OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

office message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [office] ISO 14977 EBNF grammar

Patrick Durusau <patrick@durusau.net> wrote on 04/29/2008 03:04:49 PM:

> Sure, I don't deny that you can write really useful grammars using the
> W3C style and lots of people have done it. That doesn't mean that avoids
> the rule in ISO that we should use ISO standards without compelling
> justification to the contrary.

I think we're talking different levels.  No one is talking about using any EBNF in ODF document instances.  All we're talking about is how we formally describe the syntax of formulas in the text of the specification.  This is the use of standards at a different level than when we talk about using ISO 8601 to represent dates, etc.  And there is no decision we can make here that has any influence on conformance or implementations.

To follow your logic, OOXML would have not been allowed, since its definitive schema was in W3C XML Schema format rather than ISO's Relax NG.

But I agree that ISO might have sensibly required a specific BNF format for defining syntax, but if they did I'd expect to read it in ISO Directives, Part 2.  But I don't see that.  So we should be able to describe our grammar any way we want, so long as we can describe or reference it precisely.  This could include defining our own syntax language, referring to the W3C's conventions, to IETF's or to ISO 14977.  

Remember, the W3C is an Approved Referenced Specification Originator Organization (ARO), so a reference to a W3C Recommendation is essentially pre-approved.


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]