On Sun, Oct 07, 2001 at 02:46:35PM +0200, Daniel Egger <degger@xxxxxxx> wrote:
> > really???
> I've heard there are Perl hacks as well. :)
There are hacks for a lot of other languages/environments ;) The
shortcoming of gettetx lies not int he parsing and input format...
> > which would be easy, nice and probably very small.
> Yes, but not very versatile...
Why? It contains the tips and a minimum amoutn of clutter. If you equate
evrsatile == xml because everybody claims to support it I disagree
completely.
> Also agreed, the disadvantage with the headers is that the messages
> are static after compilation while it's quite easy to extend XML and
> test it without having to recompile the application.
Yeah, but a) nobody does that (right)? and b) this could be said for ltos of
other things. Just as SGML was so feature-rich that nobody knew or used all
its features we might not need to make everything run-time-configurable.
Compare this to the trend of adding 10k of module loading and interfacing
code to about each and every program nowadays where it doesn't make sense
(pluggable protocol modules for lftp? get real... ;)
> The problem with SGML is that it's too complex to grok completely and
> that's why a large amount of people simply ignored it; XML is a subset
That's a story I never heard of before. The reson SGML was not used is
because it was very powerful and complex to implement. For humans it is
easy to grok. You are comparing sgml with xml-applications. I could just
claim that most sgml-applications are much easier to grok for humans than
xml namespaces or schemas.
> of SGML which was defined with ease-of-use in mind and that makes it
Where do you get the idea that XML was done for ease-of-use? And why do you
apply ease-of-use to humans, while XML was designed to be easy-to-use in
applications (as opposed to SGML). one of the goals of xml is:
XML documents should be human-legible and reasonably clear.
this doesn't sound too encouraging, no? yes, it was a goal to keep xml
"human", but this is a minor goal, others (ease-of-use for machines!) were
more important.
Anyways, it's getting off-topic and I'll keep it there ;)
> Exactly. If we decide to use XML, we should do so consistently. Also we
> can neglect the overhead in this case; in fact I believe that using
> GMarkup is bloatfree or even better if we throw out the configparser for
> instance.
Fully agreed. Just somebody needed to code it ;) Ha, not me, not me, I am
a lazy lamer ;)
--
-----==- |
----==-- _ |
---==---(_)__ __ ____ __ Marc Lehmann +--
--==---/ / _ \/ // /\ \/ / pcg@xxxxxxxx |e|
-=====/_/_//_/\_,_/ /_/\_\ XX11-RIPE --+
The choice of a GNU generation |
|
