In this article I am going to look as some of the practices that are promoted by the Web Standards movement, practices that cost web designers (and their clients,) time and money for no real practical advantage.
CODE VALIDATION: validating your web pages.
Web Standards zealots advocate checking of code against an ‘engine’ (W3C validator ) to verify that your code is not breaking any Web Standards rules. This is fine, but it should be the ‘icing on the cake’ and not the focus.
Before you validate using the W3C validator (who’s engine does not reflect most of the browsers being used,) you should ‘validate’ your code against the target market - Internet Explorer and then (a distant 2nd) FireFox.
THE ORDER OF REAL-WORLD VALIDATION:
- Check against Internet Explorer - the most used browser.
- Check against FireFox.
- Check against the w3c validator: if you have time to burn.
TOO MUCH FOCUS ON CODE AND NOT ENOUGH ON THE JOB.
Web design is not a mathematical equation to be balanced - it is about creating engaging websites that work well for most web surfers. The Web Standards movement has created a culture of overly anal and obsessed code monkeys.
Validated and validating code does not make a great website. Your clients will not care if you code validates and people surfing the web will not care either - unless they’re Web Standards zealots!
PERCEPTIONS REGARDING THE BROWSERS:
Web Standards zealots seem to be in denial, where they consider Internet Explorer as just ‘one of the browsers’.
This makes no sense given that IE has about a 70%+ share of the market - Internet Explorer is the ‘cock of the walk’ and will likely dominate for at least a few more years because most people don’t care about what browser they are using.
Further to this, they seem to discount and treat IE like a second class citizen because it has a buggier implementation of the standards.
Remember this rule: The percentage of your concentration ( given to a browser,) should be equal to the percentage of a browser’s penetration.
CONCLUSION:
Standards are put into to place (in any field) to save time and money. There is without a doubt, good reasons to use Web Standards based design. But, when a standard starts to interfere with practical work, we need to re-evaluate.
Currently, using CSS to create page-level design is harder (more costly,) than using tables for many common layouts. The problem comes from the fact that the constructs/methods (floats, margins et cetera,) used for page-level layout are buggy.
I would also argue that the tools that CSS provides for page-level layout are weak to begin with - we need something better . For example: we need to see the CSS3 multi-column spec to come of age.
-
I am not saying that we should all go back to table based design and drop CSS for page-level layout. I am saying that we should consider the use of tables in situations where CSS is just causing too many problems -the so-called ‘hybrid’ website is sometimes a great way to go.
… let’s face it, there are certain layouts that are a snap to build with tables but are tricky and bug prone with CSS - many times I don’t think it is worth the trouble.
- -
So ends my rant against the Web Standards movement.
PS: For those of you that I may have offended - I will accept cage-match challenges as long you can cover your travel and funeral expenses. There will also be a small charge ($29.99) to cover registration and scheduling.
Do only that which makes sense, and nothing more.
Stef
I hardly consider Validation as icing on the cake. Validation points out errors you have made in the code, that is what it is for.
Aside from professional standards of building the best site possible, those small errors that creep in can cause outrageous problems depending on what user agent is using that code, ths can cause broken designs or unuseabe sites.
In addition as more and more developers choose to go with XHTML it will be of utmost importance. XHTML when served as XML willnot continue loading like HTML, it will stopand show an error at the first mistake, so the user never even will see the site or the content. Sites must be validated to esure they will even load in a XML enviroment.
Is validation a requirement? No, but it is very important for many developers formore reasons than just web standards zealots saying so. It is proof reading for developers to be sure our code is error free which will help reduce the cross-browser problems that plague us.
As for Firefox, it is the majority browser for Linux and Mac. IE is only the majorty browser on the majority OS. The number of Mac users is growing quickly as people tire of security problems and more of the professional market are using it as well including the FBI. Linux is increasing it’s user base as more people tire of the increasing price and system requirments of each windows OS.
Also the majority browser ofr handheld devices in Opera.
So the IE is king idea only covers one OS (even if it is the most spread) and one form of internet access.
Other than that, much of the rest I agree with, though this one a little less then the other.
?
-Validating using the W3C validator has nothing to do with browsers, it just checks to ensure that the code you have written is valid in-line with the DOCTYPE you have used.
Also: “not breaking any Web Standards rules.”
-Web standards are not rules, they are guidlines - a set of standards created to help designers, developers and software and hardware manufacturers work better.
And: “The Web Standards movement has created a culture of overly anal and obsessed code monkeys.”
-overly anal and obsessed code monkeys were around a long time before web standards, the web standards movement certainly has not created them.
I have to say thank you Stef for saying what I have been feeling for a long time. I saw a post on a web site recently of a designer proudly proclaiming that his site validated to some standard and IE couldn’t render it at all.
This guy was really proud of his achievement - which blew my mind - of course it means his site wont even be viewable by the BULK of internet users.
As for the arguments of PDA compatible websites - how many are actually being used extensively - if at all? Not one of the 30 or so clients I have built web sites for have ever seen that market as a target. If they did then an alternative version might need to be coded, but to to do on the off chance that they might require it at some stage is just too costly.
As for the validation process, see how many sites incorrectly nest forms to ensure that they don’t work as a block level element and drop another line further down. It’s a ‘technique’ that any validator would error on, but one that is widely used.
I fully agree with the “client doesn’t care whether it validates or not” argument. It seems to me that several of the zealots involved in pushing the web ’standards’ (that I know of) don’t seem to build web sites for a living.
In my opinion I do not believe you can build web sites for a living without (at some stage) having to resort to ‘old school’ methods to get things done - like (god forbid) tables for layout (oh no - end of the world).
Cliff
Final comment:
I’ve relaxed my opinion regarding validating code against the w3c validator (or any other code validator) largely due to the need of building accessible websites.
Don’t get me wrong, I still think you should check against the primary/target browsers first, over the validators … but I will concede that code validation does have some value as a tool for web designers.
Credit goes to Kyle.
Stef