🚀 go-pugleaf

In <1993Jan14.194453.20233@informix.com> Robert Hartman writes:
...
>Perhaps, but I'll settle for a flat ASCII representation with standardized
>tags, and a strict separation of formatting information from content.
>It's simpler.  It will do.  But even with this simple architecture, look
>how long and difficult the process has been to deveop a standard!
>
>There are hordes of utilities of all kinds that already accept and
>operate on files with the proposed SGML text format.  What you may gain
>in response-time using a binary format, you lose in interoperability on
>the fly.  Binary representations are by no means an unqualified win.

I wanted to register my agreement with Mr. Hartman.  Viewing the use
of SGML as *only* an interchange form between authoring systems
that "hide the tags" is a shortsighted view.  My own view is that
high-function editing of SGML, compared to other tasks that having
a standardized, human-readable language for structuring text enables,
offers the smallest total benefit in author productivity, viewed over
the total set of tasks and problems faced.  Not to
say that authoring isn't important, but that it isn't the only thing,
and that having a nice editor does not solve the most costly problems
in large enterprise information development, such as supporting
multiple output forms, data interchange on a wide scale, standardization
of presentation styles, preservation of investment in data, and
so on.

Eliot Kimber                      Internet:  drmacro@ralvm13.vnet.ibm.com
Dept E14/B500                     IBMMAIL:   USIB2DK9@IBMMAIL
Network Programs Information Development     Phone: 1-919-543-7091
IBM Corporation
Research Triangle Park, NC 27709