RE: Scheme Programming Reference

Subject: RE: Scheme Programming Reference
From: "Frank A. Christoph" <christo@xxxxxxxxxxxxxxxxxx>
Date: Thu, 24 Jun 1999 08:34:20 +0900
> I find it enormously enlightening, when someone suggests a method of
> making dsssl more accessible, that there is a strong preference towards
> a more scholarly (if well done) computer-science oriented book.  The
> demonstrated lack of understanding of the *publishing* audience that
> dsssl is *supposed* to serve makes it quite clear why dsssl is now and
> will remain a guarded toy of elites.

This is an amusing indictment of DSSSL since it is exactly the same argument
being leveled against XSL by web designers who want everything served up on
a silver platter.

My opinion is that there is no free lunch. If you want your computer to do
sophisticated things for you, then you have two choices. First, you can pay
Microsoft a wad of money to try to figure out one-gazillionth of all the
different possibilities of what you might want to do, and then create a
fancy GUI to slap onto it. Second, you can become a programmer and think for
yourself. (Or convince yourself that you are free-thinking, until you
realize that you're using Microsoft Visual Studio...)

Sure, DSSSL is declarative in a sense, but that doesn't mean it is magical.
There are such things as declarative programming languages, after all, and
although they are certainly simpler and more elegant than conventional
imperative languages, that doesn't mean you can just shove one in front of
Joe Schmoe and expect him to turn into a software engineer.

DSSSL is a standard, not a marketing campaign. It's not supposed to be sexy.
A standard is supposed to be a common language for people who bother to
learn it and use it. Do people get up in arms because network protocols
aren't accessible? Or CORBA? If you do, then you have been spoiled by the
marketization of HTML that was concomittant with the swell of the Web's
popularity.

Standards aren't meant for end-users; they're meant for specialists. That's
why we have standards for programming languages, for industrial equipment
and for metric systems, but not for how to dry your hair, unzip your fly or
pick your nose. (The big deal about DSSSL, as I recall, was that the
side-effect freeness, etc. was supposed to make it easy to develop
incremental implementations, suitable for embedding in browsers, and
stylesheet GUIs.)

(BTW, IMO "Goedel, Escher, Bach:..." is popular, not scholarly.)

> You are not trying to inform analysts, you want to inform writers.  This
> seems actually droll, to even consider that any author would be willing
> to consider such a book.  Have you known any *non-technical* authors?

Actually, I doubt very many non-technical authors use SGML anyway.

--FC


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist


Current Thread