00:07
<TabAtkins>
Either can be used as a child of the other, as appropriate. In many instances, though, <section> will be a child of <article>.
00:08
<TabAtkins>
Because in many cases <article> will be used to denote the main content of a page.
00:09
<TabAtkins>
But you can, frex, have something like <article #pagecontent><section .pagesection><article .blogpost><section .sectionofblogpost>...
00:10
<JonathanNeal>
Taking from w3.org (if that's allowed here), they write "The section element is not a generic container element. When an element is needed for styling purposes or as a convenience for scripting, authors are encouraged to use the div element instead." This is important for me to understand, because I may be upgrading the html of portlets within a portal, so I'm wondering if the portlet as an application should use div'
00:11
<TabAtkins>
Smashing's use is perfectly appropriate, though, if they're only getting down to the actual *article* after having some extra data before/after it.
00:11
<TabAtkins>
Probably should.
00:11
<JonathanNeal>
Right now I have <div class="portlet"><div class="portlet-topper"><div class="portlet-title">Blog Application</div><div class="portlet-content"><!-- blog application --!></div></div>
00:12
<webben_>
JonathanNeal: I think looking at the editor's draft portlet should probably use section not div.
00:12
<TabAtkins>
Rule is: if something should appear in an outline of the page, use <section>/<article>. If it shouldn't, use <div>
00:12
<webben_>
JonathanNeal: "The section element represents a generic document or application section. A section, in this context, is a thematic grouping of content, typically with a heading, possibly with a footer. Examples of sections would be chapters, the various tabbed pages in a tabbed dialog box, or the numbered sections of a thesis. A Web site's home page could be split into sections for an introduction, news items, contact information."
00:12
<webben_>
http://dev.w3.org/html5/spec/semantics.html#the-section-element
00:12
<TabAtkins>
I'll note that your use of a <div> for the *title* is worrying already.
00:13
<webben_>
JonathanNeal: So a portal site is split into sections, some of those sections are applications.
00:13
<JonathanNeal>
Sure, that's how it's been, because portlet's can optionally remove their topper, title,
00:14
<JonathanNeal>
For instance, the web content display portlet is often configured to skip right into the portlet's content, which is why I was wondering if the html should remain div based, because it has more to do with the structure of the page, but I can also see how it's thematically grouping content.
00:14
<TabAtkins>
Just use the "outline" rule. You don't have to quibble about semantics then. If you collapse the whole page into an outline, and your bit should show up, use the <section>.
00:15
<TabAtkins>
So, it's probably the case that the outermost div should be a <section>. The others are probably good as <div>s.
00:16
<webben_>
JonathanNeal: I guess I'd suggest <section class="portlet"><header><h1>Blog Application</h1></header><!-- content --></section>
00:16
<JonathanNeal>
Would you then say that <div class="portlet-topper"> could be <header> and <div class="portlet-title"> could be <h1>. From what I've read, sections are recommended to begin new header deliniating.
00:17
<JonathanNeal>
Sure, I need to wrap that content so there you would recommend using a DIV? I really appreciate your input on all of this, by the way, all of you.
00:17
<TabAtkins>
I agree with webben_
00:18
<webben_>
JonathanNeal: Assuming you need it to be the same for all portlets, can't think of anything else than div off-hand.
00:18
<TabAtkins>
And yeah, probably just wrap the content with a div. Again, just mentally apply the outline rule.
00:19
<JonathanNeal>
All right, let me make a pastie of what I've gathered so far from what you've said.
00:19
<TabAtkins>
But if you can determine that some portlets *are* important enough to do so, consider an <article> or <section> to wrap it instead.
00:23
<JonathanNeal>
http://pastie.org/585342
00:23
<JonathanNeal>
The first one labeled "HTML4" is actually what we've been using for some time, I never liked the spans being there alongside the divs.
00:24
<TabAtkins>
Yes, except you can probably drop most of the classes.
00:25
<TabAtkins>
Keep the .portlet class, but the others can be targetted just fine with child selectors. ^_^
00:25
<TabAtkins>
".portlet > header", ".portlet > header > h1", etc.
00:26
<webben_>
JonathanNeal: I suspect that may be a menu not a nav: http://dev.w3.org/html5/spec/Overview.html#the-nav-element vs. http://dev.w3.org/html5/spec/Overview.html#menus ?
00:26
<JonathanNeal>
Well, we walk the fine corporate line of trying to move towards html5 but keep ie6 compatibility.
00:26
<webben_>
JonathanNeal: depending on what the contents are.
00:26
<JonathanNeal>
Right now, most of that can be achieved through CSS, and JS using document.createElement function.
00:26
<JonathanNeal>
Child selectors aren't available in ie6, unfortunately.
00:27
<TabAtkins>
Bah, right. I am lucky enough to have been allowed to basically throw IE6 over the bridge at my job.
00:27
<webben_>
JonathanNeal: How are you selling migration to 5 to the biz?
00:27
<JonathanNeal>
webben_ by the grace of God.
00:27
<webben_>
heh
00:27
<JonathanNeal>
I have no idea how, but they like it.
00:28
<TabAtkins>
It looks fancy!
00:28
<JonathanNeal>
They're letting me style our new website using CSS3 almost exclusively, as long as the unCSS3'd ie browsers look good and consistant.
00:28
<JonathanNeal>
They don't mind letting the new browsers take on the added improvements.
00:28
<TabAtkins>
Woo, lucky you.
00:29
<webben_>
JonathanNeal: e.g. rounded corners with CSS3 and square corners for the old?
00:29
<JonathanNeal>
Exactly.
00:29
<TabAtkins>
I'm happy I got them to accept the rounded corners issue. That's about it.
00:29
<webben_>
yeah, rounded corners is about as far as we've got where I work.
00:29
<JonathanNeal>
Well, ie6+ supports gradients, so IE actually gets the gradients too.
00:29
<JonathanNeal>
Firefox is the tricky one, I have to calculate the colors for the gradients by using the inset shadow property.
00:30
<TabAtkins>
Hmm? How are you getting gradienets in ie6?
00:30
<webben_>
http://msdn.microsoft.com/en-us/library/ms532997(VS.85).aspx ?
00:30
<JonathanNeal>
We have browser selectors out of the box and ie supports gradients using the filter property.
00:30
<TabAtkins>
Ah, k. I stay away from the filter properties.
00:31
<JonathanNeal>
Me too, but the end effect let's me get away with a lot without introducing images to the design where they weren't necessary.
00:32
<TabAtkins>
So far I've dealt with it by using a gradient-generating image script in PHP.
00:32
<JonathanNeal>
The really, really great thing about this is, if I can convince them to move forward with some of the html5 adoptions in our portal, then our clients will take those html5 adoptions with them too.
00:34
<JonathanNeal>
Which is why using proper standards is so absolutely important to me.
00:34
<TabAtkins>
Cool. ^_^
00:35
<webben_>
JonathanNeal: "inset shadow property": like this? http://markusstange.wordpress.com/ ?
00:36
<JonathanNeal>
It seems pretty straightforward, so http://pastie.org/585351
00:37
<JonathanNeal>
I think I'm in good graces because I wrote this ( http://tiny.cc/JTlyy and http://tiny.cc/KyH4B ) a day after someone asked if we could support mobile
00:37
<webben_>
JonathanNeal: class="portlet seamless" would make more sense to me. ... it's still a portlet, even if it's borderless
00:38
<webben_>
or class="portlet portlet-seamless" if .portlet.seamless is going to cause a world of ie6-inflected pain.
00:38
<JonathanNeal>
ie6 has limited support of css chaining but I'm aware of how most of it needs to work.
00:38
<JonathanNeal>
yea, adding the portlet class to them both makes sense, webben_
00:39
<webben_>
if it is seamless, tho, why do you need a div inside the seamless section?
00:39
<TabAtkins>
Ooh, shiny: http://tests.themasta.com/blogstuff/boxshadowselect.html
00:39
<webben_>
TabAtkins: Is that shiny for Fx?
00:40
<TabAtkins>
Yes.
00:40
<TabAtkins>
It's all -moz-box-shadow
00:40
<webben_>
very nice
00:40
<JonathanNeal>
Our shiny is shiny in ie6+, firefox, and safari
00:40
<JonathanNeal>
which includes any other trident / gecko / webkit browsers, obviously.
00:40
<TabAtkins>
Sure, just put -webkit-box-shadow, and some IE filters on it.
00:40
<TabAtkins>
Definitely reproducible.
00:41
<TabAtkins>
Note: I didn't make that. It's from the blog that webben_ linked to.
00:41
<webben_>
I guess one could use SVG data URIs for extra-http-request-less gradients in opera
00:42
<TabAtkins>
Man, that's *two* technologies I don't know how to use.
00:43
<JonathanNeal>
We used -moz-box-shadow as the secondary solution to -webkit-gradient and Microsoft.Gradient
00:43
<JonathanNeal>
For example @ http://pastie.org/585353
00:46
<TabAtkins>
Ah, cool, php can help write data uris for me.
00:47
<JonathanNeal>
So now I have http://pastie.org/585355
00:47
<TabAtkins>
I approve.
00:48
<webben_>
Tick! ;)
00:49
<webben_>
JonathanNeal: altho note my note about menu
00:49
<JonathanNeal>
Okay, the next big beast to tackle of the portal's page container.
00:50
<JonathanNeal>
I thought menu was gone in HTML5.
00:50
<webben_>
still here afaik
00:50
<JonathanNeal>
So, when would it be proper to use nav versus menu and vice versa?
00:51
<webben_>
nav = site links ; menu = commands
00:52
<webben_>
http://dev.w3.org/html5/spec/semantics.html#the-nav-element vs http://dev.w3.org/html5/spec/interactive-elements.html#menus
00:52
<webben_>
"control panel" sounds to me like commands/toolbar
00:52
<webben_>
so menu
00:53
<TabAtkins>
Hmm, I didn't realize that <menu> was meant to be used for in-page markup. Cool.
00:53
<webben_>
other lists of links ... ul or ol I guess.
00:53
<TabAtkins>
btw: http://www.whatwg.org/specs/web-apps/current-work/multipage/#menus for the multipage version
00:54
<JonathanNeal>
I see what you mean.
00:54
<TabAtkins>
I need to really read and digest that section soon.
00:55
<TabAtkins>
For now: grocery store, to prepare for tomorrow's grilling party.
00:56
<JonathanNeal>
Why seamless versus borderless-container
00:56
<JonathanNeal>
because of the meaning?
00:58
<webben_>
i guess various reasons
00:58
<webben_>
shorter, more abstract
00:59
<webben_>
"container" seems redundant
00:59
<webben_>
it's a section so... it's going to be a container
01:00
<webben_>
JonathanNeal: also http://dev.w3.org/html5/spec/embedded-content-0.html#attr-iframe-seamless
01:04
<JonathanNeal>
I see.
01:05
<JonathanNeal>
Great, I'm writing all of this up.
01:05
<JonathanNeal>
Thanks webben_
01:06
<webben_>
JonathanNeal: yw :)
01:09
<JonathanNeal>
All right, here's the tougher one.
01:10
<JonathanNeal>
Any number of portlets can exist in here @ http://pastie.org/585373
01:12
<JonathanNeal>
I'm curious if a section would replace wrapper, a header would replace banner, an h2 would replace current-community and be moved outside of the h1, both the h1 and the h2 would be wrapped with an hgroup, and a footer would replace footer.
01:16
webben_
is too sleepy to think, but I'm sure one of the other #whatwg will answer eventually (if not just mail the help@ mailing list: http://www.whatwg.org/mailing-list#help )
01:20
<JonathanNeal>
Should a <nav> ever go inside a <nav> for child pages?
01:39
<JonathanNeal>
I'm getting all crazy with this one.
01:39
<JonathanNeal>
You guys will have to tell me what you think. It's perhaps incorrect.
01:41
<JonathanNeal>
Here's my latest working copy @ http://pastie.org/585387
02:19
<JonathanNeal>
Anyone around?
02:26
<jcranmer>
no
02:29
<JonathanNeal>
I figured as much :-)
05:34
<JonathanNeal>
hi all
09:37
<annevk42>
aboodman2, aboodman_mac_, answered your question
09:37
<aboodman_mac>
annevk42: thanks for replying
09:37
<aboodman_mac>
looking now
09:41
<aboodman_mac>
annevk42: ok, thanks. I am working on this area of webkit now, and want to be sure I don't break anything.
09:47
Philip`
wonders how much cardboard he needs to stick underneath his laptop's system board to bend it enough that it will successfully boot
09:50
<hsivonen>
hmm. http://www.fsf.org/resources/formats/playogg tells people to install VLC which expands the installed base of encumbered codecs, too
09:50
<hsivonen>
they could tell people to install Firefox, which wouldn't expand the installed base for encumbered-format players
11:29
<annevk42>
note to self: make a simple python script instead to generate encoding tests rather than cp 001.htm 00x.htm all the time
11:49
<krijnh>
http://code.google.com/p/sessionstorage/
12:18
<Philip`>
Hmm, ten thicknesses of card under the PCMCIA slot seems to be sufficient
12:28
<Dashiva>
You probably even enjoy making it work
12:34
<Philip`>
I only had to take out 30 screws to reach it
12:34
<Philip`>
and I put 29 screws back in, which I consider a success
12:36
<Hixie>
wow, i've got less than 3 days backlog on my main folder of input
12:37
<Dashiva>
If you run out, you might have to start answering questions when asked.
12:37
<Hixie>
lots of XXX markers to worry about still
12:37
<Hixie>
and bugs
12:38
<Hixie>
(i don't like replying in realtime, i much prefer letting threads settle, since that way i can just look over the whole thread and use the best ideas.)
12:38
<Hixie>
anywya it's way past my bed time
12:38
<Hixie>
nn
15:08
<hsivonen>
wow. the StuffIt Expander download has become like downloading Real Player
15:08
<hsivonen>
and worse
15:08
<hsivonen>
very tacky
15:10
hsivonen
archives a old PPC 2005 version of StuffIt Expander and refrains from downloading the new version
15:15
gsnedders
has an old 68k version somewhere
15:33
<hsivonen>
I got tired of .zip files getting assigned to StuffIt Expander, which launches Rosetta
15:57
<Lachy>
hsivonen, couldn't you just change the default application for .zip files, without having to remove StuffIt Expander?
15:58
<Lachy>
Select any zip file, change the default app and click Change All...
15:58
<hsivonen>
Lachy: apparently not
15:59
<Lachy>
(I know it's made slightly complicated by the stupidity of Macs sometimes relying on invisible metadata for choosing default apps)
15:59
<hsivonen>
Lachy: I've gone through the usual steps in both Finder and Firefox
16:00
<hsivonen>
Lachy: but there's something somewhere that makes Firefox write the HFS creator code for StuffIt expander onto zip files
16:00
<hsivonen>
or so I suspect
16:01
<hsivonen>
could be some very, very legacy setting migrated a dozen times somewhere in the bowels of Carbon in the parts that came from Internet Config
16:01
<hsivonen>
or something like that
16:01
<Lachy>
yeah, that's what it's called. The way the creator code is used has to be one of the most annoying design flaws I dislike about macs
20:17
hsivonen
wonders what the messages sent, replies gotten stats divided by time on the list would look like now