cookiecrook²

2:04 PM #

AIR 2004 Course Notes Posted

I just got back from AIR University, a three-day accessibility conference organized by Knowbility. It was a huge success. About 95% of the feedback I heard was positive. Many of the comments were ecstatic raves.

The notes are online for the two courses I prepared… The JavaScript course (3 sessions) followed last year’s advanced training session, but covered the scripts in more detail. Jim Allan and I also taught a CSS course (4 sessions) that was a step-by-step build through a one-page example CSS-P layout. All sessions of both courses were full.

I was able to attend a few more courses: Home Page Reader taught by Phill Jenkins of IBM, Macromedia Accessibility taught by Bob Regan of Macromedia, and Making PDFs Accessible taught by Greg Pisocky of Adobe. Everyone learned a lot and I enjoyed every minute of it. Thanks to everyone involved.

More AIR University resources can be found on the Knowbility resources page.

12:01 AM #

Moving hosts

My wonderful host, ByteStacker, is unexpectedly going out-of-business, and I’m looking for a new host. I’m currently running a Virtual Private Server (VPS) on Debian GNU/Linux, and would like to find another. Most of the VPS packages I’ve found run RedHat instead. Any recommendations? Suggestions? Thanks in advance.

10:00 PM #

Brandishing the web standards cudgel

In his latest post, When semantic coders go bad, Matthew Thomas has responded to complaints raised by Matt May, Lachlan Cannon, and me. There are a couple of points, in order of his mention, that I’d like to respond to:

1. Use of b and i on my external commenting system.

   Matthew seems determined to ridicule my external commenting system (in my comments and in his response post) even though I explained the external code was not editable. I set up those comments using BlogBack a few years ago and haven’t had a chance to update since then. I would have explained this on Matthew’s site if he had a comment system.

2. My comment about deprecation of b and i.

   Matthew called me out on this one and I was wrong. Most of my posts are stream-of-consciousness, and sometimes, in the heat of a post, I get over-zealous and forget to check my facts. After all, this is a personal blog, not The New England Journal of Medicine. More important is the meaning: b and i are presentational elements and presentational elements have no place in semantic markup.

So what exactly doesn’t work?

Mr. Thomas responded to the suggestion of using span instead of i:

These two options have exactly the same semantics. The only difference is that Mr. May’s version doesn’t work in Internet Explorer for Windows. Or in Internet Explorer for Mac. Or in Opera in User Mode. Or in Mozilla with Basic Page Style selected. Or in Firefox with Basic Theme selected. Or in Lynx.

Matthew confuses the difference between browsers supported for use and browsers targeted for full support of the stylistic display. This concept is aptly explained in Jeffrey Zeldman’s article, Why Don’t You Code for Netscape? All browsers can read the text information; only CSS2-capable browsers will display the text in italics. Is it really necessary to display a French phrase in italics? No. It's a stylistic choice better left to CSS. All users get the content. So how is that broken?

I’m a graphic designer. Sure, I want my colors to display correctly and I want my italic text to display in italics... in a modern GUI browser, that is. Let the text browsers display it as plain text.

Doesn’t work in Internet Explorer? Have you seen Dean Edward’s IE7? Even Internet Explorer can use attribute selectors now. Even if it couldn’t, there is still the option of adding a class.

So what exactly is irrelevant?

Mr. Thomas also said:

Matt May’s complaint that this is not the right way for a semantic HTML coder to look at things is entirely correct, but almost entirely irrelevant. For HTML to be the lingua franca for publishing hypertext on the World Wide Web, it must also cater for the ~99 percent of Web authors who don’t care about semantics and never will.

Irrelevant? Perhaps Mr. Thomas would prefer to let Microsoft and Google set the standards... Even obscurely-semantic markup and metadata with minimal support can still be used for meaningful results. Example cases include XFN relationships and GeoURL’s location relationships.

Markup is hard.

You’re right Matthew. Markup is hard... Making it easy to incorrectly use semantic markup is a bad thing—that may be why I’ve never found a rich-text editor CMS I like—but incorporating presentational elements into semantic markup is worse.

11:35 AM #

AIR-U classes are filling up

Due to the overwhelming response for the CSS course, we’ve decided to add one more opening to the Monday afternoon schedule. If you weren’t able to enroll in the CSS morning classes, this is your last chance.

I also just learned that Bob Regan will be teaching the Macromedia Accessibility course for two of the three days. Bob is the Senior Project Manager of Accessibility at Macromedia and is credited with getting Flash accessibility to its current. I assume this course will mainly be about Flash, but may touch on the new Dreamweaver, too.

2:22 PM #

BestKungFu’s semantic kung fu is the best

Matt May has responded to Matthew Thomas’ article, When Semantic Markup Goes Bad, and Mr. May is absolutely correct: b and i should not exist. They should not be used. They have no value. Because I couldn’t post a comment on Matt or Matthew’s site, I decided to respond here.

While Mr. Thomas is correct that a one-for-one switch from b and i to strong and em is wrong, he neglects to mention the full reason for the switch: separation of content from all presentation. Presentational markup like b and i have been deprecated... They should no longer be used because they only indicate presentation, not semantic meaning.

Mr. Thomas’ article goes on to provide examples of when to use semantic markup. His article falls short, however, when the examples have no directly-equivalent semantic element. In these cases, Mr. Thomas recommends falling back on presentational elements. This is the semantic equivalent of overusing the style attribute: it packs the data full of junk presentational markup.

In the vector example, R² is a variable. It should use the var element. Even if the content was meaningless, it should not use presentational markup to style itself. The reason for bold text would then be presentational only. Markup is for data and metadata. Style sheets are for styles and presentation. As Mr. May’s example shows, when your content can use no directly-equivalent semantic element, you should use a meaningless, non-presentational element (like span) or the closest semantic equivalent instead of any presentational element. Perhaps this should be the subject of discussion in the next Simple[Semantic]Quiz.

Don’t fall back on b and i! Mark up your data by its implied meaning. Worry about style later.

4:46 PM #

Austin Web Standards Meetup tomorrow

Join us at Mother Egan’s Irish Pub tomorrow, Thursday, May 6th, at 8pm for the Austin Web Standards Meetup. Drinkin’ stout and geekin’ out. Eh heh. There will be five of us coming and apparently we’re the second largest Web Standards Meetup group in the world. Imagine that. ;)

After the brew-haha, some of us will head over to the Red Eyed Fly to check out Go Nova, Just Guns, Second Story Thief, and King Coal.

2:21 PM #

Register for AIR-University 2004

Knowbility presents AIR-University 2004, a three day conference of accessibility and web standards workshops. Join us Monday, May 24th, through Wednesday, May 26th, for classes on CSS, JavaScript, Flash, PDF, video captions, descriptive audio, screen readers, accessibility testing, and more.

If you work with a Texas governmental or educational organization, you need to attend AIR-U. Registration is cheap but register early because it will fill up. Don’t let the inexpensive price fool you; it’s cheap because the instructors are donating their time.