01:26
<heycam>
TabAtkins, yeah that's totally what currentColor is for
01:27
<heycam>
you could have a mixed HTML/SVG document, inherit your text colour through the color property into the SVG fragment, and use currentColor to set the fill of the SVG <text> elements to the same colour
06:46
<benschwarz>
Hixie: !
06:47
<Hixie>
sup
06:47
<benschwarz>
Hey mate
06:47
<benschwarz>
enjoying australian supper
06:47
<benschwarz>
erm
06:47
<benschwarz>
hah. summer.
06:48
<benschwarz>
just had a bike ride, a little fried
06:48
<benschwarz>
(its 33º)
06:49
<Hixie>
it's "winter" here but, you know, it's california, so... :-)
06:49
<benschwarz>
Hixie: what do you think about making localstorage automagically stringify a json object?
06:49
<benschwarz>
as a change to localstorage
06:50
<Hixie>
tried it
06:50
<Hixie>
browsers wouldn't do it
06:50
<Hixie>
gave up
06:50
<Hixie>
and it wouldn't be backwards-incompatible now, so not much we can do
06:51
<benschwarz>
vendors didn't go for it?
06:51
<benschwarz>
how come?
06:52
<Hixie>
i forget the details
06:52
<Hixie>
it was a while back now
06:56
<benschwarz>
Hixie: the appcache / localstorage screencast I recorded last july came out today
06:57
<Hixie>
cool
06:57
<Hixie>
that's quite the latency :-)
07:04
<benschwarz>
well, yeah
07:04
<benschwarz>
we were both very busy through the latter part of the year
07:05
<benschwarz>
but, nice to see it out
07:24
<brighthas>
IE9 nosuper __defineGetter__ ?
07:28
<MikeSmith>
benschwarz: I think the reason why browser vendors didn't want to change it was what Hixie mentioned about it not being backward-compatible
07:28
<MikeSmith>
so it would have required changing their implementations and breaking some sites that we already using local storage
07:30
<MikeSmith>
I think the browser vendors in general are now not real keen on encouraging developers to use local storage at all to begin with
07:30
<MikeSmith>
and would rather everybody just used indexed database api instead
07:31
<MikeSmith>
which kinda sucks for Web developers in that it's a lot more complicated API to use
08:57
<benschwarz>
indexdb is a solution looking for a problem
08:57
<benschwarz>
@ MikeSmith & Hixie
08:57
<benschwarz>
and MikeSmith, good to see you :)
09:11
<benschwarz>
and localStorage is easy to use and understand, useful, well deployed…
09:42
<MikeSmith>
benschwarz: http://html5doctor.com/storing-data-the-simple-html5-way-and-a-few-tricks-you-might-not-have-known/#comment-17291
09:45
<MikeSmith>
this article claims to have a way to handle deal with the race-condition problem - http://sixrevisions.com/html/introduction-web-storage/
10:25
<benschwarz>
I never thought about using an event like that
10:25
<benschwarz>
one thing the chrome team had trouble with, is that each tab runs in its own process too…
10:26
<benschwarz>
well, I guess its a software design technicality, not a single thing they had trouble with
10:26
<Velmont>
Well, I'm using websql right now, -- so i'll have to port it to idb once opera supports that. :-)
10:26
<benschwarz>
its fundamentally a more complex browser for that reason
14:16
<zewt>
the problems with localStorage tend to be overstated, too; they're valid in complex cases, but I've yet to hear of it causing issues in the real world, with the sorts of things most people use it for
14:17
<zewt>
if you're just storing the user's last selected sort column for client-side results sorting, you have to contrive pretty hard to come up with a serious problem it could cause
15:06
<benschwarz>
zewt: we used localstorage for currency.io, again, simple, but it was the perfect tool for the job
15:23
<zewt>
i suppose you could sort-of emulate it with idb+workers
15:24
<zewt>
load the entire structure (some waste here); when you change it, send the change to the (shared) worker, which broadcasts the change to anyone else using it
15:24
<zewt>
not the same since changes would be exposed to other threads asynchronously rather than synchronously, but most uses (and all sane uses, I think) don't care about that
15:25
<zewt>
or just use idb directly, I suppose, but preloading its contents so it can be read synchronously (which is the main thing people like about localStorage)
15:26
<zewt>
(and the main thing people hate about it. heh)
17:18
<annevk>
Ms2ger: where does this new-style for optional parameter values come from?
17:18
<annevk>
guess I should use that elsewhere too
17:22
<Ms2ger>
We wanted to implement that in Mozilla's WebIDL parser, and thought it would be useful for specs too
17:26
<annevk>
is it experimental or part of Web IDL?
17:26
<annevk>
#lazy
17:27
<Ms2ger>
Part of WebIDL
17:28
<annevk>
ta
17:28
<Ms2ger>
It's nice to have the editor on your payroll ;)
17:28
<smaug____>
what new style ?
17:28
<Ms2ger>
optional boolean deep = true
17:28
<smaug____>
ah
17:28
<smaug____>
that sounds good
17:29
<Ms2ger>
Next up, WebIDL dictionaries in XPIDL
17:29
<smaug____>
Ms2ger: why in XPIDL ?
17:30
<Ms2ger>
Dunno, ask emk
17:38
<mhausenblas>
annevk around?
17:38
<mhausenblas>
http://www.caniuse.com/#feat=cors ... meh
17:40
<mhausenblas>
annevk, I get a shit-load of suggestions and requests for http://enable-cors.org/ ... what do you think, is it time for a little FAQ/screencast/video that covers the most pressing issues?
18:51
<annevk>
mh...?