00:06
<gsnedders>
That station scares me.
00:07
<gsnedders>
The fact that it is the busiest station in Europe by number of trains stopping at it maybe justifies that.
00:44
<Hixie>
hober: probably best to change the lang="" to just lang="en" or at most "en-US" to avoid confusing people :-)
00:49
<Hixie>
hober: the font size is a bit on the big side, but other than that and the lang="", it looks pretty neat
00:53
<da3d>
big fonts are nice at 2am...
00:54
<Hixie>
i just zoomed it out to read it :-)
01:04
<da3d>
hmmm... I haven't heard of the placeholder attribute, I actually learnt something here! :-O :-)
05:35
<hober>
Hixie: the lang="" and the <h2> are jokes
05:35
<hober>
and yeah, I have a big font problem. the bigger the monitor, the bigger the font size in my book
05:35
<Hixie>
the lang="" joke is probably too subtle for most of your audience, and is likely to cause copy-paste issues
05:36
<hober>
hmmm. yeah, maybe I should change it.
07:26
<Hixie>
anyone have any good examples of read-only text fields? (not disabled)
07:32
<othermaciej>
I can think of common read-only textarea examples, but not text fields
07:32
<othermaciej>
(the example for text areas being clickwrap license agreements)
07:36
<Hixie>
ooh, found one in mac os x preferences
07:36
<Hixie>
User Name field in Accounts prefpane when the prefs are locked
07:41
<othermaciej>
sort of - it doesn't behave like a read-only control in that it lets you set a caret
07:41
<othermaciej>
it would be more accurate to say it is read-write, but intercepts typing and poses a challenge before typing can take effect
07:41
<othermaciej>
but close enough I guess
07:50
<Hixie>
yeah
09:37
<annevk2>
Hixie, the password example is wrong
09:37
<annevk2>
Hixie, can't use onforminput there
09:37
<Hixie>
why not
09:37
<annevk2>
Hixie, maybe onchange
09:37
<annevk2>
Hixie, because the error will hilited the moment password1 gets a single character and from thenon
09:38
<Hixie>
so?
09:38
<Hixie>
that's the idea
09:38
<annevk2>
with the UA potentially focusing the second field?
09:38
<Hixie>
why would the UA focus the second field?
09:38
<annevk2>
because it's invalid
09:39
<Hixie>
so?
09:39
<Hixie>
fields become invalid all the time
09:39
<Hixie>
the UA isn't going to focus them whenever they get invalid
09:39
<Hixie>
that'd be UI from hell
09:40
<annevk2>
hmm, maybe I misunderstood setCustom..
09:41
<Hixie>
it doesn't do anything except flip the "invalid" bit
09:43
<annevk2>
zcorpan, 21st century started in 2001 per Wikipedia at least
10:01
<Lachy>
indeed, wikipedia is actually correct in this case, cause there was no year 0, the century went from 1901-2000
10:25
<jgraham>
But, on the other hand, people who insist on using that definition despite the common usage of zero-based centuries are silly
10:33
<hsivonen>
http://www.douglasadams.com/dna/pedants.html
10:41
<Philip`>
Is it actually true that people considered 1900 to be the start of the new century?
10:42
<GarethAdams_>
of course
10:43
<GarethAdams_>
don't you remember 1st Jan 2000? well they weren't any cleverer the previous century
10:43
<Philip`>
Picking a totally random example, http://books.google.co.uk/books?id=cfAGav7e1FUC&lpg=PA750&as_brr=1&pg=PA750#v=onepage&q=&f=false mentions "The first year of the new century" in 1901
10:43
<Lachy>
Philip`, that just proves that at least some people did some fact checking before publishing back then.
10:44
<GarethAdams_>
oh, http://www.economicexpert.com/a/Zero:ists.html says differently
10:45
<GarethAdams_>
although there's no source there
10:46
<Philip`>
GarethAdams_: So it's just as well sourced as http://www.douglasadams.com/dna/pedants.html :-)
10:47
<GarethAdams_>
strictly speaking, yes
10:47
<GarethAdams_>
:P
10:52
<jgraham>
Well I do and Zcorpan did so, pedantically, that counts as "people"
10:55
<jgraham>
But I think the experience of 1999 shows that a) almost everyone considers years ending 00 to be the start of a new century and b) there is a hard core of boring people who will write endless letters to newspapers about how everyone else is wrong
10:55
<Lachy>
it doesn't really matter when people choose to celbrate the beginning of a new century. That's just any 100 year period. People could celebrate it next year on 2010-01-01 if they like
10:57
<jgraham>
But, in fact, everyone called 2000-01-01 "the start of the 21st Century"
10:58
<Lachy>
they were wrong
10:59
<Lachy>
luckily facts are not determined on the basis of popularity
10:59
<Philip`>
I hope you're not proposing to accept popular opinion over factual argument
10:59
<jgraham>
It's a naming convention. Popular argument wins in naming conventions
11:00
<jgraham>
It happens to not corresponf to 21 cenuries since the BC/AD changeover but so what? No one cares about that
11:01
<jgraham>
It is conceptually much simpler to label centuries as starting in 00 years. So everyone does
11:03
<Philip`>
Only everyone in the past few decades, it seems
11:04
<GarethAdams_>
<Philip`> I hope you're not proposing to accept popular opinion over factual argument
11:05
<GarethAdams_>
^^that's an unusual point to make in the HTML5 channel ;)
11:06
<Philip`>
Picking another totally random example, http://books.google.co.uk/books?id=NjkIAAAAQAAJ&as_brr=1&pg=RA1-PA443#v=onepage&f=false says "As the union between Britain and Ireland was to commence the next day, which was the first of the new year, and of the new century, ..." (referrin to 1 January 1801)
11:06
<Philip`>
s//g/
11:08
<GarethAdams_>
and s//g/ is not the right expression to use in a room of pedants ;)
11:08
<Philip`>
http://books.google.co.uk/books?id=JaYMAAAAIAAJ&as_brr=1&pg=RA1-PA595#v=onepage&f=false - "The commencement of the new century, 1801, ..."
11:08
<jgraham>
Philip`: Using books as a source is a big election bias because they have an unusual tendency to be pedantically correct
11:08
<jgraham>
*selection
11:09
<Philip`>
It's better than using no source at all
11:09
<jgraham>
Philip`: Not necessarily, if the answer it gives is misleading
11:09
<GarethAdams_>
http://www.googlefight.com/index.php?lang=en_GB&word1=new+century+1900&word2=new+century+1901
11:10
<Philip`>
GarethAdams_: It's been well-established that IRC regexps are highly context-sensitive and only apply in the location of the relevant string where they make the most sense
11:10
<jgraham>
Anyway it's what people believe today that's important. Otherwise we would all insist in talking about Web Founts
11:11
<Philip`>
jgraham: I thought the WHATWG approach was that any data is better than no data, regardless of potential misleadingness
11:12
<jgraham>
Philip`: That seems like a bad approach :)
11:12
<Philip`>
e.g. a video of a single user is useful evidence about how users in general interact with accessibility tools
11:12
<jgraham>
Yeah, that was somewhat dubious
11:13
<Philip`>
jgraham: It's a problem if people believe today that most people believed n (>0) hundred years ago that centuries started in '00, if that's factually inaccurate
11:14
<Lachy>
although that video was better than nothing, there were so many problems with it that made getting usable data from it impossible
11:14
<Philip`>
GarethAdams_: That just shows when modern people think the 19th century started, not what 19th century people thought
11:14
<jgraham>
Even assuming the common definition has shifted, I think you have epsilon of a problem there with epsilon->0
11:15
<Philip`>
I'm assuming n is an integer
11:16
<GarethAdams_>
so was 2000 in the 20th or 21st century?
11:16
<Philip`>
GarethAdams_: Yes
11:16
Philip`
is reminded of the old maths joke, "Let epsilon be a large negative number."
11:16
<GarethAdams_>
so which was 2000 in; the 20th or 21st century?
11:16
<and>
(One should use a different naming convention if one wants a different definition of century, like Italian Quattrocento or Swedish 1900-talet.)
11:16
<jgraham>
GarethAdams_: According to anyone you'd like to meet at a party, the 21st
11:16
<Philip`>
GarethAdams_: The 20th according to the definition that makes the most mathematical sense
11:17
<Philip`>
(and that doesn't make the 1st century either one year shorter than any other century, or start in 1 BC)
11:18
jgraham
is happy with the first century having 99 years in the same way that the year with the calendar change in it has the wrong number of days
11:19
<jgraham>
(unless people would like to measure centuries in astronomical units, in which case it isn't the first of January in any year)
11:19
<jgraham>
s/it/the chnge/
11:19
<Philip`>
Can we forget about centuries and just use seconds since the UNIX Epoch?
11:24
Philip`
remembers the original Time Machine movie was set at the start of the 20th century, but can't remember whether the dials on the machine said 1900 or 1901
11:24
<Philip`>
(The book doesn't seem to say anything about the year it was set in, though)
11:37
Lachy
sets his calendar to for a celebration on 2033-05-18T03:33:20Z
11:38
<Lachy>
that's 2,000,000,000 seconds since the Unix epoch
11:39
<Philip`>
What significance does that number have?
11:39
<Philip`>
(except in decimal, which isn't significant)
11:40
<Lachy>
it is for the same reason that 1,234,567,890 seconds was significant back in February this year
11:40
<Hixie>
Randy Pausch's "Really Achieving Your Childhood Dreams" lecture is fantastic, in case any of you care
11:40
<Hixie>
(it's on iTunes U)
11:40
<Philip`>
Lachy: i.e. it's not significant at all?
11:42
<Philip`>
(It's on YouTube too)
11:42
<Lachy>
well, I tried lookng checking what date 1000 0000 0000 0000 0000 0000 0000 0000 (binary), but I the online calculator I found broke
11:42
<Philip`>
$ perl -le'print scalar gmtime (2**31)'
11:42
<Philip`>
Tue Jan 19 03:14:08 2038
11:43
<Philip`>
That's a much more significant event
11:43
<Lachy>
oh, yeah, that's the 2038 problem
11:44
Philip`
is now prepared for that problem, having upgraded to 64-bit Linux
11:44
<Hixie>
Philip`: so it is. hadn't heard about it until just now where i stumbled across it completely by accident
11:45
<Lachy>
Apparently OS X Leopard isn't prepared for it: Fri Dec 13 20:45:52 1901
11:45
<GarethAdams_>
Hixie: really? I'm surprised you didn't know about the 2038 problem
11:46
<Philip`>
Hixie: That lecture was quite popular when it was first released
11:46
Hixie
was not talking about the quite uninteresting 32bit time_t rollover issue
11:46
<Philip`>
which I suppose is obvious given the 10 million Youtube hits
11:46
<Hixie>
Philip`: yeah, so i've now discovered from wikipedia :-)
11:47
<Hixie>
i've been watching a lot of walter lewin's physics lectures, which i've seen clips off before, but which are quite fascinating in their entirety
11:47
<Hixie>
he's far more compelling than my lecturers were
11:47
<Hixie>
it's also a lot more fun to learn this stuff when you don't have an exam looming
11:48
<Hixie>
("learn" might be the wrong word.)
11:48
<Hixie>
("revise", maybe)
11:51
<Steve^>
Don't suppose there's an HTML5 outline tool available for download?
11:51
<Lachy>
Steve^, you should be able to get the source from gsnedders
11:52
<Steve^>
ok
11:52
<Steve^>
and then hopefully someone will add it to dragonfly too
11:53
<Lachy>
it would have to be rewritten for that
11:54
<Lachy>
but I suggest you file a feature request bug for it
11:57
<Steve^>
can I still do that now that I've left Opera?
12:32
<Lachy>
Steve^, anyone can file a bug report
12:33
<Lachy>
the only problem is that our bug tracker isn't open so you won't be able to track the progress
12:42
<Steve^>
that's interesting, gsnedders' outliner understands CSS?
12:42
<gsnedders>
Steve^: No
12:42
<Steve^>
wait no
12:42
<Steve^>
I've just not used CSS properly somewhere :P
12:42
gsnedders
leaves again
12:42
<Steve^>
bye
13:06
<Philip`>
Lachy: https://bugs.opera.com/wizard/ seems to indicate it's just for bugs, and feature requests aren't suitable
13:10
<Dashiva>
Argue the feature is so useful not supporting it is a bug
13:12
<Philip`>
How do you describe how to reproduce the lack of feature in three or more steps?
13:12
<Dashiva>
Try to use feature. Nothing happens. Expected: Something happens.
13:13
<Philip`>
"Try to use the feature" is only one step
13:13
<Dashiva>
Go to a website where you'd want to use the feature. Try to use the feature. Note nothing happens. Try to locate option to enable feature. Find none.
13:13
<Dashiva>
That's five
13:13
<Philip`>
Maybe you could do "1) Look in the File menu for the feature. 2) Look in the Edit menu for the feature. 3) Look in the View menu for the feature. Expected results: I should find the feature. Actual results: I don't find the feature."
13:14
<Philip`>
Dashiva: You're mixing the outcome with the reproduction steps, which is violating the design of the form
13:15
<Dashiva>
Well, it's still three if you remove the "find nothing" steps.
15:47
<Steve^>
phew, thank god for vim backups. I got overly friendly with rm
16:32
<gsnedders>
Mr Last Week (whom I assume is reading this): Why the song for me in the latest post?
16:36
<gsnedders>
hmm, http://www.w3.org/2009/09/11-hcg-minutes.html really is fairly interesting (MO link)
16:39
<Steve^>
I need your password
17:01
<Pure>
Any help for a html4 dog to learn html5?
17:01
<cardona507>
Pure - start here http://www.whatwg.org/
17:02
<miketaylr>
diveintohtml5.org is purty.
17:03
<Pure>
Just wondering, can the new tags, like canvas, be used with Php
17:04
<Philip`>
They can be used with PHP to the same extent that any other HTML element can be used with PHP
17:05
<Pure>
I mean, can the canvas content be passed as a GET/POST?
17:08
<Philip`>
Do you mean something like uploading the bitmap content of a canvas to a server?
17:08
<Philip`>
If so, you can use canvas.toDataURL() to get a string representing the canvas bitmap (encoded as a PNG), and then upload that
17:27
<Pure>
Also, Local Storage?
17:29
<Steve^>
of what?
17:32
<Pure>
Sounds scary
17:38
<Pure>
If a browser doesn't support a certain feature, are there libraries to emulate the feature?'
17:39
<Philip`>
There are for some features
17:39
<Steve^>
there is for the non-embedded stuff
17:39
<Philip`>
e.g. there's libraries to emulate <canvas> in IE
17:40
<Steve^>
are they any good?
17:44
<Philip`>
No, unless you only want to do pretty simple things
18:05
<erlehmann>
Pure, plugins are more complicated, though. you can have theora and vorbis by adding a qucktime thingy, but the user has to explicitly install it.
18:05
<erlehmann>
also, i think there was a web forms 2.0 javascript library
18:12
<Steve^>
How would you mark up blog comments? I've chosen an article containing an ordered list of articles.
18:14
<erlehmann>
Steve^, following the spec example ?
18:14
<Steve^>
whereabouts is that?
18:15
<gsnedders>
Steve^: In http://www.whatwg.org/specs/web-apps/current-work/
18:15
gsnedders
ducks
18:15
<Steve^>
yes, that's the spec, well done
18:15
<Steve^>
I actually have a PDF version open, so shoo!
18:15
<gsnedders>
ewww, print media ;P
18:18
<hober>
Steve^: http://www.whatwg.org/specs/web-apps/current-work/multipage/sections.html#the-article-element
18:18
<hober>
specifically, "where article elements are nested..."
18:19
<Steve^>
oops, thank you
18:19
<Steve^>
I read it, but not hard enough
18:22
<zcorpan>
javascript:'<frameset cols=*,*><frame src=http://validator.nu/?doc='+encodeURIComponent(location.href)+'><frame src=http://gsnedders.html5.org/outliner/process.py?url='+encodeURIComponent(location.href)+'></frameset>';
18:22
<zcorpan>
bookmarklet to validate and check outline
18:23
<miketaylr>
ooh
18:23
<miketaylr>
rad
18:27
<Steve^>
awesome
18:27
<miketaylr>
thanks zcorpan
18:27
<zcorpan>
np
18:29
gsnedders
wonders if he'll have time to rewrite the outliner this month
18:30
<Steve^>
what's wrong with it?
18:30
<gsnedders>
It could be 100% more awesome.
18:30
<Philip`>
It's software, therefore it needs to be rewritten
18:35
<gsnedders>
annevk2: Is there any reason that scrollIntoView isn't in CSSOM View?
18:35
<hober>
Steve^: using an ordered list is semantic abuse; document order is ordering enough for this case
18:37
<Steve^>
where do you draw the line on that?
18:37
<Steve^>
comments are something that I would actually number
18:53
<Pure>
Is there still a way to use frames?
18:55
<hober>
Steve^: <ol> means that if you change the order of the <li>s, you've changed the meaning of the text. document order provides enough semantic ordering, <ol> too much.
20:09
<annevk2>
gsnedders, it's in HTML5, no?
20:09
<gsnedders>
annevk2: Yeah
20:10
<gsnedders>
annevk2: But why not have it exist on any Element?
20:12
<annevk2>
dunno
20:12
<annevk2>
couldn't care less
20:34
<gsnedders>
annevk2: It seems a needless diff. between Element and HTMLElement
20:53
<annevk2>
if that was the only one I might care more
20:53
<annevk2>
or someone else would
21:10
<othermaciej>
good day
21:23
gsnedders
squint at othermaciej
21:24
<gsnedders>
"good"!?
21:24
gsnedders
ducks
21:24
<othermaciej>
gsnedders: "acceptable"?
21:25
<gsnedders>
That'll do.
21:31
<virtuelv>
Isn't good day mostly used when the intent is insult?
21:32
<virtuelv>
wiktionary supports that definition, at least - http://en.wiktionary.org/wiki/good_day
21:32
<gsnedders>
Uh, not really…
21:32
<Philip`>
It's a possible meaning, but not the most common
21:33
<virtuelv>
the only reason I remember this is because a former prime minister used it towards the then british PM
21:35
<virtuelv>
(or some british minister, at least -- if it was the PM, all the better, because that was Margaret Hilda Thatcher
21:35
<othermaciej>
it depends on tone
21:35
<virtuelv>
of course
21:35
<othermaciej>
if you say "good day" in a huffy, dismissive voice, you are rudely telling someone that you won't talk to them any more
21:36
<othermaciej>
if you say it in a cheerful tone with an Aussie accent, you are warmly greeting them
21:36
<virtuelv>
that'd be g'day
21:36
<virtuelv>
listed separately in dictionaries, even
21:36
<virtuelv>
either way, good night
21:38
<Philip`>
If you don't do the accent, it's less warm but still a suitable greeting when you don't know the listener's timezone
21:38
<Philip`>
or at least that's how I understand it
21:39
<othermaciej>
it's pretty much never rude as a greeting as opposed to as a farewell
21:51
<Lachy>
then what would it mean if I said it dismissively in my cheeful Aussie accent?
21:58
<AryehGregor>
"Good day" isn't rude any more than anything else. I mean, you could say "thank you" in a withering, sarcastic tone of voice too.
22:00
<Philip`>
I can't imagine a way of saying "thank you" that would have quite the same 'go away, I'm fed up with you now and don't want to hear any more from you' meaning that an exasperated "good /day/" would have
22:01
<Philip`>
but maybe I'm just being unimaginative
22:03
<Lachy>
Philip`, it's easy to say thank you in a sarcastic tone, meaning that you're not really pleased with what the other person did
22:03
<Lachy>
I do frequently
22:04
<Philip`>
Lachy: That's not really an end-of-conversation expression, though
22:12
<ivan`>
often "thank you" is insulting because it's too brief and not thankful enough
22:51
<Hixie>
i need a flash app that just makes noise while loaded
22:51
<Hixie>
e.g. something that while running, beeps
22:51
<Hixie>
or plays audio
22:51
<Hixie>
or something
22:52
<Hixie>
(i'm trying to work out when flash stops running, e.g. when removed from a document, made display:none, etc)
22:52
<Hixie>
anyone know of any such files?
22:52
<_crow>
you could grab a flash music player from myspace or something
22:52
<_crow>
...but those stream music, not sure if that's helpful
22:53
<Hixie>
everything i've found so far requires you to press play
22:54
<Hixie>
before it makes sound
22:54
<_crow>
hmmm
22:54
<Hixie>
(or it takes too long to load)
22:55
<Hixie>
e.g. games take so long to load
22:55
<_crow>
yeah
22:56
<_crow>
this looks like it might be useful: http://flashbeep.net/
22:58
<_crow>
and you can bind it DOM events
22:58
<Hixie>
yeah found a few libraries but i'd really not have to rely on calling the api
22:58
<Hixie>
since that just brings something else into the tests
23:04
<_crow>
heh, not sure if this is flash or not...no time to dig thru source, but it plays automatically and loops: http://www.angelfire.com/super/badwebs/
23:06
<Hixie>
cool, thanks
23:31
othermaciej
wishes there were an updated version of http://philip.html5.org/tests/canvas/suite/tests/results.html
23:31
<othermaciej>
I guess it also needs text drawing tests
23:32
Philip`
wishes there was too
23:32
Philip`
needs more motivation to work on it again instead of doing other stuff
23:33
<Philip`>
(Text tests are the main thing, plus quite a few minor changes and additions)
23:33
<Philip`>
((like toDataURL with JPEG, and drawImage with video, etc, I guess))
23:35
<Philip`>
Hixie: By the way, did you see the few times I complained about toc.js in the past few days?
23:35
<Philip`>
It'd be nice to sort that out, so I can make any changes needed to the splitter and then get MikeSmith (I assume) to update the W3C copy
23:35
<Hixie>
Philip`: what's the question?
23:36
<Philip`>
Hixie: toc.js runs on e.g. http://www.whatwg.org/specs/web-apps/current-work/multipage/introduction.html and makes the TOC go all silly
23:36
<Philip`>
so it should be disabled except on the full TOC on the front page
23:37
Hixie
looks
23:37
<Hixie>
oh you have minitocs
23:37
<Hixie>
cool
23:37
<Philip`>
(I think I need class=toc to get the right styling, but that's the same hook the script uses)
23:38
<Hixie>
i've gotta go, but i'll look in a few hours
23:39
<Philip`>
Okay, thanks!
23:39
<Hixie>
is there some way i can tell from script whether we're in the full page or not?
23:39
<Philip`>
I could put a different class on the mini-tocs
23:39
<Hixie>
put it on the <html>
23:39
<Philip`>
(in addition to the class=toc for styling)
23:39
<Hixie>
that way i can disable the <dfn> one too
23:40
<Philip`>
toc.js should still run on http://www.whatwg.org/specs/web-apps/current-work/multipage/index.html, I think
23:40
<Philip`>
(but dfn.js presumably shouldn't)
23:41
<Hixie>
can you set class="split index" and class="split chapter" on the <html>?
23:42
<Hixie>
then when i get back i'll just have the script look for those and enable/disable subscripts accordingly
23:43
<Philip`>
Okay, can do