Re: SGML

[Chet Ensign <Chet_Ensign%LDS -at- NOTES -dot- WORLDCOM -dot- COM>]

Mark Levinson writes at length about the disadvantages of SGML.

"It takes away your control over how the text looks and how the text fits on
the page.  ... It can lead to unrealistic expectations of portability ...  And
my specific pet peeve:  If the importance of your headings is not really in
strict top-down order, there's no way you can let the reader know."

He also says that "It's sexy." Whew. And here I was thinking that I was working
in a really boring area.

Mark's comments are all on target -- for the particular system that he has had
experience with. I think it is critical to note that SGML is not a product, a
system or even a particular set of tags. It is a standard that can be used to
develop systems. The success or lack of in meeting the expectations and
requirements of the people who use the system is a direct function of how the
system developers did their jobs and how involved the users were in its design.
It is possible to design a system based on SGML that fix all these
disadvantages and also produce all the benefits that SGML offers -- mainly,
information that can live separate from the product that developed it.

There are some great database applications out there and some lousy ones as
well. We don't suggest that SQL should be thrown out just because of a few
lousy implementations. Same is true of SGML.

I've heard lots of stories by now about SGML systems that were built by the
systems people and then dropped on the writers. Receipe for failure, or at
least unhappy writers. Where the writers have been primary drivers in the
development of the SGML, they wouldn't let it go for all the world.

Paper is no longer the sole distribution channel for the information we
produce. And very few companies are going to invest in separate teams of
writers to basically rewrite the same information over and over for each
distribution channel. The focus is switching from the output -- paper, Web,
newswire, etc. -- to the information. That information touches a lot of people
and has a life that extends beyond our own departmental boundaires. It behooves
us all to begin thinking about our product that way, and take an active role in
defining that information as well as designing the processes that produce it.
The sooner we start addressing our product -- information, *not* paper -- as
an asset, the sooner we get some leverage in how that product is produced, mana
ged and distributed by the company.

Best,

/chet

Chet Ensign
Logical Design Solutions
571 Central Avenue                      http://www.lds.com
Murray Hill, NJ  07974                  censign -at- lds -dot- com [email]
908-771-9221, Ext. 152  [Phone]                  908-771-0430 [FAX]