RE: [xml-dev] Results of Open XML balloting at INCITS

["Rick Jelliffe" <rjelliffe@allette.com.au>]

Michael Kay said:

> The point is that fast-track is a process for getting a standard that is
> already published by a member body (in this case ECMA) ratified quickly
> when
> there is a high level of consensus. If there is opposition, then it's an
> inappropriate process - regardless of the technical merits or the reasons
> for the opposition.

I don't think that is a presumption you can make. The things that make a
ISO standard are first a market requirement, second an adequate text, and
third enough national interest to get an agreement. All fast-tracking does
is bypass the Committee Draft stages, because it has been developed to
draft stage by some external group.

> For one person's summary of other national body responses, see
>
> http://consortiuminfo.org/standardsblog/article.php?story=2007022819130536

That is about responses to the contradictions period, which is a different
stage in the process. ISO basically said "No", or rather "None of what you
say is a showstopper that requires extraordinary intervention, these
issues can go on and be addressed by the normal process": the current
ballot produces the big issue list to get resolved.

> Presumably a country votes yes if it believes that the existence of the
> standard is in that country's economic interests. I would think this is
> only
> distantly related to the technical quality of the specification.

A national body certainly may get more people interested if there is some
strong economic interest, but many times nations will vote yes because
they can see no reason to abstain: if some group of nations thinks they
need a spec, other nations don't stop them just because they are not
enthusiastic. That goes to the basic "good faith" requirements of
standards work: you don't diss the other bloke's needs.

The other thing to realize is that no means yes. When a nation gives a no
vote, they have to give the technical reasons why not and suggest their
preferred fixes. The system is rigged towards keeping different parties
talking, not pressing the eject button. So "we don't like ODF" is not a
technical reason. And "our people don't need a standard for this kind of
fine grained legacy access" is not a technical reason. Having a "No with
comments" vote with good meaty comments is actually a really good thing,
for getting a good standard.

> In practice, of course, many countries will vote based on the opinions of
> a
> small number of individuals, who will not always take a purely objective
> approach to the decision...

The issue of what is good objective evidence is a really good one. I would
say that the best objective evidence of the technical quality of a
standard is whether people can successfully implement it. Open XML seems
to have no problems there. Apple's Numbers, Gnumeric, and so on.

Cheers
Rick Jelliffe