Building blocks for Web applicationsThursday, June 24, 2004
As Dave Shea mentions, Web Apps are Hot. The clamouring for the ability to do web applications is there, however the big concern is which collection of building blocks will be the better option for web applications.
There are a number of alternatives, ranging from the simple enhancements, right through to massive chunks of new software.
The current web platform
From a practical point of view, this effort has more potential to be a realistic option for web developers. WHAT WG started off as a result of the disillusionment of parties toward the W3C Web Applications and Compound Documents workshop.
The strength and flexibility of Mozilla as an application framework is neatly exemplified by its browser application Firefox. Recent evaluations of browser conducted by magazines have scored Firefox above Internet Explorer 6 as a browser. Considering Firefox is "merely" an application build on-top of the Mozilla application framework this is an incredible achievement against a specialised browser.
The Mozilla application framework is available to be used now. There's no waiting around for specification groups and vapourware.
The RICHIE project
RICHIE is a loose collection of developers building tools for handling XML user interfaces. Whatever your preferred scripting language, there's probably a XML user interface library ready to be used.
The problem with this group is that the main area of overlap is using XML as a GUI descriptor. There's no evidence of a push toward interoperability between applications. At the moment, its a lack of focus or direction that keeps the movement being eclectic rather than mainstream.
W3C Web application and Compound Documents
Earlier this month the W3C held a workshop on web applications and compound documents. The proposed direction from the original workshop looks to be clear. Web applications based on XHTML2, SVG and XForms. The belief is that the browser is a dead platform, and work needs to be done outside of that enclosure.
There's a drawback to this group's long range efforts. The time needed to produce a mature, deployable platform to enable web applications. At the moment, interested parties know who they are and what they individually want to achieve. The collective path is not yet fully defined, and the process of creating a specification is not yet underway.
Microsoft's Avalon and XAML
Alongside the W3C interest, Microsoft's approach is also at the vapourware stage. Nothing is deployable at this stage, and nothing is expected to be deployable until 2006.
At this stage, only MSDN subscribers have an genuine inkling as to what Avalon and XAML can deliver. It is all heresay at the moment. But its enough to worry open-standards developers.
The browser is an integral part of the web. Moving people off that to other applications is essentially starting from afresh. Mozilla and RICHIE are the two web application frameworks that works today. The others are in the various stages of specifications and prototyping.
Web applications is a fertile ground for the web. I firmly believe that the building blocks are based on open standards and are not strongly tied to vendors.
In my opinion, the two groups to focus on are the WHAT WG and the Mozilla platform. But keep a watchful eye on the W3C working group - that will be interesting, if not immediately useful. Avalon and XAML is something to be weary of.