ntg-context - mailing list for ConTeXt users
 help / color / mirror / Atom feed
From: Hans Hagen <pragma@wxs.nl>
Subject: Re: Context against XSL
Date: Thu, 30 Sep 2004 21:06:31 +0200	[thread overview]
Message-ID: <415C5937.5050900@wxs.nl> (raw)
In-Reply-To: <415C1B8B.3030709@hotmail.com>

Dirar Bougatef wrote:

> Does anyone have informations about tex macro packages and their 
> advantages over XSL (previously known as XSL-FO) ?
> 
> I have read somewhere that tex is a good implementation of the XSL 
> standard !
> 
> I think this is in regard that tex thinks in matter of boxes (Which is 
> the equivalent of XSL blocks). I this case, is the difference between 
> the two in the fact that at the end tex and macros are only algorithms 
> for typesetting blocks automatically ?

xsl is mostly a specification, and there are program soutthere that 
implement parts of is. The page model that xsl uses is not that 
advanced. Also, because you more or less make up the page, you also sort 
of disable all kind of clever things that batch processors like tex + 
macropackages may do. This means that xsl (fo) is suited for a certain 
range of typesetting tasks. From my experience your expectations should 
not be that high with regards to complex layouts.

I'm on and off implementing an fo engine (foxet) and run into fuzziness 
with regards to the specs (a bad omen is that that there i could not 
find a good manual and the ones i have are made up rather poorly, which 
indicated that we're not so much dealing with high end typesetting, but 
with regular batchprocessing of not too complex documents).

Recently i've been playing with css (from which xsl inherits much, which 
does not add to a clear design imo) and i'm surprised that browsers are 
so different that one ends up hacking around as much as one would using 
tex -) In many ways xsl is driven by the web, and not by real 
typesetting (is my guess). paper and screen are different things.

What you use depends on what you need it for. For a long time, the 
midset of designers has been shaped by what page maker, quark, etc can 
and cannot do (therefore all those ragged right docs, where the 
limitations have become the standard). I fear that in the next couple of 
years the limited possibilities of for instance xsl will bring down the 
standards (if it can't be done, one will just lower the demands), which 
also fits in the short lifecycle of most documents.

So, what to use when:

- here i find that using tex directly (using the context xml parser) in 
most cases is rather efficient; the problem is always in getting 
(frequently inconsistent) designs done. In that respect my motto has 
become 'the problem does not change'

- xslt is nice for preprocessing and manipulating documents and often 
one can get away with clean coding

- some scripting is often needed as well (for instance in order to add 
typographical detail, which is rather easy to do with regexps in 
scripting languages)

- xsl (fo), well for the moment i see it as a kind of 'placed xml'; when 
customers want us to use it, we'll do it (gives a feeling of 
independence), but in most cases using some direct mapping onto tex is 
not only easier (cheaper) but also gives a bit more control. It all 
depends on the design.

- so: just use the best of all worlds (which is what xml is about: it's 
consistent -when used all right- and it can be transformed; 
interestingly there are quite some organizations out there that bind 
themselves to just one kind of xml handling app thereby contradicting 
the concept.

In de time i want to write down something on these matters.

Hans

btw, there is a special mailing list for foxet; a preliminary version is 
in the alpha zip


-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
      tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com
                                              | www.pragma-pod.nl
-----------------------------------------------------------------

  reply	other threads:[~2004-09-30 19:06 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-09-30 14:43 Dirar Bougatef
2004-09-30 19:06 ` Hans Hagen [this message]
2004-09-30 20:53   ` Dirar Bougatef
2004-09-30 22:35     ` Hans Hagen
2004-10-01  6:52   ` Taco Hoekwater
2004-10-01  9:25     ` Hans Hagen
2004-10-01 10:53       ` Nikolai Weibull
2004-10-01 17:18         ` Matt Gushee

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=415C5937.5050900@wxs.nl \
    --to=pragma@wxs.nl \
    --cc=ntg-context@ntg.nl \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).