« Ruby on Rails : Impressive, will surely shake up web development. | Main | AJAX : Demystifying the buzz for all platforms. »
May 3, 2005
XQuery not likely in .NET 2.0, strange to say the least.
It's common for certain specifications -- the W3C kind -- not to be completly adopted in commercial products, yet companies generally go to great lenghts to align themselves as much as possible with these type of standards, emerging as they may be -- Browsers come to mind as an example.
| [Entry continues to the left and below ad ] |
Yet, it is not every day you see a company say flat out it's impossible to adopt a specification even though they have a say in the standard working group, and on top of it all, have their most distinguished group of users petition to incorporate a particular standard.
This is the case for XQuery and Microsoft's .NET 2.0 version. As it turns out, a group of MVP's (Most Valuable Professionals) has petitioned the company to support XQuery in its upcoming .NET 2.0 release .
While the SQL-type language for XML is not completly being ignored by Microsoft -- it is partially being supported in other products like SQL Server 2005 Express -- what is surprising is its current stance on XQuery for the more influential and milestone release of .NET 2.0 .
Considering the fact that other companies are incorporating XQuery into their core products -- even though the specification is not a final release yet -- such as BEA's Liquid Data and SoftwareAG's Tamino XML Server, it may raise some questions about Redmond's commitment to the standard.
For now, we will have to wait and see if Xquery makes it into .NET 2.0, and wether this decision was strategic in nature or simply a matter of avoiding scope creep in the release, and in the process, learn how Microsoft deals with petitions from their power users.
| [Comments below ad ] |
Posted by Daniel at May 3, 2005 9:11 PM
Comments
Hi Daniel
You may want to check out the official Microsoft statement on this at:
http://msdn.microsoft.com/XML/XQueryStatus/default.aspx
And my explanation why SQL Server does indeed support a subset in its next release at:
http://sqljunkies.com/WebLog/mrys/archive/2004/12/20/5654.aspx
I can tell you that we got badly burned by doing XSLT early...
PS: Most of the people on the petition are not MVPs...
Posted by: Michael Rys at May 22, 2005 1:30 PM
Michael Rys already hit the most important points, but let me make a couple more. First "commitment to the standard" requires there to be a standard. What there is with XQuery is a working draft of something that will presumably be a W3C Recommendation someday. It cheapens the idea of "standard" to say that every draft spec from the various relevant standards groups should be impelemented by the major vendors. That would create chaos, not interoperability.
Also, products such as Tamino and Liquid Data (and SQL Server for that matter) are in a different category than .NET -- they are not shipped as a core infrastructure component on hundreds of millions of computers. There simply has to be a higher standard for what is shipped to a large percentage of the world's software users vs that which is shipped to people with a specifc need for some draft Recommendation.
Finally, this is not a petition from MS power users, this is a marketing ploy by a company that invested in building XQuery front-end tools, betting that the market would have emerged by now. It hasn't, and they want help creating the market. Perfectly sensible from their point of view, but not a reality that Microsoft should address until we hear from our REAL power users and MVPs. It's my job to listen for this, and I've asked repeatedly on a number of public forums for people who care about XQuery in .NET to let us know their needs, priorities, timetables, etc. The response has been, for the most part, deafening silence.
The input we get from real MVPs is mainly a) XSLT 2 is more interesting than XQuery on the client side, invest there if you have to choose; and b) thanks for NOT playing 800lb gorilla and leaving some interesting niches open for specialist companies and open source projects to fill.
Posted by: Mike Champion at May 22, 2005 7:16 PM
Post a comment
Track back Pings
Track Back URL for this entry:
http://www.webforefront.com/mtblog/mt-tb.cgi/7.
[Any other Reader]
