There is a long-term view of Mobile AJAX, and a short/mid-term (or current view) reality about it…

Let me start with a brief overview:

AJAX is web-browsing (UI) technology.

AJAX implies a set of technologies: it's client side scripting that uses HTTP, JavaScript, (X)HTML, DOM (as in Document Object Model), and maybe even CSS (style sheets).

AJAX caught up the way it did because it allows web applications (web-based user interfaces) to be more dynamic and responsive, thus more interactive and friendly. With AJAX, screens/pages don't have to refresh completely, but instead parts of it can update/refresh independently as needed. In addition, JavaScript client-side scripting allows for others nice bells-and-whistles, resulting in a “wow” effect.

What about mobile handsets?

Mobile AJAX is mobile browsing, but mobile browsing is not necessarily Mobile AJAX. Mobile AJAX is not services on the web, but might be used to consume such services on the web.

With their small screens, does AJAX really matter for handsets? Will it have the same effect as it did on the desktop?

The answer is that it will matter for mobile, but it won't have the same effect as in desktops. Not until 1) AJAX is consistently supported across micro-browsers, and 2) not until the primary screen on handsets are large enough to make it exciting (and the day the mobile handsets primary screen become large, then the whole mobility thing may have been defeated).

So AJAX is not supported well across mobile/micro-browsers, with Opera probably being the mobile browser with best support for it (if not the only one). So this means that Mobile AJAX is not mainstream today, and won't become mainstream until support for AJAX itself becomes mainstream (across mobile/micro-browsers).

Once AJAX becomes pervasive/mainstream across mobile browsers, things will be different. The third screen will never be as large as a desktop screen, so AJAX's wow-effect will be a limited kind of wow-effect. But leveraging client-side scripting will be beneficial for things such as search and filters (where as you type you can see the filtered results), or downloading resources progressively, resulting in a better experience, and if properly designed, bandwidth can be preserved, by only refreshing data, vs. the whole page.

And once Mobile AJAX has gone mainstream (across all micro-browsers), and regardless of screen size, one thing will be (for some of us, very) exciting — Widgets: those smart, visual-rich components (on the client) that are very focused and very good at a few/specific functions, such as weather, data feed, access to personal information, and so on. But these Widgets will never be as rich as the “local” Widgets (rich Widgets written in Java ME, or Symbian, and so on) with access to native functionality. Mobile browsing will never be as rich as local applications until micro-browsers provide access to native functionality. And when that happens, we will enter a new phase on Mobile browsing, and new concerns on provisioning and security.

As I've previously written, this means that for this to catch up, we need a common and cross-platform way to code and run these Widgets — see Widgets: the next portability nightmare?.

I would like to encourage all micro-browser vendors to get together and lead and agree on providing consistent support for the core technology for mobile/micro-browsers and AJAX: JavaScript, XMLHttpRequest, DOM, HTML and CSS — note this core functionality is the W3C Web Integration Compound Document (WICD) Mobile 1.0. But also, lead and agree on how Widgets are authored, provisioned and executed (runtime).

Last but not least, Mobile browsing and AJAX will not make other types of mobile clients become obsolete – not local clients written for Java ME, or native clients, or Flash… And it won't because the different mobile application types may or not provide and address different value propositions, different use-cases, different experience, different cost of ownership and ROI, different actors… to both users/consumers, as well as the producers/developers – there are many factors to consider, making the decision not as “simple” as some people may want you to believe…

In conclusion, Mobile AJAX today is not mainstream, and is not ready to deliver the goodies it has delivered on the desktop, not because the technology per-se isn't ready, but because it is not consistently offered across handsets, and because its effect (and experience) on Mobile is just plain different from the desktop. So when can we expect Mobile AJAX to go mainstream? Well, the sooner the better, but it will probably take about 2 years — hopefully I'm wrong and it's before that.

ceo