Tim Gallant Creative Blog
Thoughts on graphic design, web development, and related technology. And maybe other nonsense too….

Euro Forced “Browser Choice” Changes Nothing (So Far)

April 1st, 2010

http://www.sitepoint.com/blogs/2010/04/02/browser-voting-results-month-1/

These results do not surprise me. While a lot of people involved with web standards have been on an anti-Microsoft crusade for a long time, and politicians and courts in Europe have their own agenda against big business, the truth is that for the average person, the whole kerfuffle is a non-issue.

I personally am pretty conscientious about developing with web standards. Not because the elite told me I should be, but for a number of very practical reasons:

  • Coding well according to web standards both makes things simpler now and easier to alter or troubleshoot later.
  • Deploying web standards means better accessibility for a significant handful of people.
  • Given the steady trend to better web standards support across the board, it just makes sense as far as future-proofing.

But the truth is that, while people may justly complain about IE’s handling of the DOM, for most uses, IE8 deals just fine with good code, and I rarely need to throw more than a couple lines of code to IE6 in a separate stylesheet (by way of a conditional comment) to get it to behave.

As time goes on, if Microsoft continues to lag behind in CSS support (which in fact doesn’t look all that likely, if the prerelease versions of IE9 are any indication – they’re ahead of everyone else’s curve), one can code in a way to take advantage of more cutting-edge code while allowing more archaic browsers to degrade gracefully.

While that is more work, that is already virtually inevitable anyway. Why?

We web developers are geeks, and our line of work means that we cannot afford to allow our computers to get too far out of date. But there are plenty of normal human beings still out there who cannot afford the opposite: to keep their computers up to date. Try running Firefox 3.6 on (say) a ten year old computer. But real human beings are still using machines that old. If we don’t at least serve up web sites that are usable for them, in effect we are punishing them for being poor (or in some cases, just having a life that doesn’t revolve around the Internet), just as Microsoft was arguably punished for being rich.

That’s quite aside from the issue of whether web developers and standards zealots have the right to demand what browsers people use, or more particularly, don’t use.

That’s rather akin to Superlube saying they won’t service a certain kind of vehicle because it’s not sufficiently technologically advanced. To put it another way: It overlooks the fact that the user is the one who holds that choice. If he wants to drive a 1979 Pinto, that’s his prerogative, and you should be thankful that he comes to you for an oil change.

Discovery / Tip: Controlling Textarea Dimensions

March 25th, 2010

That previous post about controlling the dimensions of a form’s textarea got me thinking.

Given that we can control the look of form inputs via CSS (using input[type='text'] ), it struck me that it might be possible to at least specify a given number of rows in the HTML markup, and assign a width in the CSS. That is the purely aesthetic concern about using rows and cols in the (X)HTML rather than widths and heights in the CSS.

Sure ’nuff: it works!

Consider this XHTML:

<textarea name=”message” rows=”8″ cols=”55″></textarea>

… and this CSS:

textarea[cols='55'] { width:470px; }

(Note that the numbers must match.)

That will give you valid CSS, and it renders perfectly in IE8, Firefox 3.6, and Opera. Unfortunately, WebKit browsers (Safari, Chrome) go a few pixels narrow for some reason. IE6, as can be expected, just ignores the CSS altogether, and renders the cols the way it would have otherwise.

Naturally, if you go this route, you’ll want to get as close as possible with your number of columns first. But err on the side of narrow, so that browsers that don’t obey the CSS won’t break your form.

Validation & Semantic Markup (or, an apologia for my non-validating front page)

March 25th, 2010

Over the past few years, web developers have heard quite a ballyhoo about “semantic markup”,” which in large part means separating stylistic/presentational elements from structure and content. Presentational elements should be handled off-screen from the HTML markup – in most cases, with CSS (Cascading Style Sheets).

Meanwhile, online validators check for well-formed CSS and (X)HTML.

For some time now, I have consistently refused to sacrifice semantic markup on the altar of validation, as I believe the two standards should be wed. (And in most cases, they can be.)

For those who care about such things, every page on this site should validate as valid XHTML, except one: the front page (of all places), where my contact form resides.

I’m unbending on that one. Why? Because the two “errors” are, in my opinion, poor judgment in the XHTML standard. These errors are because I have not declared rows and columns in the textarea in my contact form. It is my philosophy that this is a presentation issue and should be handled in the CSS. I am choosing semantic markup over an arbitrary and ill-advised validation standard.

Any other web developers out there who find themselves choosing semantic markup over validation (or vice-versa)?

Blog On Line

March 25th, 2010

The Tim Gallant Creative blog is now live!

I will modify the theme/template to coordinate with the rest of my site in the next few days. Watch this space for various posts on matters related, not only to my work, but also to graphic design, related technology, and probably some CSS/HTML/web dev tips.