Re: F2F5 Agenda

On Tue, Feb 27, 2007, Dave Reynolds <der@hplb.hpl.hp.com> said:

> 
> Dave Reynolds wrote:
>> 
>> I like some of the better names but I don't quite see how this version 
>> of the sort sub-diagram solves the issues. Though that might just be the 
>> lateness of the hour :-)
It is not a sub-diagram. I don't see your point here. Any term is types by
some sort. What's wrong in that?
>> 
>> First, Constants include things like numbers and dates, so they can't 
>> uniformly have an xs:anyURI id. You could split Constants into Symbols 
>> and Literals (or TypedValues or something) - Symbols could have an id 
>> (xs:anyURI) whereas TypedValues would have a lexicalForm (an xs:string).

We have all this things already in R2ML but I just put a question to the
participants if we consider to sub-class constants.

>> 
>> Second, just considering Symbols I don't see how this resolves the 
>> divergence between the metamodel reading and the abstract syntax 
>> reading. Your BNF suggests that you intend this to have an abstract 
>> syntax reading but I don't think we are saying that a Symbol used in a 
>> predicate position (for example) would have an ASort specification 
>> inline in the syntax at the point of use (at least I hope not).

Here you have partially right. Probably you we have to use a reference to
an already specified sort. But in this case RIF Core needs a specified
vocabulary too. I don't know if this is agreed by all participants.

> 
> Sigh, of course that should have been either function/Asort or 
> predicate/Bsort ... definitely too late ...

A uniterm can be sorted by any sort. If it is a predicate or function it
depends of the context in which it is used.
> 
> Dave
> 
I hope this clarify a little bit,
Adrian
>> 
>> Dave
>> 
>> Adrian GIURCA wrote:
>>> Dear F2F participants,
>>> Find below our proposal for the MOF metamodel of the RIF Core. This 
>>> proposal considers all concepts from the RIF Core Positive Language. 
>>> We try to find good names which are proposed by using /better-name/ 
>>> property. After the MOF/UML class diagram you'll find a proposal for a 
>>> mapping from MOF to (E)BNF. The EBNF conforms to ISO/IEC 14977:1996 
>>> (similar with OWL abstract syntax).
>>>
>>>
>>> Mapping rules:
>>>
>>>    1. Any abstract class (italic name in the diagram) translates into
>>>       EBNF non-terminal using capitals (ex. /Condition /translates into
>>>       CONDITION)
>>>    2. Any concrete class translates into EBNF terminal (ex. And
>>>       translates into 'And')
>>>    3. Any composition is captured using round brackets as terminals
>>>    4. As in OWL abstract syntax using curly brackets { and } denotes 0..*
>>>    5. Right brackets [ ] are used to indicate 0..1
>>>
>>> According with these rules we obtain the the following EBNF:
>>>
>>> CONDITION ::= 'And' '(' {CONDITION} ')' |
>>>
>>>               'Or' '(' {CONDITION} ')'  |
>>>
>>>               'Exists' '(' 'declare('Var {Var}')' CONDITION ')' |
>>>
>>>               POSITIVE
>>>
>>> POSITIVE ::= 'Equal' '(' TERM TERM ')' | 'Uniterm' '(' 'functor' '(' 
>>> Const ')' 'arguments' '(' {TERM} ')' ')'
>>>
>>> TERM ::= Var | Const | 'Uniterm' '(' 'functor' '(' Const ')' 
>>> 'arguments' '(' {TERM} ')' ')'
>>>
>>> Var ::= 'Var' '(' name ['type' '(' SORT ')'] ')'
>>>
>>> Const ::= 'Const' '(' id ['type' '(' SORT ')'] ')'
>>>
>>> SORT ::= PSORT | 'ASort' '(' SORT {SORT} ')' | BSort
>>>
>>> PSORT ::= xs:integer | xs:decimal | xs:time | xs:dateTime | xs:string
>>>
>>> name::= NCNameRef
>>>
>>> id ::= URIref
>>>
>>> You may notice our proposal for using xs:NCName for local names and 
>>> xs:anyURI for constants.
>>> Also there is a question which arise:
>>> Why the symbol set Const is not subclassed by (possibly overlapping) 
>>> subclasses IndividualName, FunctionSymbol and PredicateSymbol?
>>>
>>> Looking forward to your comments,
>>>
>>> Adrian
>>>
>>> Chris Welty wrote:
>>>>
>>>>
>>>> We had to do a little reorganization of the agenda to deal with the 
>>>> problems created by the weather here, and the new agenda for the f2f 
>>>> is now on the wiki.  Of course things may change further as we 
>>>> uncover new issues.  If someone who plans to attend the meeting 
>>>> remotely is planning on attending for a specific session they should 
>>>> let us know.
>>>>
>>>> -CC&S
>>>>
>>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>> 
>> 
>> 
> 
> 
> 
> 

--

Received on Tuesday, 27 February 2007 13:36:52 UTC