Re: Jade and display-space

[lex@xxxxxxxxxxxxxx (Alex Milowski)]

> Alex Milowski wrote:
> > 
> > I don't see that Jade supports the display-space type.  Am I wrong here?
> 
> Yes.

Ah ha!  Good!  I was hoping to be wrong! ;-)

> > For example, the space-before property is suppose to have a value of
> > type display-space.  The current stylesheets for Jade use a quantity.
> > 
> > Display spaces have different semantics than quantities.
> > 
> > This is cause us problems in our DSSSL engine.
> > 
> > What's the official word on this?
> 
> >From the proposed TC (http://www.ornl.gov/sgml/wg8/document/1883.htm):
> 
> > Defects in Clause 12: Style language
>  
> > 12.5.4.1 
> 
> >      A length-spec should be allowed for any characteristic with a display-space type (just as a length is allowed
> >      for any characteristic with a length-spec type).

Another *very important* change in the TC.  Now, the parenthetical remark 
doesn't make much sense.  Should that read "just as a display-space is allowed
for any characteristic with a length-spec type"?  The sentence associates
an ability to exchange length-spec with display-space.  If the parenthetical
remark intends to associate a previously describe reversal of this--that is,
that display-spaces are allowed for length-specs--it should state so
more clearly and refer to the specification of such a ability.  

In any event, the parenthetical remark is not clear.  The information before
it is very clear.
 
> 
> This is what Jade implements.
> 

Another nit-pick, but, this document is a proposed TC for DSSSL.  Doesn't
this mean that Jade, Seng, and others who adopt this TC are out of sync
with the official standard.  If the TC changes, we are then conformant to
something that is *not* DSSSL and not the revised DSSSL.

For compatibility sake, how "hard-line" should all DSSSL implementors be?

For example, I would like the ability for stylesheets to be exchanged
between Jade and Seng.  Unfortunately, there are already application
extensions that are not support in Seng that exist in Jade.  Although
someone could add these to Seng, they do not exist at the current time.

Hence, given a stylesheet that does not take advantage on processor
specific external procedures or application flow objects, how do we
ensure that between "feature sets" and the TC that our environments can
inter-operate?

I suggest that we need a common core specification--maybe dsssl-o?

==============================================================================
R. Alexander Milowski     http://www.copsol.com/   alex@xxxxxxxxxx
Copernican Solutions Incorporated                  (612) 379 - 3608

 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist