OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

 


 

   RE: [xml-dev] Microsoft Hypes Up XUL As The Greatest Expiriment S ince A

[ Lists Home | Date Index | Thread Index ]

Yes, and the extensibility looks interesting.  The whole 
transform bit is exactly how Goldfarb described these 
systems a decade ago, so this in some ways as Gates 
admits in his Longhorn interview at CNET.COM is a 
lot of old dreams coming to fruitiion.

My favorite part of the article is where he says it 
is just an old Windows resource file.  That is exactly 
what the U of Waterloo paper suggested, and exactly the 
inspiration for MID I.  Sometimes, XML seems to lose 
track of the "markup is a verb, not a noun" adage.  It 
is so obvious:  look for anything with names, properties 
and values and tag it.  

This will be fun.  We never got to the 'compile it' 
stage in the MID demos.  It was interpreted, but even 
then, we knew the performance of that would suck.  We 
weren't attempting to remake the operating system in the 
image of a browser; we were decoupling behaviors from 
presentation for IETMs.  Turns out to be the same problem 
in either case.

Now I wonder if this helps the "the web is stateless 
so one has to linearize" problems that turn performant 
QBE GUIs into lost-in-space wizards.  If XAML is extensible, 
will business rules move back to the client?

len

-----Original Message-----
From: Mike Kozlowski [mailto:mlk@klio.org]

On Wed, 29 Oct 2003, Michael Champion wrote:
> On Wednesday, Oct 29, 2003, at 09:23 America/Detroit, Bullard, Claude L
> (Len) wrote:

> > It is the performance of the mush of different XML
> > languages and objects that would make me wince.

It gets compiled down.  XAML is the source-code format, not the executable
format.

> >  One
> > has to ask, why do this in XML at all.
>
> I've been wondering that myself.  Isn't the whole point of XML
> portability (or interoperability) across platforms, at the price of
> "bloated" data and "inefficient" performance?  That's usually a good
> tradeoff, but not if you stay in a proprietary box.

That's a point of XML, but I don't know that it's the only one.  If you
were making a declarative, human-readable UI-description language,
wouldn't you use XML even if you suspected that the only application to
use it was your own resource compiler?  Bloat and inefficiency aren't
concerns here; developer/designer ease-of-use is.

And if you add in the possibility of making XAML a target of XSLT
transforms, or an output format of graphic design tools (apparently, Adobe
demonstrated something like that at the PDC), you do get some exchange and
interoperability, too.




 

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS