Hello, allow me to highlight Kurt Cagle comments on XAML written up in the blog story titled "XML In the Crystal Ball: The Rise of XML Frameworks".
Kurt writes: XAML- the 800 Pound Gorilla? I've deliberately held off talking about XAML/Avalon until last. XAML of course was one of the major stories of 2004, an XML based application framework development being developed by Microsoft, originally to be deployed as part of Longhorn. XAML represents a critical turning point for Microsoft, although it has been moving towards a more XML based representation of their underlying API for years. XAML definitely falls into the frameworks category - with thousands of tags and a rich, robust application model, XAML represents an attempt to take Microsoft out of the API model that predominated during the 1990s and into a just in time architecture that stresses sophisticated data binding and a fairly daring move into the 3D space. There is no question that XAML is THE 800 pound gorilla in the XML framework space. For a product that is still at least 18 months from seeing the light of day, it is sending shockwaves throughouth the industry, though its long deployment time may very well end up working against it as it has given a deadline for many other companies and open source movements to establish their own frameworks as a consequence. As far as the technology goes, XAML is large and comprehensive, an effort to define much, if not most, of the Microsoft API in XML terms. In many ways, XAML will end up most clearly articulating which of two competing approaches to declarative frameworks development will be dominant through the next couple of decades. The XAML approach is very monolithic - namespaces are used to identify particular packages of classes at a very granular level, which in turn means that the entire XAML namespace packages number into the hundreds. This is the approach that is taken by other proprietary frameworks as well (such as Flex), with much of the inherent functionality handled by "Code Behind" blocks that define functionality through the use of C# or Java classes.The w3C approach, on the other hand, has been to build at a less granular level, but with more emphasis on an XML-ish orientation to code development and data binding. Just as Flex has, XAML has created an entire cottage industry of XAML "viewers", which use the XAML specification to create stand-alone XAML implementations. Xamlon, run by Paul Colton, is a good case in point - because the specific behavior and interfaces of XAML are published, Xamlon has implemented an application (including, ironically, an SVG converter and Flash generator) that can create XAML based applications. Similar products are being handled through Mobiform. This of course raises an interesting set of questions. One of the purposes of XAML is to effectively replace much of the web's infrastructure with a XAML infrastructure, providing superior tools for people to write sophisticated applications, provided of course, that they run on a Windows platform. This was a fear expressed early by Ximian founder Miguel de Icaza. Ironically, I suspect that this fear is overblown. XAML will be making its way to XP in 2006 (one of the more surprising announcements of the year, as Avalon was supposed to be a major enticement technology for upgrading to Longhorn), but it will also be sharing that space with a supercharged Firefox that is beginning to breath real life into many of the open-standards XML technologies that have seen relatively little light such as XForms and SVG. XAML also won't be running on earlier versions of Windows, let alone on the Macintosh or Linux (unless some open source group decides to build a XAML implenentation there). While it may in fact end up replacing certain intranet development efforts,I suspect that it will face a much larger hurdle on the home desktop, especially as Avalon running on top of XP likely doesn't really solve many of the security problems which currently plague Internet Explorer. Thus, it's likely that XAML may end up taking a reasonably healthy share of the XML frameworks space by 2006, but nowhere near the penetration that Microsoft marketing would like to see. Additionally, given that XAML is an XSLT transformation away from SVG and XForms (albeit, a LARGE XSLT document), it's much more likely that you'll see bridge technologies springing up that will make it possible for XAML to interoperate with the rest of the XML world, and VICE VERSA. What this may end up doing is actually help the formation of certain open standards that currently have been in limbo (I'd love to see XAML's 3D capabilities spur such standards, for instance, as I'm not impressed with X3D. Source: http://www.understandingxml.com/archives/2005/01/xml_in_the_crys_2.html What's your take? Do you agree Kurt that XAML is sending shockwaves throughouth the industry? - Gerald PS: Kurt Cagle will speak next week at the VanX. See http://vanx.org/next_meeting.htm for details. ------------------------------------------------------- The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt _______________________________________________ xul-talk mailing list xul-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xul-talk
