Dog shows and parades are boring, but the Turkey's out of the oven. We heard from my brother in Afghanistan this morning and he's doing well. We're in East Texas but the weather's actually nice. My nephew Julian's first Thanksgiving is today. Jessie and the rest of my family are here.
Happy Thanksgiving everybody.
So I've realized this geeky goal I have to get an article published on A List Apart. I've had this goal for sometime, but I wanted to mention it here to give myself more motivation. Now you know about it, and you can hassle me if I don't get off my ass and make it happen.
I've been mulling over this idea for an article on CSS and how cross-browser scalable font sizing can help with accessibility. Now that I know what to write about, the biggest problem I have is that my idea takes a conflicting opinion to two articles already published on the subject: one on using pixel sizes only and one about making font-size keywords work. The former was written by Jeffrey Zeldman, ALA's editor, and world-renowned godfather of the web. He's a very open-minded guy, and I'm sure he'd like to read it, but somehow I've got to convince Mr. Zeldman to publish an article that flies in the face of his own. Hmm... I must proceed carefully. Any ideas?
I'm not sure what it is about Mountain Dew, but it's my favorite beverage of all time; I actually fiend for it. For a while I thought there might be some addictive ingredient (You know: Coca-Cola was originally made with cocaine), but of course that wouldn't fly with the FDA. Then I remembered that it was my brother's favorite soda as well, though I'm not sure if he liked it as much as I do. Maybe it's genetic... I wonder if my sister likes it.
Anyway, I was recently up to sometimes 5 or 6 cans a day, when I realized that I was just drinking too much damn soda. I've been able to cut it back to about 2 a day, but like I said, "I fiend for it." When I drink it, there's never enough. The can is gone too quickly... like a drug. Is this weird? Am I the only one?
Yesterday I bought some Diet Mountain Dew with the assumption that it's a little bit healthier not to ingest so much sugar. It tastes alright, I guess. It's better than I thought it would be, but it's still got that "fake sugar" flavor so it's definately not as good as the original. I don't fiend for Diet Dew...
While I'm on the subject, did you know "Mountain Dew" means moonshine? It was originally a lemon-lime drink to be mixed with whiskey... I learned this useful tidbit on the Mountain Dew webpage which, by the way, has a long flash intro and looks like they took tips on webpage building from Strong Bad.
I’m contracting at an office building on the opposite side of town from where I live. It takes me about 45 minutes to drive one-way in traffic. It's not the time that I mind, it's the fact that the time is wasted.
When I worked downtown, I could talk, relax, read, or even sleep on the bus ride while letting someone else worry about the traffic. It was a smooth transition into and out of the workday. Now the worst part of my day is the bumper-to-bumper traffic. I seriously think that my days were less stressful because of the difference in commute. I was less tired, yet slept better. I learned more, and worried less.
Most of the time, I had my nose crammed in some technical book to keep up with the latest web technologies, but sometimes I would just watch the people. I miss seeing the headphone rapper, the 8am drunk, and the albino with the telescope on his glasses. Though I was sometimes annoyed with the obnoxious people, I’d gladly trade my daily commute for a bus pass if I could.
Yi May says girls are smarter than boys.
My first nephew, Julian Cyril Buntinx, was born at 4:39 AM this morning to my sister-in-law, Jennifer. Congratulations to Jennifer and her husband, Olivier. By the way, that's pronounced "Oh-liv-ee-ay", not "Ah-liv-ur". Will post a picture soon.
Steve Champeon and I must be thinking about the same things. He just wrote a wonderful article on semantic markup for Webmonkey. Of course, Steve's well planned and very eloquent article puts my recent weblog rant on the same subject to shame...
I really wanted to quote him here, but all the clinchers come from the closing paragraphs. I don't want to ruin it for would-be readers... (pause)... but I can steal his opening jokes without ruining the article:
For extra credit, sneer at those who use the term "tag" when they mean "attribute". Sophisticates may wish to also use the versatile Russian term nekulturny, and act as one might if confronted by a leper. This is a field rife with opportunity for snobbery. Try to be kind, or at least correct. Let's move on.
In my never-ending quest for the perfect search engine, I ran across one called All the Web. Of course, it's too soon to tell, but this one might be the next replacement for Google. All the Web seems to be as fast as Google and returns similarly relevant results, but it also uses compliant XHTML and CSS. This makes it fully customizable, highly accessible, and the pages render faster.
It doesn't have Google News, but then, that's totally separate from the search engine. Speaking of Google News, is there any talk of syndication via RSS? I use a little headline viewer app called SlashDock on my Mac instead of going to every news site. A Google News feed would be useful.
The latest article on A List Apart has some great news for Flash authors. Everyone previously thought that you could not get Flash to work in Netscape 4 with validating XHTML. As the author discovered, Netscape can use the <object</object> tag instead of <embed></embed> if you change a few attributes.
PHP has a great little feature called variable variables. So you can call a variable whose name is the value of another variable. Here's how it works. All PHP variables start with a dollar sign ($). Here's some sample code:
$foo = 'bar';
$bar = 'one';
Now to reference a variable variable, you add another dollar sign to the beginning. $$foo is currently equivalent to $bar and will return the string 'one'. Think of it as $($foo).
Now the dilemma. What's the best way to do that in client-side JavaScript? I could use eval() to evaluate the expression as a line of code:
eval('var ' + foo + ' = "one"');
But eval() is really slow because the interpreter has to run through each line twice: once to evaluate it, and a second time to execute it. I think the best way would be to set up an object as an associative array, but then you have to predefine all those variables as properties of some generic parent object. That doesn't seem any more efficient. Does anyone have a better idea? Has anyone tried this before?
Before I start on another long boring post about web coding, I wanted to warn the non-coders that this is the time when your eyes should glaze over.
You may have heard a lot about the benefits of Cascading Style Sheets (CSS) and the move away from tables as a layout element for webpages. Using elements like <div></div> in combination with CSS-P can greatly increase the accessibility, portability, and maintainability of data. However, it's main benefit is achieved when the XHTML is used properly... Semantically.
What that means in relation to XHTML is that HTML was intended as a way to transfer data in it's own "self-describing" format. The tag <h1></h1> signifies that the contained text should be treated as the first-level heading of a document. The tag <p></p> should be used as a way to mark up paragraph text. The tag <table></table> should refer to a table of related data, not a visual page layout. Some tags, like <font></font>, <i></i>, and <b></b>, refer to visual styles but have no semantic meaning. These are some of the elements that CSS was meant to replace and because of that, they have been deprecated (removed) from XHTML. Other elements, that have semantic meanings, should be used in their intended context.
Tables are not deprecated... While CSS-P is meant to replace incorrectly used layout tables, it's not meant to replace tables altogether. I've seen more than a few sites recently with developers who are so excited to remove tables from their layout, they miss the point and remove data tables that should remain. I've seen people do so much work finding CSS margin and alignment settings to emulate tables when they should have just left the table in. For example, a staff contact info list is a great use for a table. A list of website articles with their titles, dates, or other fields of data is another great use for a table. Any relational data that needs to be arranged in rows or columns should be marked up in a <table></table> tag.
The point of XML and XHTML is that the data should describe itself. Pretend you are going to markup a web page without caring how it looks; only think about the tags and what they mean. After you figure out what tags you need, then you can go back and style the CSS however you want to. For example, look at this page without styles. Is the page readable? Do the headings read as headings? Do the paragraphs read as paragraphs? Does the link list read as a hierarchal list? (The answer to that last one is currently 'No', but I'm going to change that soon.)
Common HTML use reached a peak in display markup because there was no other way to get the visual look designers wanted. Now that most browsers support web standards such as CSS and XHTML, the markup use should return to it's orginal purpose. Not only should the style be separate from the data, the data should be "self-describing". Use the language how it was intended.
The Chimera version 0.6 web browser came out today. It's available for Mac OS X and stands to be one of the best browsers out there. It's built for OSX and uses a beautiful aqua interface.
The only noticeable bug I've seen relates to CSS fixed positioning. Fixed positioned elements leave artifacts behind. The reason I know about this bug is because it affects my site, too.
Today is my father's birthday. That's him holding the pop gun back in 1958. Happy birthday, Dad...
The photo was sent by my Aunt Margaret. That's her standing next to him with a sly look towards the camera.
When Microsoft Office XP came out, there was a big stir in the web development community about a new feature that Microsoft dubbed "smart tags". Users of Office XP can get added "links" to relative information based on keywords. For example, if a user views an email with a street address like "123 Fake Street", the user is presented with a link to a map. While this seems relatively harmless and even useful, the problem is that Microsoft determines all the content of the links. The map location is a Microsoft affiliate site and all other links are controlled by Microsoft.
The problem gets worse when it applies to web pages. A bank site we maintain at work has information about mortgages they offer. The bank owner called Will once when he saw the website on his son's computer. The son had Microsoft Office XP and Internet Explorer 6 installed and the "smart tags" were parsing the web page for links. The mortgage information on his site was linking to another bank's website, a competitor, and a Microsoft affiliate. This is just wrong...
There is a way to keep the M$ vandalism off your webpages. Simply add this meta tag in the head of your HTML. It sucks that you have to add weight to the page to remove a feature that shouldn't be there in the first place, but at least it's not much weight.
<meta name="MSSmartTagsPreventParsing" content="TRUE" />
This problem (and fix) have been around for at least a year, but it recently came to my attention again when I saw it in action. I thought I should re-emphasize it for everyone.
I just got back from the Amon Tobin show with DJ Food, Bonobo, and P-Love and I must say it was well worth it. Strictly Kev, the main man behind DJ Food, stole the show in my opinion, but all the sets were great. Kev played what had to have been my favorite live DJ set. It was so diverse; he played all kinds of music and most impressed me with his ability to drop out half beat from hardcore drum-n-bass into a Fine Young Cannibals track. The only thing that would have topped it, is if he had thrown in some old school electro like Whodini or Soul Sonic Force. The interim DJ, P-Love, was an added bonus. He's a turntablist and went back-to-back beat juggling on Sade and even a Justin Timberlake tune; all thrown together in an underrated hip-hop style.
Amon Tobin played for at least an hour and a half. His set was also diverse, but I wasn't as impressed with the first half as I planned to be. It was mainly a jungle DJ set, but the second half pulled in more of his own unique stuff. Jazzy, crazy drum beats and downtempo funkiness is what I came to hear and Amon Tobin didn't dissappoint, though he took a while to get there. All in all, the show was well worth coming out of hibernation for.
Dane found a page that mentioned part of the T-Mobile site I worked on with frog. Bill Kim wrote most of the JavaScript and I wrote most of the CSS for T-Mobile. It's kind of ironic that the article talks about this, but I'm about to update cookiecrook to use lists for my navigation menu. While that's of no concern to anyone who's not a web developer, CSS-styled lists are the way to go when it comes to navigation menus. To you other developers out there, get on the bandwagon...
XHTML 1.1 | CSS 3.0 | RSS 0.91 | WCAG 1 AA+ | | Blogger
