It's time for practical XML!

[len bullard]

Jonathan A. Borden wrote:
> 
>         Perhaps of the greatest attractions of XML is hopefully the ability to
> bridge disparate systems, communications protocols and interfaces (in the
> broad sense of this term).

Yes, that is what markup technologies have been about since Goldfarb,
Mosher and 
Lorrie released their seminal work.  XML has by dint of its web origins
brought 
many people outside the tradtional publishing crafts into the game.  So
now 
ideas that were once considered only by the left wing lunatic fringe of
SGML are 
being fought for and embraced.  Any of the MID team vets would love SOX,
VML, etc.
 
>         If the world were 100% java there would be less need for XML solutions to
> interoperability ... all objects would talk to eachother in RMI and
> serialize themselves as java objects ... and theoretically all systems would
> interoperate - right?

Well... err... if a VM wins yes.  I don't think we ever get more than
80% done 
before the last old thing becomes the next new thing and the CS business
takes 
another round though the Moebus loop.  Interfaces are critical but let
the 
ISO Java committee have that problem.  ;-)

>         I doubt this will be reality in my lifetime and at least for what I can
> forsee as the future, there will be lots of good work connecting and
> bridging legacy (and new :-) systems. XML is poised to play an important
> role in this critical activity.

Sure, yet the longer I've been around, the more I've come to appreciate
the 
80% victories.  The other oft overlooked benefit of markup is lifecycle 
support.  Moving tagged content onto new machines without having to 
worry about which way a platform counted significant bytes was a lot 
easier when the publishers (usually DoD contractors) bit the bullet 
and did SGML.   IETMs were dammed expensive and hard to produce, so 
not too many were done.  When you consider that the text in the 
caption property of the object is moreorless the same as the footer 
of a page in a book, one realizes just how expensive the next round of 
moving content to a new platform can be if we don't use markup.  

If you saw what goes on trying to deliver and validate the 
contents of a relational database to the FBI (eg, incident records) 
when they are still using byte offsets as record separators, 
as a taxpayer, you would revolt.  It is expensive yet most of 
your public safety systems vendors have to vet their software 
for precisely that kind of delivery.  Now think that this 
is a foodchain starting from the time a 911 call is logged, 
through the local police department, then to the state, then to 
the Feds.  At the final step, those records are used to set 
policy for administering the law enforcement agencies and 
advising law givers worldwide.  Scary.  Done every day, though.

IOW, while XML-based components and objects are exciting to the 
programmers, the grunt problems of delivering and maintaining 
very large databases of mixed document types are still major 
drivers for using markup.  That is why where some still like 
to fear or castigate Microsoft, their support is the vital event. 
Now it is an issue of selling MS-based products for XML to 
communities of developers weaned on RTF and Do While Not EOF to a 
recordset.  

len

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)