afongen

WebAIM is doing another online web accessibility training this year. Last year’s was good (and free) enough that I bought the training materials on CD-ROM. This year’s is not free, but I like some of the things they’re doing, namely breaking it out into tracks. And there’s a handy education discount. Looks worth it.

In case you haven’t tried it, I suggest that you try the updated version of WAVE, an online accessibility tool that is now developed by WebAIM. It’s in beta but is still a far sight better than most other tools.

A lot of people like Bobby. I don’t. The problems that Bobby can check automatically are almost all things that a good markup validator checks for: missing alt text, DOCTYPE declarations, and so on. Most pages turn up a fair number of user checks that need to be done, which is fine except for two things: messages from Bobby are cryptic and unhelpful, and it’s not clear how to evaluate some of what Bobby flags as possible problems (especially to novices). The W3C‘s markup validator has much more helpful messages, even to those who don’t know HTML well.

So how do you do the user checks? That’s where WAVE comes in. Plug in a URL, and WAVE returns tha page at that URL marked up with icons that you can use to evaluate the accessibility of that page. For example, rather than simply flagging missing alt text, WAVE goes a step further and displays the alt text so you can determine whether it’s actually useful. It identifies structural markup (headers, lists, etc.) to help you evaluate whether the page is marked up properly. WAVE calls your attention to mouse-activated JavaScript events, so you can decide whether they should also be keyboard-triggered.

When evaluating and improving the accessibility of a web page, I do at least these two things:

1. Clean up the markup, with the aid of a validator or HTML Tidy. This alone addresses many/most accessibility problems.
2. Use WAVE to help identify anything else that needs cleaning up.

There’s always room for improvement, but I find that’s usually enough — or at least is all that automated tools can help with. Other accessibility issues tend to be global in nature, relating to navigation, content, and so on, and require more sophisticated analysis.

Tempting though it’s been, I have not gone out and bought Apple’s Keynote presentation software. It looks slick, easier to use than PowerPoint, and I like how the presentations look. I’ve held out, though, because

1. I can think of better ways to spend a hundred bucks.
2. I don’t make that many presentations.
3. As much as I love my iBook, the few presentations that I do make need to be made available to people who don’t use Macs. Yes, I could export and import PowerPoint presentations or PDFs, but I’d have to see the interoperability at work before I shelled out.
4. I’m holding out for a decent OpenOffice port. Alright, the most recent release is pretty good, but still not comfortably workable on my iBook.

Yesterday Apple released Keynote’s XML schema, for those of us who might want to access or create Keynote presentations programatically. This makes Keynote a teensy bit more tempting. The schema is certainly easier to grok than OpenOffice’s XML file format, although that’s not entirely fair since OpenOffice is a suite of applications while Keynote is a stand-alone app. Too, there may be more to come: Apple mentions a “four-part Technical Note ‘Deconstructing a Keynote Document'” that is apparently a work in progress.

So am I planning to buy Keynote? Not unless I start doing enough presentations to merit the price. I don’t think a hundred dollars is extravagant, but still more than I should be tossing around right now. I do have to say that I applaud Apple for making this move. They have been doing such a good job of working with open standards, balancing openness with proprietary needs.

Oh yes. If you’re using Safari, don’t miss the new beta.

Interesting: Googlert

Googlert is an experimental free service that keeps you updated on what the web is saying about you or your interests. It does this by performing regular Google searches on your behalf and sending you email alerts of any new results that appear.

My first question was how they get around the restrictions on the Google Web APIs, mainly the limit of 1000 queries per day. Duh: Googlert uses your Google license key for your searches.

On the back of a snowplow I saw a sign today: “Stay Back. Stay Alive.” I suppose that’s better than “Stay Back or Die” which is what I would most likely have written. I don’t imagine that would sit well with people.

I decided a couple months ago that it was time to renew my acquaintance with either Python or Java, but vacillated between the two for some time. In the end the decision was simple: although Python has some sex appeal, Java makes more sense. It’s much more in keeping with the direction that I want to go as a web developer. Whereas there’s very little that I want to do with Python that I can’t or don’t already do with Perl, there are exciting things happening in the Java world that I’d rather not miss, including some of the most appealing Apache projects. I have a strong suspicion that my employer will be doing quite a bit more with Java in the not-too-distant future, which means that I might be, too — assuming I don’t get laid off in the upcoming budget cuts, god forbid. And I really did miss working with the language. It’s been fun diving back into things.

Then I stumbled into How to Think Like a Computer Scientist: Learning with Python. For a sec I thought this would muddy the waters. But wait! There’s a Java version. And of course there’s always Bruce Eckel’s Thinking in Java.

Thing is, though, the only one of those three that I’m likely to read is the Python one. Go figure. Maybe next year.

This evening I took care of the kid while Kiara was in class. Nothing unusual about that, except we stayed in town instead of going home. We ended up at the TeaSource, where he fussed a bit but eventually settled down so we could read a couple of his favorite books. A couple guys at the next table had a copy of a children’s book that they’ve written and illustrated, due to be published in early March. I mention it here so i don’t lose this information: If I Had Thumbs Like People Do. ISBN 1-4010-8261-0. Looks like a fun book, keep an eye out.

Update: You can buy the book here.

Exciting news: Matt Sergeant has released the code for the AxKit OpenOffice Provider, which makes delivering OpenOffice SXW files to the web a snap.

You may already be familiar with AxKit, an XML application server for Apache written in Perl. The OpenOffice filter transforms OpenOffice files to the output format of your choice…say, XHTML. I’ve been eyeing it for some time now, but it was a commercial product so not easily available to me. (If we’d decided to use AxKit at work, it may have been an option, but we’re still quite a ways away from making that sort of decision.) Since Matt’s dissolved the company that sold the OpenOffice filter, he’s open sourced the code. I cannot wait to dig into this.

I’m curious, too, how I could incorporate it into something like Callisto CMS, a content management system built using AxKit. I’d also like to know out how easily it could be ported for use in Cocoon. I prefer to work with AxKit but have to be open to other options.

The Space Shuttle Columbia appears to have broken up on re-entry. Oh no. And I just now heard that the flag at the Kennedy Space Center has been lowered to half-staff. I guess that’s unofficial confirmation. NASA should hold a press conference in half an hour or so.

I’m up at Lake Superior College for the next couple days, doing web accessibility training for their faculty, so will be incommunicado.

I’m looking forward to it, should be fun. (Especially after the bureaucratic hell it took to arrange my involvement. But that’s a story for another day.)

I’ve switched the DOCTYPE of these pages to XHTML 1.0 Transitional, even though I will continue to “code” to Strict. Why? Because Dean Burge makes sense: the XHTML media types summary from the W3C suggests that XHTML 1.0 Strict should not be sent as text/html. Until user agents can consistently support application/xhtml+xml (yes, I’m looking at you, Internet Explorer), I’m treating the markup as Transitional and sticking with text/html.

I thought about serving up Strict to user agents that claim support for application/xhtml+xml. Paul went that route a while back, before he went off the deep end and moved to HTML (albeit for sound reasons). It just doesn’t seem worth the trouble, though, especially since I’m not doing anything very XML-like with the XHTML at this point. Until I do, I’m not entirely comfortable calling it Strict. Although one could claim that technically it is, it’s the spirit of the matter that sways me.