This is a lovely example of how quoting out of context can replace giving a
counter-argument. Here's the _substantive_ part of Liam's note:
>Whitespace treatment needs to be specified in the CML specification,
>for example, and then any conforming CML processor will do the right
>thing and there's no problem. Taking CML and passing it to a CDF processor
>will result in different whitespace treatment, I expect... and also different
>treatment of all the non-whitespace too! And that's fine.
The point is (and I also made this at length before) there are few ways to
meaningfully process markup without knowing the DTD or having a stylesheet,
there are even fewer ways to process such markup (w/out stylsheet or
knowledge of the DTD) such that ignoring whitespace is something that you
_need_ to do. XML already provides you with _all_ the whitespace (unlike
original flavor SGML) -- so there's no problem with the parser hiding
significant whitespace. The only question is whether we should be addding
features to note that some whitespace is _insignficant_. In fact I believe
that small set of ways to process markup (that you don't know the meaning
of, without a processing spec), and where you _have to_ collapse or
otherwise mangle the whitespace is the _null set_.
As I asked before, I'd like to see even one example of of an application
that needs this.
-- David
_________________________________________
David Durand dgd@cs.bu.edu \ david@dynamicDiagrams.com
Boston University Computer Science \ Sr. Analyst
http://www.cs.bu.edu/students/grads/dgd/ \ Dynamic Diagrams
--------------------------------------------\ http://dynamicDiagrams.com/
MAPA: mapping for the WWW \__________________________