RE: Nostradamus (was Re: FO. lists as tables)

> -----Original Message-----
> From: pandeng@xxxxxxxxxxxx [mailto:pandeng@xxxxxxxxxxxx]
> Sent: Thursday, October 14, 1999 1:36 PM
> To: xsl-list@xxxxxxxxxxxxxxxx
> Subject: Re: Nostradamus (was Re: FO. lists as tables)
> 
> 
> On Thu, 14 Oct 1999 12:20:12 -0400, you wrote:
> 
> >I think the XSL WG *is* tuned to the needs of typesetters.
> 
> I'm not so sure. From the XSL Requirements Summary (thanks to Tony
> Graham for reminding us of its existence; it's been a long time since
> I looked at it):
> 
> --------8<--------
> Predictability
> 
> Page fidelity is neither a requirement nor a goal. Presented with the
> same document and the same stylesheet, a given renderer should always
> produce the same results. Different renderers should produce similar
> results.
> --------8<--------
> 
> To me, it looks like HTML all over again. HTML is non-portable,

Well, sort of, but not really.  This basis of this is the entirely pragmatic
recognition that once you split the content structure notation (XML) and the
style constraints (XSL/CSS/etc), composition is pretty non-deterministic.
To get complete fidelity you'd almost have to specify the algorithms.  Take
TeX, for example.  It produces a certain set of line breaks for a given
text.  If I use a different algorithm I get different line breaks in all
liklihood.  Ditto for other breaks - para, page, etc.  I think it is
possible to design a language that would allow the page designer to
constrain composition at a very finely grained level (think of expressing
TeX's linebreaking in terms of "policy"), but it would be (is, actually; I'm
working on it) pretty difficult to do.  And even given such a language,
would it be worth the trouble for vendors to conform to such an exacting
degree when something well short of 100% fidelity will satisfy just about
everybody?  So the real design goal is to find the sweet spot, where
composition is sufficiently tightly constrained to ensure a relatively high
degree of similarity of outcome across implementations (i.e. this ain't no
html) but not so tightly as to frighten off implementors.

At least that's my take on it; I'm not speaking for the WG here.  If I had
my druthers we would have 100% fidelity; but consider what this means.  We
would need to be able to determine exactly what a given document+stylesheet
ought to look like based solely on the semantics of the language,
independent of any implementation.  Otherwise we end up with the situation
we have today, where the meaning of language semantics is almost always -
intentionally even! - based on implementations, and conformance is dictated
by whoever gets to market first with the biggest PR budget.  We would also
have to have levels of granularity, since the high-end and low-end segments
of the marketplace have different requirements in this respect, and
supporting such finely-grained constraints ain't cheap.  In any case, until
somebody comes up with such a language, 100% fidelity is not an option.
IMHO.

Sincerely,

Gregg Reynolds
standard disclaimers apply

P.S.  Forgot to mention, different "units of fidelity" have different
importance for different users.  Page-level fidelity is very important for
some users (e.g. in loose-leaf composition you really, I mean really, want
that nuclear warhead technician to receive accurately paginated revision
pages for his handy repair manual).  For most other users its not that
important; whether Madame Bovary gets snuffed on page 365 or 367 doesn't
matter much in the grand scheme of things.


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