Re: args tag in RIF XML serialization

Harold,

Writing down our reply to Thomas Krekeler's question, below, I checked 
again what FLD and BLD said.

You pointed out that BLD said that: "For the subscript n it is understood 
that n?0, i.e. the ellipsis indicates zero or more terms" [1]. But nothing 
says (or I could not find it) where it says that the empty args element 
must be omitted (and, no, it is not abvious; for instance, I had to check 
for the items sub-element in lists... and I found out that PRD required 
the element even if empty, whereas BLD and Core did not [2] => we need to 
issue an erratum on PRD).

Furthermore, FLD does not even seem to mention the difference between n 
and m subscripts (i.e. that n>=0 and m>=1) => we probably need to correct 
that in BLD and FLD too (not sure if this is an erratum properly said, 
though).

Errata aside, do you (and everybody) agree if I tell TK that the empty 
args element, like any other empty role element, must be omitted in 
RIF/XML and that this will be made explicit in the next release of the 
specs? Or are there cases where an empty role element is valid RIF/XML?

Cheers,

Christian

[1] 
http://www.w3.org/TR/rif-bld/#Mapping_from_the_Presentation_Syntax_to_the_XML_Syntax

[2]  I traced back the difference to the an ambiguous resolution that we 
passed on May 4, 2010, and that was implemented differently in PRD and 
Core/BLD/FLD: I implemented the decision litterally, copying the schema 
fragment in the reference email [3], whereas Harold implemented it in a 
way that made the items element optional.

[3] http://lists.w3.org/Archives/Public/public-rif-wg/2010Apr/0032.html

IBM
9 rue de Verdun
94253 - Gentilly cedex - FRANCE
Tel./Fax: +33 1 49 08 29 81




From:
Thomas Krekeler <krekeler@ontoprise.de>
To:
public-rif-dev@w3.org
Date:
22/09/2010 16:03
Subject:
args tag in RIF XML serialization
Sent by:
public-rif-dev-request@w3.org



Hello,

which of the following (partial) XML code is valid RIF XML?

1) No args tag:

<Atom>
   <op><Const type="&rif;iri">http://test.org/P</Const></op>
</Atom>


2) Empty args tag:

<Atom>
   <op><Const type="&rif;iri">http://test.org/P</Const></op>
   <args ordered="yes"/>
</Atom>


In [1] there is no question mark characterizing the args tag as optional,
so I would go for 2 and view 1 as invalid.

By reading the XML schema ([1]) on the other hand I think 1 is valid but 2 
is not.

Thanks for your help.

[1] 
http://www.w3.org/TR/rif-fld/#Mapping_of_the_Non-annotated_RIF-FLD_Language

[2] http://www.w3.org/TR/rif-fld/#Baseline_Schema_Module

Yours,
Thomas

- - -
Thomas Krekeler
Development
ontoprise GmbH ? know how to use Know-how
- - -

General Electric selects SemanticGuide from ontoprise as a corporate 
standard for providing intelligent advisory systems for Jenbacher

http://www.ontoprise.de/en/home/news/news-en/general-electric-uses-semanticguide/


- - -
An der RaumFabrik 29; 76227 Karlsruhe; Germany
Tel.: +49 721 509809-51; Fax: +49 721 509809-11
mailto:krekeler@ontoprise.de, www: http://www.ontoprise.com
Registered Office: Karlsruhe, Germany, HRB 109540

Managing Directors: Prof. Dr. Jürgen Angele, Hans-Peter Schnurr
- - -





Sauf indication contraire ci-dessus:/ Unless stated otherwise above:
Compagnie IBM France
Siege Social : 17 avenue de l'Europe, 92275 Bois-Colombes Cedex
RCS Nanterre 552 118 465
Forme Sociale : S.A.S.
Capital Social : 612.509.964 ?
SIREN/SIRET : 552 118 465 03644