Re: New/old pattern syntax, why can't we have both ?

At 10:34 AM -0400 8/26/98, Paul Prescod wrote:

>#2. If the concept of a query is going to be generalized beyond XSL (which
>it inevitably will have to be) then being parallel to the generation
>syntax is not a benefit.
>
>
>Even if it is not generalized, being parallel could also be confusing.
>Let's please remember the lesson of the overwhelming popularity of
>languages in the LISP family. People don't like so much consistency.
>"Consistent inconsistency" is easier to scan visually.

Seems to be a question of what the currently "in" consistency is.
The aforementioned confusion, I'm told, was one reason SGML DTDs were
not written as tagged documents themselves.  And we know how successful
FOSIs were.  But today consistency of notation for disparate things
is in.  Perhaps the difference is that original SGML, like LISP, saw
the disparate things as part of the same whole--all part of the program/
document--whereas the XML/FOSI point of view is that they are separate
documents/data-sets, linked by system-dependent, nonstandard means--we
aren't encouraged to consider them at the same time.

Dave Peterson
SGMLWorks!

davep@xxxxxxx



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