W3C Validation - The Controversy Continues
by Miriam
November 26th, 2006
The highly-opinionated, and alarmingly on-target Aaron Wall of SEO Book just put up a truly provocative post entitled Bad Advice that Sounds Good which basically points a finger at some of the most common, generic statements people in the Web/SEO industry make to clients, and I suppose, to one another. His list of 5 comments he hates are certainly worth a read, but one of them, in particular, happened to really hit home with us this week:
“Validate your site. Why do I hate it? Most successful sites do not validate.”
Almost immediately, and quite predictably, a reader responded that Aaron must be joking to say that people shouldn’t worry about validating their websites.
I nearly choked on my tea when I read Aaron’s brief yet eloquent retort:
“My website does not validate.
Neither does yours.”
One of the ensuing comments only intensified the insanity of this debate. The reader pointed out that neither Google.com, Yahoo.com nor Live.com validates. So, folks, if the big guys in search are running websites that don’t validate, how important can this issue really be?
3 W3C Camps
In the first camp, we have the purist coders who would no sooner put up a non-validating website than a chef would serve up apple pie a la mode, sans the ice cream. I’ve run into these brainy fellows on forums, and while I respect their brilliance, their adamance about the W3C issue sometimes borders on the fanatic. I envision them having nightmares about putting up a page that is - oh horrors - actually missing an alt tag on one of the images!
In Camp #2, we have more middle-of-the-road folks who say, yes it’s good if your website validates. Why not take the time to make it validate? These folks seems to have gained some peace of mind knowing that their site does pass validation, and I’m guessing that professional pride is what is most at stake here.
In our third camp, we have the fellows like Aaron Wall who are laughing at the whole thing, pulling in top rankings, huge traffic, etc. with websites that break all 10 of the W3C Commandments.
Current thought on W3C and SEO as of November 2006
I have yet to see it documented that validation does diddly squat for your prestige in the search engines. On the contrary, websites that were built in Microsoft Frontpage in 1997 are still topping the SERPs in many fields. Talk about code that doesn’t validate! Accepted experts seem to agree that it just doesn’t matter whether your website validates, in terms of how your site ranks.
It is, however, important to mention here that websites that do validate may be more accessible to individuals with handicaps. As we all know, Target is finding out that not being friendly to their differently-abled users is not a good way to get good press. Will someone sue your grandmother for not making her Jams and Jellies website validate? Probably not in the near future, but legal action may become a concern for all website owners at some point, and in the meantime, being courteous to blind and deaf people is certainly not a step in the wrong direction. Indeed, accessibility is really the ONLY convincing argument I’ve heard from the pro-W3C camp.
Why Aaron’s post hits home
We’ve recently taken on a contract with a company who was adamant about having their website validate. Of course, we agreed to provide this if it was a priority for them. Then came the headache. My husband decided to try validating our own homepage first, to get the feel of what this W3C stuff is all about. Yes, you heard me right. Our homepage, which ranks Top 10 for web design in our local county, did NOT validate. Please, code fanatics, don’t send us hate mail. Liam has just spent the entire afternoon and evening rectifying the most trivial issues in the homepage code and, wonder of wonders, the happy This Page is Valid HTML 4.01 Transitional! salutation finally appeared. He will now need to go and lie down for several hours. I don’t like to think of him having to perform this task with the other umpteen pages on our site. Why do I feel we’ll be procrastinating on this???
However, we will give the new client what they want, because it will make them happy. That’s what we’re being paid for after all. But, our own personal experience with this is that W3C was never necessary for our homepage to rank well and it was a big pain in the neck dealing with it.
What I really am not liking about this issue is that I am hearing it become a cop-out for when problems are happening with a website. “Your site doesn’t validate”, is starting to have the ring of, “Clear your cache.” 9 1/2 times out of 10, I would doubt the advice of either statement when being given it in order to solve a serious problem.
In conclusion, W3C validation is an extremely tedious process, which appears to have no measurable effect on SEO, but which may be a good idea anyway in order to make your site a better place for blind and deaf visitors. And to those of you in camps 1 & 2, please, don’t look at anything but our homepage. I KNOW the other pages don’t validate. I’m busy. I have to make dinner. I have to do the laundry. I….oh, all right, I’ll try to get to it.
Next Article: 11 SEO blogs you probably missed Previous Article: Ultimate Onpage SEO Guide
November 27th, 2006 at 2:35 am
Great post Miriam!
My thoughts on validation are pretty simple though I used to be a purist.
Validation does not help ranking well in the SERPs. It does however ensure your code can be read accurately by the search engines. There’s some peace of mind in that.
SEO really comes down to two things: content and links. I’ll leave the argument over which is more important to someone else!
Validation provides a minimum level of accessibility for disabled users which are around 10% of all internet users. That’s a pretty big chunk to ignore and being able to serve them is a competitive factor when almost nobody else does. Disabled users tend to be very loyal as well. They know the extra effort put in to make the site work for them and they appreciate it.
Validating makes the user experience almost identical in all browsers. Well, not really but it does shorten the development time to make a site work well in all browsers which is important.
Being able to code at the level where a site validates gives you a much better understanding of your product (as a web developer).
I think most people have forgotten the original browser wars between IE and Netscape and the expense and stupidity that came from it. The W3C guidelines were an answer to that and while the guidelines themselves aren’t perfect, the implementation across various browsers leaves a lot to be desired. As someone that codes by hand I know that better support from browser manufacturers would make coding much faster and easier and even… less expensive.
PS. The argument that the search engine sites don’t validate isn’t entirely true. I have caught Google and MSN on days where they did! Besides, they are the big dogs. They can “do as I say, not as I do” all day long - at least for now.
November 27th, 2006 at 2:11 pm
“The argument that the search engine sites don’t validate isn’t entirely true. I have caught Google and MSN on days where they did! Besides, they are the big dogs. They can “do as I say, not as I do” all day long - at least for now.”
Haha, very true, Rand. I’m glad you like this post.
You know, apart from being a pain in the neck, in good conscience, I should add that it was kind of interesting to see why our code didn’t validate. Like you, Rand, we’re old school editor users…we use Textpad. No WYSIWYG for us. Most of the problems the validator found were related to what I’m assuming became outdated tags that my husband learned when he first learned HTML years ago. Oddly, these tags were for things like backgrounds, color attributes, etc. that technically always worked just fine. But, I guess the way we coded them isn’t ’standard’ so this is why our pages didn’t validate. That seems to us to be the funny thing about this whole validation issue…even if a tag works, it still may not be valid. Weird!
Now, I have a gripe with Liam. He taught me how to code and he taught me wrong! LOL.
Kind Regards,
Miriam
December 17th, 2006 at 6:38 am
I think the big issue is not full validation - rather it’s being able to parse a page on a block level or not. Google doesn’t care less if your site only uses the allowed attributes, it doesn’t even understand XHTML. However, if your page is not parsable, it will get the “text-reader” treatment and that could mean that things like context and semantic markup (headers, highlighting, etc) get lost. I don’t think there is anyone here who would code a page without using markup like that - why risk that it’s ignored?
December 17th, 2006 at 11:24 am
How do you know Google doesn’t understand XHTML? Google does understand elements within it like font, bold, italics, headline, anchor, etc.