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] Another view on conformance?

Dave Pawson <dave.pawson@gmail.com> wrote on 02/28/2009 12:00:52 PM:

> Re: [office] Another view on conformance?
> 2009/2/28  <robert_weir@us.ibm.com>:
> >> What are you afraid of?
> >>
> >
> > It isn't a question of fear.  It is a question of mathematics.  If you
> > have a way test whether given number is random, using Relax NG, XML
> > Schema,or Schematron, then I'd love to hear about it.
> If the spec was firm, the test would be easy.
> In this case, ask ODF TC to define random;
> then define the test wrt that?
> Exactly as Rick suggests.
> Logical sequence rob?
> E.g. random = changing,  0 < n < 1,  different n times over m?
> Test? Easy.
> Weak definition, weak test. Your choice.

The point is not the definition of random.  The point is that a single 
value of RAND()is neither random nor not random.  Randomness is a property 
of a sequence of values.  Automated validation techniques in the markup 
world validate XML instances, not values returned by sequences of calls to 
methods declared within the markup.  This is the disconnect we get when we 
have a bunch of markup jockeys trying to evaluate something that is 
essentially a programming language.  We need to appreciate that the 
specification and validation of programming languages constructs are 
different beasts than the markup, and that different techniques and 
criteria apply.

It makes me wonder whether OpenFormula would be better referred to 
JTC1/SC22 than JTC1/SC34 as the committee with the closest expertise in 
this area.


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