08:41
<hsivonen>
shepazu: we are in the process of removing svg load event from non-<svg> elements. https://bugzilla.mozilla.org/show_bug.cgi?id=552938
08:42
<hsivonen>
when I emailed the WG about the event, I didn't realize it applied to other elements per spec
12:20
<openstandards>
Hi there
12:21
<openstandards>
currently putting together an html 5 marked up site and was wondering about some semantics
12:22
<openstandards>
I've got a menu for a hotel and i'm wondering if an article with nested sections would be the best way to go
15:55
<alise>
So I shouldn't use Ruby html5lib since it's unmaintained, yes?
17:33
<alise>
html5lib using cElementTree is returning None for a perfectly valid html5 doc
17:33
<alise>
what gives?
19:40
<Philip`>
Woah, the STIX fonts got released
21:20
<kamathln>
Hi, I have a suggestion for html5 I would like to make. (I am a total noob here, so plese, go gentle)
21:21
<kamathln>
where do I propose the new idea
21:27
<kamathln>
anybody around?
21:27
<Philip`>
kamathln: The WHATWG mailing list is probably the best place for detailed proposals
21:28
<Philip`>
though you could mention it here first since people might be able to quickly point out common mistakes :-)
21:30
<kamathln>
Philip`: cool! Then here it is: I hope I am not suggesting something already suggested
21:30
<kamathln>
My idea is to have an option attribute to all tags that need to load external resources than the page itself . The name would be "priority"
21:32
<kamathln>
The user agent can, depending on priority of the external resource, decide whether to download the resource, the order , the bandwidth , etc ..
21:33
<kamathln>
Like I may have a small css file which contains essential classes like error class, and those that differentiate the widgets . I would like to give it high priority
21:34
<kamathln>
Another css that pulls in a bunch of whacky, heavy backgrounds and other images that spruce up that page can be given low priority
21:34
<kamathln>
the same can be applied to javascript as well.
21:34
<kamathln>
Philip`: what say?
21:36
<Dashiva>
That doesn't sound all that useful
21:36
<Dashiva>
For stylesheets, at least
21:36
<kamathln>
Dashiva: Okie, may be wrong example .. Think in terms of images ..
21:36
<Dashiva>
If the style changes randomly a while after loading the page, that wouldn't be a good user experience
21:38
<kamathln>
Dashiva: If the connection is that slow, wierd behaviour is gonna happen anyway. So its better we give priorities to essential resources
21:39
<Dashiva>
If the connection isn't slow, priorities don't matter
21:39
<kamathln>
yes. Thats the point
21:40
<kamathln>
it does actually! If I am doing research, and I dont care about the looks of the page as much as content, I could switch off *non-essential* images
21:40
<kamathln>
priorities need not always be in numbers
21:41
<kamathln>
they could be something like "essential" "non-essential"
21:46
<Philip`>
kamathln: Have you seen http://wiki.whatwg.org/wiki/FAQ#Is_there_a_process_for_adding_new_features_to_a_specification.3F ?
21:48
<nielsle>
kamathlin: There are firefox plugins to disable flash advertisements, but I don't think that many website providers will help us avoiding to read advertisements. Maybe there is a conflict of interests
21:48
<kamathln>
My suggestion to whoever wrote that section: Examples give much better voice to the bulletted lists.
21:48
<kamathln>
Philip`: thanks for pointing though, Will try following it.
21:49
<kamathln>
nielsle: My idea is to hardly avoid advertisements. In fact , for my specific case, I use elinks as long as possible. When I really need to look at an image or there is videom I switch to firefox
21:51
<Philip`>
kamathln: Probably the most important thing is to start by identifying problems that are worth solving (I guess it's something like wanting pages to be usable as quickly as possible), rather than starting with a solution (a priority attribute) and then trying to list problems it might solve
21:52
<nielsle>
I think that gmail has a link that allows you to use a simpler html-based version of the site. Maybe such a link is the best way to solve the problem, but of course it would be nice if the webservice could discover automatically from the data provided by the client.
21:52
<kamathln>
Philip`: thanks
21:52
<Philip`>
since e.g. the problem of loading pages faster might be better solved by browser developers spending more time on speculative parsing and HTTP pipelining etc, which would benefit all pages, rather than adding support for a new attribute that would only ever be used by a tiny fraction of pages
22:14
<kamathln>
Philip`: nielsle: thanks a tonne for the feedback and guidance .. its past my bedtime. Cya on Monday