Re: Are we losing out because of grammars?

[Charles Reitzel <creitzel@mediaone.net>]

Sure, fine.  Is it streamable?  If not, it ain't a basic requirement.  Once
a basic layer reaches recommendation status, you can do rules till the cows
come home.  Until then, time's a wastin'.

It could probably be done in fewer lines of Java code.  Schematron is
competing with Java, C/C++, Prolog, RDBMS stored procedures, expert systems,
etc., etc.  All such rule implementation languages, including Schematron,
need access to a stable, predictable, static data definition language.  The
basic layer should *not* impose DOM-like requirements on all layers over it.
From this common starting point, interesting things can proceed in the
dynamic space.

take it easy,
Charles Reitzel


At 08:51 AM 2/5/01 +0100, you wrote:
>> >Or what about "the twelfth <month> in a <year> has 31 <days>"?  Is that a
>> >schema requirement? That can be expressed in some grammar languages but not
>> >others.
>> 
>> Not a basic requirement.  I don't know of any simple grammar that would
>> express that easily.  Rules+DOM are probably needed.  I put this at layer 3.
>> 
>
>A case for Schematron (an extension taking care of February 29 would be
>easy to add):
>
><pattern name = "Year months">
>	<rule context="year">
>               <report test = "count(month[1]/day)=31">January
>OK.</report>
>               <report test = "count(month[2]/day)=28">February
>OK.</report>
>               ...
>               ...
>               ...
>          </rule>
>     </pattern>