After following some recent discussion on the WHATWG HTML5 mailing list something has become apparent to me that I find interesting to say the least.

From reading Ian Hickson’s (the editor of the HTML5 standard) posts to the mailing list it is clear that he believes the HTML5 specification should only document what the browser companies implement. Here is a quote of Hickson from a recent thread in the mailing list.

…HTML5 is largely built on the idea of speccing the de-facto standards, either long after they were implemented, or in tandem with them being implemented. Very little of HTML5 has been ahead of implementations.

The most interesting discussion of late is whether or not the spec should include language that would specify Theora as the baseline video codec browser companies should support in the display of <video> tags.

Hickson believes that the inclusion of this language in the spec would not be productive. In his own words:

I don’t see that the spec saying “must…Theora” would have any effect. If anything, I think it would be a negative effect…

Supporters of the open web believe Theora support should be in the spec since it is the foremost patent and royalty free codec. I posted this comment to mailing list in the midst of the discussion about codecs, Hickson’s reply follows:

…the fundamental question here is whether or not the spec should be concerned solely about creating a standard that is satisfactory for implementers to follow, or if it should go further and try to make the standard work well for everyone involved including developers and consumers.

I am certain that most would like to have a standard that best serves the entire community.

It doesn’t help the community if it’s not implemented.

What has become apparent to me is that the HTML5 specification is effectively strengthening the stranglehold that browser companies (and the technologies that they support) have on the web. Apple refuses to support Theora. They cite the possibilities of submarine patents as the reason. However, it much more likely that they want to use their influence to promote H.264, a patented codec in which they have invested heavily.

Another interesting tidbit is that YouTube (owned by Google) seems unwilling to switch to Theora when they make the move to HTML5. That means that if the spec included Theora YouTube would theoretically have to use it to utilize the <video> tag. Using the <video> tag would be a great step forward for YouTube as they could finally drop Adobe’s Flash. The reason YouTube would not want to use Theora is because they already have all their videos encoded into H.264, and it would be very costly to move to Theora. Did I mention Hickson is employed by Google?

I find it very disappointing that the browser companies have achieved this much control over the web standards process. Instead of letting the community of web developers, designers, users, and browser companies collectively decide what the web should be the WHATWG, led by Hickson, has derailed the standard to fit what the browser companies decide to implement.

I believe this kills innovation and competitiveness in the browser market. Instead of pushing the web forward this standard just lets the browser companies market themselves as standards compliant when in fact the standard is merely browser compliant.

In the case of the <video> tag, HTML5 in its current form enables proprietary codecs to continue to control the use of video on the web. This flies in the face of the open web.

The bottom line is that this practice of merely documenting “the de-facto standards” cements the power browser companies have over the web. It effectively butts web developers and designers out of the process and replaces them with what works best for the bottom-line of the browser companies.