Re: idea for developers

[Valeriy E. Ushakov]

On Sun, Mar 25, 2001 at 11:31:53 +0400, Dmitry Kirsanov wrote:

> Imagine an XML-based format (let's call it LoutML) ...

Check list archives (see http://www.ptc.spbu.ru/~uwe/lout/list.html).
This topic was discussed (to death, I might add) just recently.


> This will promote Lout to a huge XML user community

XML user community does not need Lout (or TeX, or troff), it needs a
printing solution.  If Lout could be a part of this priniting solution
is entirely different matter and for the *user* community is an
irrelevant implementation detail.  E.g. I don't care a hoot if gcc on
solaris uses gnu binutils or sun's binutils, or if my email is
delivered by sendmail or qmail, or if your web pages are served to me
by apache or some other web server.

Of course for implementors the potential choice of Lout is an
important implementation decision, but then for implementors the lack
of XML syntax for Lout is a non issue.


> Ideally, I would love to see the next version of Lout support the
> XML-based input format directly.

This wouldn't buy *anything*.  For implementors seriously interested
in using Lout as part of XML printing solution the syntax of Lout is
not an obstacle.  And no, this is not "play by our rules or go away" -
this is a simple engineering fact.  Also, as far as engineering is
concerned, I would say that it's much more *practical* to target
existing Lout syntax than to bolt on an XML fronend to Lout.

In the previous incarnation of this debate I discussed at some length
that Lout syntax was designed with specific uses in mind and that
providing an XML mapping for it is not that simple (HIE-wise) as one
might think.

So, please, check archvies.

SY, Uwe
-- 
address@hidden                         |       Zu Grunde kommen
http://www.ptc.spbu.ru/~uwe/            |       Ist zu Grunde gehen