xml-dev - Creation vs. Long-term Maintenance of Declarative Language Scripts

["G. Ken Holman" <gkholman@CraneSoftwrights.com>]

At 00/08/25 16:39 +0200, Eric van der Vlist wrote:
>While I foresee a wide opportunity and generalization for the usage of
>XSLT transformations, I don't think that the development of such
>transformations will take a significant part in the costs of development
>and maintenance.

Bingo, Eric!  Thank you!  You've identified the biggest problem.

What is the length of time to develop the first version of a stylesheet (or 
a DTD, or any file in whatever declarative language you care for) compared 
to the time taken to maintain it?

Even if you don't write the file yourself ... if you contract the initial 
creation of something and you are left with the maintenance, your WYSIWYG 
tool must be able to accommodate a declarative script written by 
hand.  ***AND*** your WYSIWYG tool must be able to accommodate the 
subsequent writing or maintaining a script by expert hand without 
interfering with previously embedded comments, constructs, additional 
information, etc.

>Compared to the costs related to the design and content creation of a
>site (or of an IT system), I strongly believe that it's not significant.

Agreed!

Plus, consider how much of maintenance depends not on the constructs used 
in the declarative language, but in the comments, the documentation, the 
flow diagrams one may edit into their file to help the maintainer, etc., etc.

I feel that skill and customer satisfaction should be measured *more* by 
how successfully the customer is going to work with the results supplied by 
a craftsman than by the initial task being solved in the first place.  The 
craftsman may have changed employers, may be too busy to help, or may have 
taken up glider building instead of IT (a good friend left the SGML 
business to build gliders in his garage) ... where is the customer then?

The same goes for the tool itself as what I just said about a craftsman.

If the WYSIWYG tool does not successfully mesh with the needs and practices 
of the craftsman, then the WYSIWYG tool is not meeting *all* of the 
customer's needs.  I've seen WYSIWYG tools marketed solely based on getting 
people started using a declarative language, which is fine for doing just 
that, but the users don't think to look in the long term or in the bigger 
picture.  And neither do many tool manufacturers (or at least their bosses 
who won't fund the programmers who are creating the tool to properly 
research the needs of the craftsman!).

Anyway ... sorry to get carried away ... this has been a sensitive issue 
for me since 1993. :{)}

.................. Ken

--
G. Ken Holman                    mailto:gkholman@CraneSoftwrights.com
Crane Softwrights Ltd.             http://www.CraneSoftwrights.com/x/
Box 266, Kars, Ontario CANADA K0A-2E0   +1(613)489-0999   (Fax:-0995)
Web site: XSL/XML/DSSSL/SGML services, training, libraries, products.
Book: Practical Transformation Using XSLT and XPath ISBN1-894049-05-5
Article:          What is XSLT? http://www.xml.com/pub/2000/08/holman
Next public instructor-led training:     2000-09-19/20,2000-10-03/05,
-        2000-10-09/10,2000-10-19,2000-11-12,2000-12-03/04,2001-01-27