COMMENTS FROM MICROSOFT SEMINAR ON XSL/XML/ASP/XLL/ et al

Subject: COMMENTS FROM MICROSOFT SEMINAR ON XSL/XML/ASP/XLL/ et al
From: Duane Nickull <webmaster@xxxxxxxxxxxxxxxxx>
Date: Wed, 14 Apr 1999 15:26:50 -0700
Okay boys and girls:

WARNING - LONG THREAD!!!

I just spent a beautiful sunny day in a Microsoft XML seminar in
Vancouver and have to justify it as a day not wasted (I'm a cyclist and
I wanted to go riding instead).  Here are some brief notes from this
seminar.  It should be noted that this was the first XML seminar they
have hosted away from Seattle since the release of IE5.0 (according to
the Canadians).

XML/MICROSOFT & THE FUTURE:

Nothing unexpected here.  Microsoft basically pumped XML as the future
of Electronic Data Interchange for the future.

XSL vs DSO, DOM

Once the XML demo was over, I got a bit piqued when they announced there
were only 3 preferred ways to display XML other than an unstyled xml
page.  The first one presented was using the XML DOM from within an HTML
page.  The language they used for this was VBS to manipulate the data. 
The second method expressed was the use of Data Source Object binding
into an HTML page (basically an XML island within an HTML page.  This
method was expounded as better than the first although the reasoning
wasn't quite clear.

The method that Microsoft seemed most anxious to expound as the best was
XSL.  I found this quite interesting as I always viewed XSL as a
competitor of active server pages (*.asp).   Although no reasoning was
given,  it appeared that XSL is in heavy favor with MS.

They did give a demo of how active server pages can use XSL along with a
mix of DOM and DSO binding to have endless possibilities.  Although this
was also aimed at selling their own server products, I think they should
be applauded for their support of XML et al. (I know I'll get flamed for
this) 

THE QUESTIONS
At this point there was a Q & A period.  When no one out of the 400+
people offered any question, yours truly stood up and asked:

"Why did MS not fully implement some XSL functions from the W3.org
working draft while adding two proprietary functions <xsl:eval> and
<xsl:script>?"

(hehehe)  Well , you could have heard a pin drop.  The poor guys who I
put on the spot offered the answer "we don't know".  They did however
reply that they would look into this. 

Grand UNIFICATION THEORY AND MS

I then asked a second question regarding the unification of XSL,XPointer
and what the future of MSIE 5.0 support for X link (XLL)would be?  They
responded that there was no clear policy on this and until the W3.org
finalized the drafts and adopted them as an official version, there
would not likely be any further changes in forthcoming support from IE.

They then admitted that bringing out a browser as first to market
without a final draft of XSL was also likely the most probably reason
for not fully supporting XSL and adding the <xsl:eval and xsl:script
tags.

I personally view these two tags as useful and hope that the final
release of NN 5.0 supports these.  

EXAMPLES:
AMIT - they used your problem as an example and the exact syntax used
was:

In Javascript:

var doc = Island1;
var doc = new ActiveXObject ("Microsoft.XMLDOM");

In VBScript:

dim doc
set doc = CreateObject ("Microsoft.XMLDOM")

In Visual Basic:

dim doc as new DOMDocument

In C++

IXMLDOMDocument *pIXDDoc;
hr = CoCreateInstance(...,&pIXDDoc);

Now some more confusing stuff

XML OBJECT MODEL: 

Just when you thought there were enough questions to be answered, the
seminar leaders started talking about the XML Object Model.  We ran out
of time and I didn't get a clear picture of what is being proposed but
it certainly appears that someone at MS seems to recognize the need to
unify the future of XML/Xlink/Xpointer/XSL/DOM/XQL and HTML.  The XML
Object Model seems to be a move in this direction and I personally will
look forward to seeking more info in this area.

USEFUL TOOLS/READING

We were presented with a demo of Architag - an neat little XML
validator/writer.  The URL to view it is at http://architag.com/xml/

MS also released a new book on XML (Okay - I bought one) called XML in
Action - Web Technology.  It's from Mircosoft Press and my first glance
seems to indicate it will be worthwhile reading.  ISBN 0-7356-0562-9

VADLIDATING/NON VALIDATING PARSER JOBS

Okay - I hate to open up this can 'o' worms again but MS provided some
quotes that I felt could be shared.

" The Parser:

    - A software program that recognizes the rules of XML
    - The XML specification also calls this an XML parser

  In Well Formed XML:
  
    - Checks document to see if it follows the well-formedness rules

<!--See how it stops short of clarifying whether or not it actually
enforces those rules-->

  In Valid XML:
  
    - Checks an XML DTD to see if it follows the rules of XML, THEN
    - Checks the XML document to see if it follows the rules of XML then
      checks to see if the doc conforms to the DTD"

There you have it - MS 's version of Validating VS Non Validating ??? 

To me it seems as close as anyone can get to explaining the rule without
actually stating what action will be taken should the parser find the
document not well formed or valid.

If this opens up that discussion again - please don't blame me -> I'm
just the messenger. ;-p

WINDOWS DNA

MS also says that XML will be deeply embedded into the Windows 2000
DNA.  THis was not clarified but seemed logical.  

<opinion type="Duane Nickull's>
This would also indicate a strong investment into the future of XML by
MS.
</opinion>

CONCLUSION:
I would hazard a guess that we will see MS taking a lead in the future
development of XML technologies and address some issues it must face
with respect to it's own browser's abilities.  I would also say that MS
is hedging on the fact that it's new server software is ideally suited
for delivering such content in a way that will be more efficient.

MS also will continue to develop XSL and expand it's capabilities while
recognizing the need for tighter unification of the meta language spec.

Comments. replies anyone???

Duane Nickull
webmaster@xxxxxxxxxxxxxxxxx


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list


Current Thread