Hi michael

I think that the meta-model of the format is not really important as 
soon as meta-model of the contents is coherent. For example the format 
of VW is not really good because you have to parse in the XML to get 
information. For extracting the method name you have to parse.

For me I would have no problem to have:

MethodDefinition
	inClass: #MyFooClass;
	inPackage: #TheFooPackage;
	methodName: #foo:bar:;
	arguments: #(one two);
	fullName: 'foo: one bar: two';
	mainComment: "jkljljlkjl";
	body: 'jlkjlkjkj lkjkl j lkj ljkl ';
	timestamp: '10/02/2003';
	version: '1.02';
	author: '....'

So this is not XML and does not require a special parser, we can read 
it and understand it.
For me the key aspect is that we would have a declarative exchange 
format of Smalltalk code
we could load that expression, manipulate it, store it, analyze it.

I think that XML is just about the interexchange format while the 
context in the current
meta-model of the contents with changeset contains holes 
(globalVariable Definition).

I think that what is important is that the meta-modele of the contents 
should be extensible.
So that we can add new properties.
	
So I'm even not against XML :) except that you need another parser and 
that after people may think to be smart to do checking based on DTD and 
when you do that you are basically done with extensibility.

So I like the way Ginsu or Monticello did it. No parser extension still 
a good meta-model of the source code.

Stef


On Monday, July 28, 2003, at 06:38 PM, Michael Rueger wrote:

> Daniel Vainsencher wrote:
>
>>> Actually we are. We need a way to attach attributes/semantic 
>>> information to the source code. Although it is possible with the 
>>> current chunk format it definitely isn't the best way to do it.
>>> All this of course is MHO.
>>> Well, maybe NSHO ;-)
>> What do you think IS the best way to do it?
>
> Maybe not "the best", but one choice would be
>
> <taking cover behind a 42" flame proof wall>
> XML
> </cover>
>
> :-)
>
> Michael
>
>
>