Crawlable AJAX? Not Really.

Perusing Google's Central Webmaster Blog, I came across a proposal written as an attempt to make Client-side, Rich Internet Applications (RIA) built in AJAX crawlable by search engines.

Upon reading the introduction paragraph, followed quickly by the goals of the proposal, an inkling of skepticism and pessimism quickly grew within me: this proposal had been attempted before -- many times before -- by many different people. But this was from Google, or at least, from somebody who has permission to write on a Google blog. Perhaps there's more to this than meets the eye. But, alas, in spite of the Google wrapping, as I continued to read on, it appeared to be nothing more than the same proposal to a complicated problem -- another proprietary standard loosely based on URL structure.

The popularity of Client-side RIAs is seen throughout the Internet: dazzling us with their ability to transcend beyond the world of simple http protocols and static pages. But as popular as these sites may be, there is no consensus amongst RIA developers as to how to build an RIA. While this "without-borders" approach promotes creativity and ingenuity in RIA development, it also creates a scenario reminiscent of a two-year old feverishly colouring outside the lines - and the lines are accessibility.

While this particular proposal addresses some of the needs of the search engine (actually, only Googlebot), it de-couples itself from the actual content of what the users may see. At the end of the day, a search engine must have a static URL (or deeplink) representing an instance of the RIA. This means that RIA developers must build a correct framework, and follow a standard protocol for the search engine to index it. But the problem is, if every search engine releases their own proposal, you're looking at many standards, and more complication. So what should the proposal be?

It amazes me that search engine companies, including Google, have never come to terms with what their products truly represent: visually-, and auditorily-impared people. Starting with that fact in mind, search engine companies should really be focusing on promoting web initiative projects like WAI-ARIA: projects that are attempting to bridge the gap between disabled users and RIAs. From that perspective alone, you are not only promoting RIA developers to use a set of defined standards to make their RIA pages accessible to the seach engines, but are also making them accessible to everyone else.

The WAI-ARIA standards are already being adopted by many screen-readers, which are helping disabled users to view RIAs in much the same way as the normal user. Priming these standards with other semantic Web initiatives like RDFa into a singular, extensible, microformatted definition list would further enrich the user experience, and remove the current trend of de-coupling what a normal user sees, vs. what a disabled user (including search engines) would see.

Hopefully, the next proposal from Google (or any other search engine company for that matter) will follow a pattern similar to the one mentioned above. It would really be dishertening to have to create a different version of an RIA for every search engine.