The World Wide Web Consortium Promotes Standards for Web Markup. But Do Its Members Practice What They Preach?

As most within the web community know, the World Wide Web Consortium, or W3C, is “an international consortium where member organizations, a full-time staff, and the public work together to develop Web standards,” quoted from their site. Their mission is to “to lead the World Wide Web to its full potential by developing protocols and guidelines that ensure long-term growth for the Web.”

The W3C has been publishing recommendations for web standards since 1994. While it is not limited to, these recommendations include ones for web markup (including HTML and XHTML). Using valid code in the recommended standard structure ensures interoperability (technical compatibility between systems), functionality, consistency, accessibility, and a host of other benefits.

One would think that W3C member organizations are on the forefront of web standardization efforts and practice what they preach. I’ve come across research by others and conducted my own research to help answer this question.

Do the Websites of W3C Members Adhere to W3C Standards for Markup?

In 2002, Marko Karppinen tested each W3C member site for valid markup. His test found that only 18 of 506 W3C member sites. That’s right, 18.

Four years later, triin.net performed a similar test. The results were better, but not by much. Of 352 W3C member sites capable of being validated, only 63 passed the test.

I am unsure of how these tests were conducted. I assume each website URL was run through W3C’s markup validator. However, neither published their test’s design or methods.

My Methods

The following describes my methodologies on a broad level. On Tuesday, March 27, 2007, I compiled a list of W3C member sites on my computer. This list was fed into a pre-built web bot, which scanned and checked each URL for DOCTYPE declared (if any) and number of validation errors (with this DOCTYPE) in the site’s home page. If no DTD is present, this is noted and the website is validated using HTML 4.01 Transitional (the default behavior of W3C’s online validator). URL’s that could not be accessed by the bot were validated manually. (I’ll watch this thread if anyone has more specific questions.)

Findings

In short, W3C members are bringing their sites up to par with web standards. The percentage of valid W3C member sites is currently as high as it’s ever been and continues to steadily increase.

Of 436 W3C members, 426 sites were successfully scanned. Right off the bat, 106 sites (25%) had no DOCTYPE declaration! Technically, they automatically fail. When W3C’s validator encounters a page like this, it defaults the DOCTYPE to HTML 4.01 Transitional and continues the validation process after making a note of the initial lack of the DOCTYPE declaration. My test followed this behavoir, defaulting these 106 sites to HTML 4.01 Transitional.

The pie graph above shows how 94 W3C member sites (22%) have homepages that validate (all had DOCTYPE declarations). The percentage of valid sites in 2007 is much higher than those in the previous two tests. An additional 24% have between 1 and 10 errors, indicating they are somewhat close. Only 24% have 51 or more errors. So far, this is an improvement over last year.

The scatter plot above spreads out the distribution of members sites. The clumping near the bottom indicates how many sites hover around the ‘0 - 20′ errors mark. Fewer sites appear as the number of errors increases. *There are some stragglers that appear above the upper limit of this graph (between 200 and 1,000 errors), but only a handful.

Trends

These number might not seem that impressive; we’re not even close to having the majority of W3C member sites validate. But let’s take a look at the current scenario in comparison to those in the past.

Using data from all three studies, the line plot above shows how the percentage of valid W3C member sites has steadily increased over the past five years. However, it’s still a long way to the top (100% valid), especially at the present rate. * Data for 2003 - 2005 is unknown. Data for these years in the graph above are estimated values.

Limitations

• A single link, image, block of ad code, etc. can create multiple validation errors.
• This test only covers the index page of each site. Markup of subpages can differ greatly.
• This test only covers markup (not CSS or accessibility outside of the markup spec.)
• Data from 2003 - 2005 is unknown and not considered in this study.

Certainly there are more ways this test can be improved, and I invite reader suggestions.

Conclusions

Web standards are in a transitional phase and are on the rise. Many smaller organizations and agencies have been practicing them for years. Larger W3C member organizations are adopting these standards at a slow, but steady rate.

What will happen to the adoption rate in the coming years? Will it increase? Continue at the same rate? Or plateau long before it nears 100%? It would be interesting to think what these graphs will look like in 2010.

And will sites learn to declare a DOCTYPE???

Share This

↑ Jump up to the comment form

The posts Free Publication Subscriptions, Lets go on a journey, Free Publication Subscriptions, Show me your site…, Is readability a black and white issue?, are related to this post.

Or why not take time out to find out about the author of the post.