Re: indentation (was Re: About the article)

[Brandon Ibach <bibach@xxxxxxxxxxxxxx>]

Quoting Didier PH Martin <martind@xxxxxxxxxxxxx>:
> <Comment>
>    Well, I could go off for a while here, but back to the original
> topic.  My point in this is that you may not want to go too far to
> draw someone in to DSSSL if they wouldn't be willing to invest the
> time to really learn how to use it properly.  I guess I wonder if
> we're looking for more users in terms of those who use DSSSL by using
> Norm's Docbook style sheets, or terms of those who use DSSSL by
> writing their own style sheets.
> </comment>
> 
> <reply>
> Restating what you say in other words: Let's limit the DSSSL users to the
> Docbook DTD users only.
> 
   I don't want to jump in to this again, but I do want to clarify a
couple of things.  I did *not*, in any way, mean to say that DSSSL
should only be used for Docbook.  The contrast I was attempting to
make in the above comment was between a user who is primarily just
running pre-packaged DSSSL scripts to get their output, and *may* even
go so far as to do some minor customization, as opposed to a user who
is writing their own stylesheets.  Norm's Docbook style sheets just
happened to be a convenient example.
   In addition, this question was truly meant as a question.  Are we
looking for more people to just use DSSSL, or are we looking for more
people to write DSSSL scripts?
   Lastly, I still say there is validity to the idea of layering XSL
and such on top of DSSSL.  As Mark (I believe it was) pointed out, XSL
is probably going to go way beyond what DSSSL ever will, because you
will have so many web guys who don't think of themselves as
programmers, don't want to use something that requires them to
program, and don't think you can use DSSSL without programming.  But
if we can give them their supposedly so-much-simpler syntax on top of
a real DSSSL engine, we may be able to lose the battle, but win the
war.
   Let me give an example of why I think this layered approach is
worth it.  Configuration/administration of a Unix system can be a bear
for someone just getting started (ie, your average Windows user who
has never seen a Unix prompt that has just bought and installed a
Linux CD).  Thus, you have a number of graphical administration
utilities such a linuxconf, Caldera's COAS, Sun's tool (whatever
that's called), etc.  But do these tools actually replace the
underlying mechanisms that are used for configuration, such as shell
scripts for the basic startup of a system, and the numerous
configuration files of various formats for different tools.  No!  They
provide simple abstractions for these mechanisms and write out the
appropriate stuff for what the user wants.  So, when you bump into
something that the tool doesn't address (like some very obscure
setting that doesn't get used very often), you can always dig into the
configuration file and tweak it by hand (or get someone who knows how
to do it for you).
   Look at Jade itself.  Theoretically, James could have just gotten a
basic Scheme compiler/interpreter going, then built all of the DSSSL
specific stuff in Scheme, but instead, he did a lot of it in C++ (for
efficiency, I'm guessing).
   Okay... enough rambling...

-Brandon :)


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