00:03
<Bevan>
picture++
00:03
<Bevan>
img set--
00:41
<zewt>
Bevan: not helpful
00:58
<Bevan>
zewt. Acknowledged.
00:58
<Bevan>
:)
03:45
<inclement_bore>
’twas two nights after srcset, and all through the chat
03:46
<inclement_bore>
not a creature was stirring, not even a... sig cat
03:46
<inclement_bore>
the use cases were hung on anything near
03:46
<inclement_bore>
in hopes that efficient scalable image delivery over the web ...
03:46
<inclement_bore>
... would soon be here
03:46
<inclement_bore>
developers were nestled, all snug in their beds
03:46
<inclement_bore>
visions of media queries still danced in their heads
03:47
<inclement_bore>
implementers too, not impressed
03:47
<inclement_bore>
had just settled down for a long spring rest ...
04:43
<paul_irish>
gy
04:44
<paul_irish>
(whoops. IRC doesnt start gyazo. ignore.)
05:50
<heycam>
what's the reason appendix sections in CSS specs aren't numbered (like A.1, A.2, etc.)?
05:51
<heycam>
is is that the numbering scheme is ugly?
06:53
<hsivonen>
heycam: at least some CSS appendix naming forms a joke, but maybe that was only for CSS 2 and 2.1
06:53
<heycam>
hsivonen, ah yes, for the appendix letters
06:56
<heycam>
hsivonen, though notice how if you jump to the bottom of http://www.w3.org/TR/CSS21/cover.html that appendix sections get numbers while those in http://dev.w3.org/csswg/css3-text/#contents don't
06:58
<hsivonen>
maybe the letter thing is for CSS 2/2.1 only
06:58
<hsivonen>
Hixie: I've successfully browsed the Web with 1rem == 18px for years
07:36
<webben>
What is the current pipeline for turning http://svn.whatwg.org/webapps/source into http://wiki.whatwg.org/wiki/FAQ#What_are_the_various_versions_of_the_spec.3F ?
07:44
webben
spots http://dev.w3.org/cvsweb/html5/spec/Makefile
07:55
<jgraham>
webben: Brown paper and glue. And magic.
07:56
<webben>
jgraham: So I can see.
07:56
<jgraham>
Hixie has a highly-custom pipeline that is mostly calling various services that others are providing e.g. anolis, the (HTML-specific) spec spliter
07:56
<webben>
it looks like the w3c bits have makefiles
07:57
<webben>
is there a makefile for the whatwg "complete" version somewhere?
07:57
<jgraham>
No
07:57
<webben>
k
07:57
<jgraham>
Most of Hixie's pipeline is closed source because it is very environment specific
07:58
<jgraham>
But the complete version of the spec might be relatively easy to make since it is the most, well, complete
07:59
<jgraham>
http://pimpmyspec.net/ + the right options might get you a lot of the way there
08:00
<jgraham>
Of course, the output is conveniently provided, so I wonder what the motivation is ;)
08:36
smaug____
doesn't always understand what people mean with 'polyfill'
08:37
<smaug____>
ahaa
08:37
<smaug____>
, silly word
09:17
<smaug____>
We should decide at some point what to do with FileSystem API
09:23
<smaug____>
hmm, webapps is the right place for that
09:39
<AryehGregor>
Is there a command-line HTML5 validator that's easy to install and use?
09:44
<MikeSmith>
AryehGregor: I put https://bitbucket.org/sideshowbarker/vnu together experimentally
09:45
<MikeSmith>
can download from https://bitbucket.org/sideshowbarker/vnu/get/2759dcb15031.zip
09:45
<MikeSmith>
or just clone the reop
09:45
<MikeSmith>
*repo
09:45
<MikeSmith>
see the "How to use the pre-built jars" part at https://bitbucket.org/sideshowbarker/vnu
09:46
<MikeSmith>
you can feed it a list of filenames
11:18
<Ms2ger>
<TabAtkins_> othermaciej: Heh, I can't believe we didn't actually define when MQs are true. ^_^
11:18
<Ms2ger>
Really, you couldn't believe that?
12:29
<annevk>
in an async algorithm, are all observable changes queued tasks?
12:29
<annevk>
e.g. manipulation of the fullscreen elemetn stack
12:34
annevk
has the feeling nobody really cares about task queues
13:39
<zewt>
well, task queues are really just partial order guarantees, right?
13:39
<zewt>
(so if an implementation orders *all* queued events, they mostly don't have to care about them)
13:58
<EyePulp>
'morning. [to whom it applies]
16:18
<dglazkov>
good morning, Whatwg!
16:18
<dglazkov>
I think I need to start a Good Morning CG.
16:19
Ms2ger
strangles dglazkov
16:19
<Ms2ger>
Good morning to you too
16:19
<dglazkov>
:)
16:31
<annevk>
hsivonen: curious (re twitter), what should the EU have used instead?
16:32
<companyhen>
Morning all. :D
16:37
ChrisWilson
mumbles something incoherent and goes and gets another cup of coffee
16:48
<Philip`>
dglazkov: You ought to start a wiki page listing the use cases for having a good morning, before anyone will take you seriously
16:54
<EyePulp>
heh - I just got that.
16:54
<jreading>
hee
17:05
<dglazkov>
Philip`: let's not forget the support forum, er.. mailing list.
17:07
<hsivonen>
annevk: E for example
17:08
<hsivonen>
annevk: you know, like FF and DM were plain ASCII before euro
17:12
<annevk>
heh
17:12
<annevk>
apparently our former currency didn't have a code point http://en.wikipedia.org/wiki/Florin_sign
17:14
<Philip`>
On (at least) UK keyboard layouts, the € symbol is stuffed pretty inelegantly into a spare corner of the 4/$ key, which doesn't seem like great design
17:14
<Philip`>
(The ¬`| key is the only other one that typically follows that pattern)
17:15
<annevk>
hsivonen: the French had U+20A3 ₣ but also used F/FF indeed
17:18
<annevk>
some more browsing reveals that http://en.wikipedia.org/wiki/Azerbaijani_manat does not have a Unicode symbol while they do have a distinct sign
17:18
<annevk>
weird
17:19
<Philip`>
http://www.tcmb.gov.tr/yeni/iletisimgm/TurkishLira.php - "The upward pointing parallel lines symbolize the "steadily rising value" of the Turkish lira and the Turkish economy." - does that mean they'll redesign the currency symbol if their economy collapses?
17:20
<annevk>
heh, they got this one because it devalued so much
17:21
<hsivonen>
before euro, Finland use an ASCII-only currency abbreviation "mk" and it worked great
17:22
gsnedders
finds £ annoying with just ASCII — going for GBP as a suffix just seems weird
17:22
<Philip`>
gsnedders: Use "lb" instead
17:23
<annevk>
hsivonen: I typically use the three-letter code because that works great in Google :)
17:24
<annevk>
1000 pen in eur please
17:25
<hsivonen>
in other news, it's terribly inconvenient that Ubuntu 12.04 ships Wine64 and the wine binary that ships with CrossOver (which looks like 32-bit Windows even on 64-bit host) is incompatible with winetricks
17:26
<hsivonen>
of course, CrossOver is supposed to make winetricks unnecessary, but CrossOver is broken
18:43
tantek
scrolls up and reads about currency symbols.
18:44
<tantek>
There's definitely quite a variety of ways that folks publish currency amounts on the web.
18:44
<tantek>
The three-letter code as annevk suggests is probably the easiest to both publish and search for.
18:44
<tantek>
But not what every locality is used to seeing - so...
18:46
<tantek>
anwyay - if anyone is interested in either reading more on currency publishing practices on the web or has further real world examples to add, please take a look at: http://microformats.org/wiki/currency-examples
18:48
<tantek>
and currency related formats respectively: http://microformats.org/wiki/currency-formats
18:58
<gavinc>
RDF HTML datatype in terms of DOM4 and HTML5 http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-concepts/index.html#section-html
19:45
<annevk>
gavin: why do you use <code> for what are <var>s?
19:45
<annevk>
euh gavinc
19:45
<annevk>
sorry gavin
19:47
<annevk>
gavinc: you don't need to invoke normalize()
19:48
<annevk>
gavinc: and if you don't pass a context element you'll get <html> and <body> etc. implied and you might get quirks mode parsing even...
19:59
<gavinc>
annevk: isn't normalize needed before using isEqualNode?
20:01
<gavinc>
annevk: What should the context node be in that case?
20:19
<annevk>
gavinc: normalizing already happens during parsing afaict
20:19
<annevk>
gavinc: dunno, kind of depends on what you want I guess
20:20
<gavinc>
annevk: A floating document fragment
20:21
<gavinc>
Something that came from, and be put back in a complete HTML document
20:22
<annevk>
hmm see discussion on public-webapps
20:22
<annevk>
we don't have a generic thing for that which works
20:22
<gavinc>
:(
20:23
<gavinc>
which discussion?
20:24
<gavinc>
all of it is going to take a good deal of reading... somewhere to start?
20:27
<annevk>
i guess you could go with what you have now or wait until contextless parsing is defined