i remember when i first saw html
(I think)
there was some sense to it
RDF OTOH...
i read about in a book
and i couldn't make any sense of it
my first experience w/ html
actually
was in 7th grade
7th grade?!
when the class nerd turned in his homework assignment printed out from a web browser -- from HTML that he'd written to format the tables
i think so
I hate young people
;)
b 1981
that's like 13, right?
it would be
I didn't encounter HTML until about '96
when I was... 18
well, that's not that big a difference
i actually had a blog before i knew i did
http://web.archive.org/web/20010404233502/http://www.factorycity.net/
heh
all my beautiful html table layouts...
from back in the day
wow
1998
http://web.archive.org/web/19981212032517/http://www.sandstormpublishing.com/
too bad all the images are gone
If only you'd used @alt...
heh
did y'all see the mf icons?
drewinthehead is the author of hKit and a developer for Yahoo! Europe
http://factorycity.net/projects/microformats%2Dicons/
[[icons]] http://microformats.org/wiki?title=icons&diff=0&oldid=9876 * Chris Messina * (+57)
What would the 'CV' button do?
hresume
What's the action that occurs when you click it, I mean
ah
well
you'd subscribe to someone's resume
the h(Atom/Card/Calendar) ones all have obvious purposes
like you're supposed to do w/ hcard
in your resume subscription program :)
right
like emurse or linkedin
oh yes I forgot about my resume reader....
;-:
(I've been working on them... they're getting there!)
but yes
so you'd publish hresume on your website
and you'd tell them to grab your hresume from your website -- to "subscribe" to it
so that when you make changes, it's always up to date
and i'm a job seeker
sure
i look at all your rel-me links
and find the hresume
and put it in my hresume-reader basket
you're an employer, you mean
and when you come in for your interview i had up-to-date info
err
yes
that's it
Does anyone have a list of attributes that can take URLs?
I'm writing a web service thingy and need to resolve the URLs in some HTML to be absolute
@href, @src ... ?
ah, @action
never mind, found it from the HTML spec
@data?
heh, too slow
there seem to be Looooads
URL?
(to the spec)
http://www.w3.org/TR/html401/index/attributes.html (the ones with Type=%URI)
ta
that's a few more than I was expecting
@action, @background, @cite, @classid, @codebase, @data, @href, @longdesc, @profile, @src, @usemap
I'm not 100% sure which is easiest, Xpathing all of those or just finding http(s)://es
@valuetype in a param might be a URI if @valuetype is set to "ref"
that might be a bit of a stretch though :p
uh, that should be @value in a <param> might be a URI if @valuetype is set to "ref"
[[rel-home-issues]] http://microformats.org/wiki?title=rel-home-issues&diff=0&oldid=9877 * Phae * (+38) Rel-Home of Rel-Index -
hmmm
true, but let's assume that's out of scope for uF parsing ;-)
if they're relative paths you're converting to absolute, they wouldn't have a protocol
d'oh, true
I'm basically thinking of writing an include-pattern proxy
and might not begin with a / either
hrm
that can adopt slightly different include-path methods
so that you can use it with services that haven't implemented it yet
And hopefully it'd spark a bit of debate about which include strategy to use
[[chat-brainstorming]] M http://microformats.org/wiki?title=chat-brainstorming&diff=0&oldid=9878 * Ben Ward * (-200) hChatLog Strawman Proposal - – Typo corrections.
BenWard is Ben Ward of http://ben-ward.co.uk (+0000/+0100 GMT)
well, it's friday
and so I give to you: www.catsthatlooklikehitler.com
cats that look like data would be better :D
einsteins cat?
that was pretty quick, KevinMarks :)
you're more awake than I am, clearly
by data do you mean noughts and ones?
or the android in star trek?
oh, is this a reference to this? http://monster-island.org/tinashumor/humor/eincat.html
indeed
very good
Ronnos is Ron Kok, a friendly student Communication and Multimedia Design in The Netherlands
how annoying.
if you use -moz-border-radius on a UL, and a background on the contained LIs, the background of the LI obscures the rounded corners of the UL
and I thought the slapdash aliased rendering was bad enough as it was...
I guess that's why it's still a moz-* ;-)
I imagine so.
I can't figure out why they'd use a non-antialiased method anyway
surely creating anti-aliased rounded edges is trivially easy in any modern 2D graphics API
I dunno how mozilla renders stuff.
well neither do I. I am making uninformed guesses that are quite possibly spectacularly wrong.
gripe #2 for the morning: list-style-image
:|
you had me thinking my griping was excessive there.
well it is a little ;-)
What's wrong with list-style-image?
if the image is bigger (taller) than the height of the LI, then the text in the LI flows to the bottom
I can't think of a reason why you'd ever want that by default
either you'd want the text to align to the top, or you'd want it centred
sorry, I don't get what you mean.
I'll show
I'd *expect* it to clip
the image.
it doesn't do that either
maybe you can specify it with attributes I don't know of
http://www.kapowaz.net/teamphone/damnyouli.html
huh, it's changed the height of the LI?!
indeed!
what cockwittery is this?
ah it's not fixed-height
intersting though
you can likely work around it using background image instead
not in my case
oh?
that would require the background colour I have extend to full width of the UL
and I want the UL colour to show through behind the list-style-image
reload and you'll see
Ah you want it to stand outside the LI
yes
hm
I'm baffled by this behaviour.
I cannot for the life of me think why that would be a desirable effect
I really dont' see why it's reflowing like that
me either
drewinthehead is the author of hKit and a developer for Yahoo! Europe
I don't suppose you're on a Mac are you?
I'll test it in Opera
does the same
although interestingly, it appears the image is more right-aligned than the Firefox one
IE7 does something entirely different
although equally baffling
maybe you could just use a padding and a background image that contains the check mark as well as the solid background?
possibly, but that's really quite hacky
I should point out that this test has a bit of a caveat; the tickmark has a lot of padding in the image itself
but the point still stands that the text oughtn't reflow like that
I'm curious to hear what Safari does
expandes the li
but I think it's the right behaviour
it's an image
not a background image
indeed
well, it's neither
it's a list-image
a more common way of doing that is to make it the background image
aye, and why is that more common?
perhaps because the specified way of doing it is problematic?
because if it's the "bullet point" the li will be made tall enough so the "bullet point" is totally visible
that itself is logical
but I'd argue the text shouldn't reflow that way
either that or the visibility of the image should be tied to the height of the LI
I'm thinking of all the other places where an image is used instead of a bullet in lists, things like file listings, icons next to menu items etc.
nowhere would you want the text to reflow as if it was aligned to the bottom of the image
kapowaz: can you set text-align?
maybe it works
well hang on, that's only for horizontal alignment isn't it?
you mean vertical-align, right?
ah, vertical-align
and no, that doesn't do anything
some things in CSS just don't do what they say on the tin
oh wait!
yeah, I think vertical-align only works on table cells
it works in IE7!
*** IT WORKS IN IE7 BUT NOT FIREFOX 2 OR OPERA 9! ***
but IE doesn't support display: table-cell on anything
if it works in IE but not in the other browsers then it certainly must be a bug ;)
in the other browsers...?
;)
in this case, IE is behaving as I'd expect the specified attribute to
I have a containing element (an LI) and within it, another element (an A)
and the LI has vertical-align:top; and the A is flush to the top of the LI
that's behaving as expected. the question is whether or not LIs are allowed to have vertical alignment specified
I can't think why they'd not
isn't the rule of thumb that if they're in a display mode of block then they should be allowed, but not if they're inline?
lis have display: list-item I think
oh yes, of course
actually I am inclined to think it is a bug after all
vertical-align:top; and vertical-align:middle; produce the same result in IE7
it's at the bottom in safari as well
and I know why
if you have text and image, then the image normally is aligned to the baseline of the text
and that's what's happening here
it is hideously inappropriate though
why would anyone ever want that?
as if that would be of any concern to them ;)
quite.
(what people want, not that you're shaking your fist)
aye.
so... time to ask WHAT-WG to propose a new CSS attribute
list-bullet-image or something
that behaves how a bullet would
or maybe list-image-align
nah, that's overkill
I think the CSS3 LI::marker pseudo-selector might do what I want
since you can specify margins etc. in there
but what supports it, I wonder?
just use a background image, multiple background images, generated content or what you just said :)
well for now obviously I'll have to consider one of those
ha, well, your newly invented spec isn't even a draft yet ;)
but ideologically it seems pisspoor that the intended purpose of list-style-image is handicapped by this nonsensical adherence to traditional image flow
okay, take another look at http://www.kapowaz.net/teamphone/damnyouli.html
tell me that's not insane
so long as the image is less tall than the line-height, it will be positioned logically, i.e. at the centre of the first line
no
in safari, it's always at the bottom
just try li img { vertical-align: top } or li { vertical-align: top } and see what happens
already tried that, didn't make a difference
I don't suppose I could see a screenshot of what it looks like in Safari?
the whole point about it reflowing to accomodate the image is asinine given that it allows the image to flow off the left edge of the screen (in FF2 at least)
http://static.last.fm/depth/screenshot.png
hmm, that's essentially the same as I get in Firefox
there is no way I can bodge this to work though, IE7 is just too damned weird
use a background image
yes :)
I shall
anyway, now I shall go to lunch
bon appetit
csarven is Sarven Capadisli and can be found online at http://www.csarven.ca
Atamido is Paul Bryson, http://orangeman.commo.de/
kensanata is blogging at http://www.emacswiki.org/alex/
pnhChris is Chris Casciano, blogs at http://placenamehere.com/ , and a member of the Web Standards Project.
drewinthehead is the author of hKit and a developer for Yahoo! Europe
this might seem pedantic, but I'm wondering why the hReview spec (http://microformats.org/wiki/hreview) outlines 'Semantic XHTML Design Principles'
isn't that a bit beyond the scope of such a spec?
it seems XHTML design principles is a general microformats thing
well, whichever. I mean, it's giving you technical information which extends past the purpose of the document.
would it make sense then to simply refer to another wiki entry on XHTML design principles in each spec?
http://microformats.org/wiki/semantic-xhtml-design-principles
there's a similar section in xfolk
it appears to use the exact same text.
so maybe just nixing that section and saying 'read about it here' would make sense?
brb, checking something relevant
from the wikipedia practice in this instance: {{main|semantic-xhtml-design-principles}}
followed by the first paragraph
XHTML is built on XML, and thus XHTML..... blah blah etc etc
I don't follow, is that your suggested alteration?
assuming that template is on the microformats wiki, or can be added there, then yes
sreynen is Scott Reynen, who makes things at makedatamakesense.com
hell, you could template up that blurb into a single article, so it's just a case of adding in {{semantic-xhtml-design-blurb}} so the data isn't duplicated
anyway a more pressing issue has my attention right now... where can I get the microformats logo?
I don't see it on the wiki, and it isn't obviously available on microformats.org
kapowaz: http://microformats.org/wiki/spread-microformats
SVG and PNG on there
cool
I don't know why, but that page is the 17th search result when I search for 'logo' on the wiki
bad SEO :-D
yeah loads of pages have "logo" in them
I'll create a redirect
I just remembered someone mailing uf-discuss with one and asking if they could put it up
[[logo]] N http://microformats.org/wiki/logo * Kapowaz * (+33)
well, there you go.
hmm, doesn't quite do what I intended... it's preserving the URL
briansuda is brian suda of http://suda.co.uk and is at (-0000 GMT) and is author of "Using Microformats" for O'Reilly [http://www.oreilly.com/catalog/microformats/]
would anybody be interested in EPS and Photoshop versions of the microformats logo?
GIMP supports SVG
although it probably would be good to post EPS versions to make it easier for peeps
can GIMP export to Photoshop foramt?
er, format
polluted with shareware toss that doesn't do anything of the sort, but has all terms on the page
hmm
Photoshop can read EPS files right ?
yes but not in editable form
ah
well
no wait, I'm thinking of Photoshop PDF there, yes, it can read EPS but it rasterises it immediately
imports them as raster doesn't it?
boneill: yup
Photoshop PDF would be the ideal format for distribution IMHO
what is the earliest version of photo shop to support PDF importing/editing ?
a good question
and which version of PDF, also
so it may not be well supported in earlier versions
nice :)
those search results were polluted with pure SEO evil
yeah, I really want to be redirected to a PC hardware reseller's homepage
I've got the PNG here
the weird thing is, the logo being used on the wiki isn't the same as the PNG
those white borders are being cut out
instead of shown
SignpostMarv: are you going to start extolling the virtues of GIMP? Because I'm quite, quite bored of hearing that... it's on digg every five minutes.
i've no idea how good GIMP is because it runs too slow on my windows box- there's like a 2-3 second lag on it when i'm using my graphics tablet
fair enough, just so long as you're not about to start proselytising
yes it can export to EPS
what about to PSD?
yep
EPS isn't really much better than PNG, except that I can choose arbitrary rastering size
ooh
would you be able to save a copy and host it up?
ah, no. Not got any hosting space that wouldn't collapse
ah
wait a sec- it appears GIMP rasterizes the SVG upon opening it
joy.
I would suggest contacting the original creator
I'm doing so :)
if I had Illustrator here it'd be fine
I've a program which can open SVG and export as PDF
Inkscape
horribly Linux-y interface, but it does the job
does it export as native vectoryness ?
drewinthehead is the author of hKit and a developer for Yahoo! Europe
Inkscape's primary format is SVG
so yeah
I can actually do what I need with it for now, I think
since I can export to PDF then raster the PDF in a chosen size
but the layers won't be editable
anyway I've emailed Remi
I shall revisit the matter once he replies
ajturner is Andrew Turner, a simulation and geolocation nut who blogs at http://highearthorbit.com
drewtimestwo
well, sometimes i think i'm with two also :)
like the moments i hear those voices in my head
cgriego is Chris Griego (-06:00) and a front-end architect with rd2inc.com
...
sometimes i wish i was 2
would get a lot more done
any prototype-aware folks in the house?
the #prototype channel is, as ever, dead as a duck
a dead duck
that is
lol
cloning and re-assimilation is quite handy indeed pnhChris
speaking from experience, eh?
you must be identified to NICKSERV on freenode in order to send PMs
ah
I wish Trillian would tell me these kinds of things. I forgot the password I registered this nick with
Colloquy does the identification automatically
Trillian! Got skinz?!?!1onelolz
mirc does it for me when i connect
most *proper* IRC clients do it
i can recieve PMs you send me though right ?
sorry SignpostMarv. that's hypocritical of me after what I said about GIMP earlier.
hehe :P
so are you free to talk in PM Tantek ?
not really SignpostMarv
well fingers crossed on the cloning thing then.
danja is Danny Ayers, http://dannyayers.com
hey, we're gonna need to turn off editing over the weekend for this sever move-operation... shoot.
tantek is Tantek <http://tantek.com> and works on Technorati and develops microformats <http://microformats.org>
woaw
the uF wiki is going crazy or something like that :S
database errors
Ronnos, could you provide URLs where u r seeing errs?
http://microformats.org/wiki/Main_Page
hm, looks like there are more pages with this problem
hmmm... just went there and didn't see any database errors or any other problems
what in particular r u seeing/
?
well, the error message is in Dutch :)
MediaWikiBagOStuff:_doquery". MySQL gaf the foutmelding "3: Error writing file '/var/log/mysql/mysql.log' (Errcode: 30) (localhost)
syntaxerror in the database
tantek, only this page ( http://microformats.org/wiki/Special:Recentchanges ) works fine
vmarks is in NC
tantek, working on the wiki?
Ronnos: we're working on it
filesystem problems
aha okay!
have to send someone to the colo to reboot it
kingryan: where is it?
actually, I don't know
commercenet takes care of it for us
they can't get anyone there until 4pm, though
but its past 4 :P
not where the server resides
mere technicality
drewinthehead is the author of hKit and a developer for Yahoo! Europe
BenWard is Ben Ward of http://ben-ward.co.uk (+0000/+0100 GMT)
whoa. http://dig.csail.mit.edu/breadcrumbs/node/166
remi is Remi Prevost, a web developper (yeah, that's how we spell "developer" in french) from Quebec and blogs about web stuff at <http://remiprevost.com/>
wow. "The attempt to get the world to switch to XML, including quotes around attribute values and slashes in empty tags and namespaces all at once didn't work."
wow
"The plan is to charter a completely new HTML group."
re: "As always, we will be insisting on working implementations and test suites." - if so, then all so-called RECs without test suites (<cough>XHTML</cough>) MUST be set back to CR status immediately. Otherwise the "as always" / "insisting" qualifiers/statements are false.
blog that
wow
I wonder if Hixie, et al are involved
no links yet http://technorati.com/search/dig.csail.mit.edu%2Fbreadcrumbs%2Fnode%2F166?sub=toolsearch
yes, i do, i was reading his blog
he's wrong that whatwg has no process
it has a very well defined process
it's also a very simple process
yeah
maybe its too simple for him to recognize as a process?
i listen to people, i make a decision, i publish, rinse, repeat.
and you promise to respond to all feedback, right?
yeah
sounds like a process to me
that promise is so gonna bite me in the ass :-P
well, you don't promise *when* you'll respond
yeah
but i have literally a thousand or more outstanding e-mails by this point
och
ouch*
you need help :D
hehe
i don't trust anyone else :-P
hehe
you know what they say...
neither would I
Ian "Mulder" Hickson
"if you want it done right, you have to do it yourself"
any ideas on who/what the new group will look like, or would that be pointless speculation?
DanC might be able to help :D
since he's been dropping hints about this new WG for awhile, it seems
I'm not optimistic about the new group being able to catch up with WHATWG.
hmm... I was expecting something like http://dig.csail.mit.edu/breadcrumbs/node/166 , but I was expecting it in http://www.w3.org/QA/
If Hixie decides to continue WHATWG either in parallel or inspite, I know where I would put my bets on specifications that are implemented and stable first, despite the fact that he is 1000+ emails behind.
Until the weekly telcon + quarterly f2f mtg working group model is dropped and rethought, the rate of progress of working groups will be quite limited.
I find weekly telcons pretty important as sync points to go along with mailing lists.
and there's a time and a place for f2f meetings; we're not using them for the GRDDL WG, for instance.
I find sync points to be limiting by the nature of the single threaded/speaker audio medium and the requirement of temporal synchronicity.
weekly telcon sync points that is
I found the "office hours" thing pretty much essential to effective collaboration on hCard tests.
IRC > phone
could be just my work style.
imho telecons are useless, face-to-face meetings more so, at least in terms of cost-per-unit-benefit.
IRC bandwidth > audio speech bandwidth
semantically speaking
there is a great benefit to 2-person and 3-person meetings in person with a whiteboard
but anything more is usuless
useless, rather
though ironically requiring less bitrate bandwidth
as a means of collaboration
DanC: IRC office hours are much different than a formal weekly meeting
it's easy to do a f2f meeting poorly. but done well, they're worthwhile, in some cases.
voice/speech is better at conveying emotion, but that's about it
how so, kingryan ?
IRC is easier to slice in while multitasking
DanC: in every case where i've been involved in a group with f2fs or telecons, the rate of progress has been dramattically -- orders of magnitude -- slower than in groups where i have had just a mailing list.
the informality of office hours allows us to meet as necessary. no one is required to come. there's no organizational overhead of voting/scribe-ing/etc.
(or just a mailing list and informal meetings online or in person)
and no participant stress of not voting/scribing/ etc.
it would seem that converting the weekly telecon's to IRCcon's could lead to a productivity boost
makes it easier to informally participate and incrementally complete small tasks
Hixie, it's entirely possible that all the times when you've done email-only collaboration have been groups of like-minded, trusting people. It's easy to go fast in that situation.
DanC - no need to assume that without supporting evidence
we use IRC during telcons
but it is secondary
DanC: the whatwg is a group of several hundred people, and believe me, they're not all like-minded :-)
voice still requires too much attention bandwidth
I didn't assume; I'm just challenging the causality
good point Hixie, I don't know of any W3C wg with several hundred people
Hixie: however, it does seem that people trust you in whatwg
:D
also, the "several hundred people" can participate informally
there are lots of W3C mailing lists with several hundred people; e.g. www-tag
DanC: but having said that, any group where the participants have different goals is going to result in a poor standard and little progress. You're far better off having two standards each targetted at the actual use cases than a compromise standard (since in the latter case neither use case will actually be well handled)
at least in my experience
which i admit is limited
true, it's better to not pretend people want the same thing when they don't
(i've only been involved in a dozen or so standards so far, in six or so groups)
well, mostly. sometimes you just have to have a standard for railroad track size, even when people don't like each other.
i disagree
if you have a train that goes up a steep hill, it makes no sense to use the same gauge as a train that goes on a plain.
and indeed, they don't use the same gauge.
DanC: I'm not sure that analogy effectively maps to the digital realm
similarly, in model trains, if you want to make a model of a single station and have a lot of room, you'll want a much bigger gauge than if you want to model an interstate on a small table.
fair enough; I'm happy to leave the head-butting standards efforts to somebody else in any case.
:-)
DanC, the incidental costs of experimentation and competition among various choices in the virtual world is much lower than the same in the physical world.
(the original xbl group was an example of this, btw)
Jon Bosak's charter for the XML WG was carefully designed to produce a standard even if the participants disagreed. I suppose the results speak for themselves, in a way. ;-)
no comment
XML sucks in various ways, but I think it has done more good than harm, on balance.
sure
but it's utterly failed in some spaces
sure
XML really needs to be split in two, imho
but that's another problem altogether
tantek: actually, it's better than hearing bandwidth, not speech bandwidth, right?
gah... missed the timing on that :( ignore me
I really like the wiki/irc/mailing-list combo. But I do miss 2 things: (1) a roadmap (2) periodic checkpoints/updates to the roadmap
bewes1 - no, I can copy and paste into IRC, I can't copy and paste into the phone.
i.e. if I have a goal for hCalendar, there's no telling how much time I need to budget to get it done. Granted, nobody has a perfect crystal ball, but I think planning, with some flexibility, is useful.
danc: speaking of roadmaps (and i mean this without the slightest intent at offense), what is it with wg charters in the w3c and crazy unrealistic timetables?
DanC, you have to have more faith in organic incremental process than top-down hierarchically planned incremental process.
DanC: e.g. the webapi charter had a first public wd date of _before_ the group's start date
indeed, Hixie, I find that sort of thing unacceptable. Note that the groups I'm involved in keep public schedules, and we update them from time to time.
Even the word "roadmap" assumes that you can/should map out where the roads should go any appreciable point into the future.
As opposed to carefully following the cowpaths and natural foottrails.
you can have small roadmaps though :)
csarven, but that's not what happens in practice.
in practice, roadmaps typically map out roads into fields that the market couldn't care less about.
i supposew3c is taking a more holistic approach in most cases
and then people blindly believe/follow them, because they like the prefer artificial stability of an imaginary roadmap to the the natural instability of the real marketplace.
s/the the/the
true that
i think the problem is there is too much of an attempt to solve bunch of things as supposed to solving a specific small problem
or to generalize even further, people would rather have blind faith in illusion and myth, than accept the uncertainty of the natural world
csarven, yes, that is one problem
DanC: (i just tried to send you a /msg but it looks like this network only allows /msgs if you're authenticated)
tantek, do you completely rule out any sort of risk aggregation at all?
yes, I got your /msg Hixie; I guess you can't get my replies
odd; you'd think it would go the other way
i'm on w3's network too if that helps
well, I think I'm gonna go back to "day off" mode.
tim would pick my day off to launch that thing. oh well.
heh
DanC, I'm not sure what you mean by risk aggregation
another time, I hope, tantek
np
there are several of us that are organically committed to roadmap-ey-type things... see all the efforts on reorganizing the wiki, IA...
bewes1 - projects are not roadmaps
roadmaps typically have artificial datetimes associated with them
the efforts you describe are simply a set of projects / designs in progress that will complete somewhat organically and unpredictably
and with uneven spurts in progress
but the value isn't in the list of dates
all of which is just fine
i wish i knew more about how w3c manages their budget
money -> Semantic Web
:)
im not sure if i understand that inference :)
if you have money, then semantic web?
money goes into semantic web?
csarven, boneill: Perhaps that admittedly debatable inference can be reversed if we finish the currency Microformat?
lol
it's how academics work, similar to the underpants gnomes of South Park.
1. Semantic Web, 2. ?, 3. Profit
or Research -> ? -> Profit :)
You have to have 4 points for that to work Ben
1. Something 2. Profit
I realised that after I'd started, now i just look silly
Maybe _that's_ why we're not all writing RDF, they're missing a point.
tantek: regarding speech bandwidth, I was going to point out that the I/O aren't equal bindings in this case... everyone can talk at once with no bandwidth problem. the problem resides in the fact that it becomes difficult to understand more than one thing at the same time :-)
? how many "Ben"s are there in here?
not enough
3?
Sounds like it.
oy
There's also Ben Buchanan on uf-discuss, but I don't know if he IRCs
And a few posts from a Benjamin Carlyle. Is 5 enough to have something important named after us?
I should think so!
hBen
A microformat to differentiate between people asking for 'Ben' in IRC, to prevent our IRC client's causing alerts at the wrong time?
Maybe I should roll that into the hChatLog strawman…
Although I think there's prior art in surnames.
BenWard: have you taken a look at the stuff I've put up on the hChat pages?
specifically ULF
the site's down
good luck
bewes1 - the other difference is in hysteresis and catchup
e.g. much easier to scroll up and scan an IRC window than to record an audio stream, rewind and attempt to catchup
so I can pop in and out in terms of attention/focus and still participate
kingryan: So I see. Is it intentional downtime?
nope
filesystem corruption problems
Eek.
ironically, the unplanned downtime has gotten in the way of the planned downtime
kingryan: commented
kingryan: i'm quite impressed that i learnt of your blog post from a google alert, btw.
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.
