Jonathan

Thanks for citing that example. I have added a summary to the minutes of yesterday's WG call.

Please go ahead and trawl the document for implementation defined/dependent things.

Please also raise a public comment to track, at http://redmine.ogf.org/projects/editor-pubcom/boards/15.

Regards

Steve Hanson
Architect, IBM Data Format Description Language (DFDL)
Co-Chair,
OGF DFDL Working Group
IBM SWG, Hursley, UK

smh@uk.ibm.com
tel:+44-1962-815848




From:        "Cranford, Jonathan W." <jcranford@mitre.org>
To:        "dfdl-wg@ogf.org" <dfdl-wg@ogf.org>,
Date:        17/09/2013 16:11
Subject:        [DFDL-WG] action 224: add section for implementation defined limits
Sent by:        dfdl-wg-bounces@ogf.org




All,
 
Action item 224 was raised two weeks ago during the WG call.  
 
224 Add section for implementation defined limits (All)
3/9: Several places in the spec cite this, should be grouped. Currently partially listed in section
2.6.
Also note distinction between 'implementation defined' and 'implementation dependent'. Check
spec for correct usage.
Resolve during public comment.
 
The action item was created based on a comment I made during the call, so I thought it’d be good to provide an example of the distinction I was trying to make.
 
The W3C XProc specification does a great job of differentiating between implementation-defined and implementation-dependent features, with a convenient list of each in the appendix.
 
Appendix A (http://www.w3.org/TR/xproc/#conformance) contains the following text.
Conformant processors must implement all of the features described in this specification except those that are explicitly identified as optional.
Some aspects of processor behavior are not completely specified; those features are either implementation-dependent or implementation-defined.
[Definition: An implementation-dependent feature is one where the implementation has discretion in how it is performed. Implementations are not required to document or explain how implementation-dependent features are performed.]
[Definition: An implementation-defined feature is one where the implementation has discretion in how it is performed. Conformant implementations must document how implementation-defined features are performed.]
 
Section A.1 then lists all the implementation-defined features, and section A.2 lists all the implementation-dependent features.
 
I think the XProc spec provides a great example to follow on two counts.  First, it formally distinguishes between implementation-defined and implementation-dependent features.  The choice of terms isn’t nearly as important as the distinction itself, of course:  implementations must document how certain features are implemented.  In the DFDL realm, section 2.6 lists some implementation limits which always constitute schema definition errors; surely these are the types of details that must be documented by any DFDL implementation.  Using terminology such as “implementation-defined” and “implementation-dependent” would flag these types of documentation requirements for implementations within the specification.
 
Second, all the implementation-defined and implementation-dependent features are listed in one place in the specification.  I think doing the same in the DFDL spec would provide a great resource for DFDL implementers.
 
Comments?  If everyone agrees, I don’t mind taking the action to search through the document looking for candidates for inclusion in such a list.
 
Sincerely,
 
--
Jonathan W. Cranford
Senior Information Systems Engineer
The MITRE Corporation (http://www.mitre.org)
 --
 dfdl-wg mailing list
 dfdl-wg@ogf.org
 
https://www.ogf.org/mailman/listinfo/dfdl-wg

Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU