Re: lambda was RE: W3C-transformation language petition

Didier Martin wrote (in two separate messages)

> You may end up with something even more un-lisible than DSSSL. After all, a
> dsssl document is also a SGML document where the begin tag delimiter is "("
> and the end tag delimiter is ")". The problem is not so much with dsssl
> interpreters but with tools to help not get lost with the "()".

> So if people get sick with "()" and like to write a lot, then we can replace
> the delimiters and omitags in order to make dsssl xml compliant :-)

A dsssl stylesheet is an SGML document but the SGML declaration has not
been changed to make (foo   ) the SGML element foo with a non standard
tag syntax, and omitted name in the end tag.

DSSSL uses the standard <  > / & ; syntax for SGML elements.

The SGML parse happens essntially before the content of the
style-specification-body gets passed (as a NOTATION) to the scheme
parser of dsssl. So you can use SGMl comments (and entities) in ways
that cut across the normal list syntax, as it is the _result_ of that
parse that is passed to the scheme parser. (Conceptually.
Implementations may I suppose interleave these things, I don't know).


So while it is true that one could of course define a language
that was 100% functionally equivalent to dsssl but used <> syntax,
it does not follow from the fact that DSSSL is an SGML document does it?
In particular the change would involve recoding the dsssl parser, not
simply a matter of changing the SGML declaration to change the start tag
syntax from <foo> to (foo).

> Your proposal is not so crazy. If people get sick because of parenthesis and
> because Dsssl is also a sgml document therefore we can change the begin tag
> and end tag like in the example below:

David


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list