Re: Tokyo document for review

[Martin Bryan <mtbryan@sgml.u-net.com>]

William

This one is obviously down to me as the guy who tried to do the
classification in 8.1 without the aid of an atlas!

> But I have to ask: wherever are you going with 7.3 Regional Context? By
> the appearance of the asterisk, it seems to be a preferred context
> category recommended by the ebXML ccWG. There are any number of uses for
> the ISO 3166 country code, but trying to anticipate how an application
> (or pair of trading partners) intend to use it may be a futile
> proposition.

No one is trying to define all the uses. What we are trying to do is to
define the commonly used ones as a "starter set". There was a long argument
in the group as to whether continents were relevant at all. The case we came
up for needing them was based on the fact that every country in the world
currently has quarantine regulations that apply specifically to anything
originating from Africa. Other than that the continents are probably not
necessary, but they do help to identify groupings.

> Looking at the appendix in 8.1 Regional Classifications, I appreciate
> the yeoman effort that went into geographically classifying every
> political unit, especially since, like the typical American
> high-schooler or President-presumptive George W. Bush, I wouldn't be
> able to place any more than 10% of the countries on a globe.  But I
> can't see any relevance to e-business by this rather arbitrary
> classification - and if it's of no use in business, why would we bother
> to carry this information around in a repository or use it to drive
> "context"?

The key factor is which economic groupings each country belongs to, and what
to do with countries that do not belong to one of these groupings. This is
why Europe is split into EU, EFTA and Eastern Europe (i.e. the rest). I
would have liked to do the same for America and Asia, but unfortunately I am
in the same league as Mr. Bush when it comes to identifying economic
groupings outside of my own economic block, Europe.

> And now you have me started: why the mixture of organizational
> classifications with geographical? "North America Free Trade
> Association" seems out of place in this scheme which is describing the
> objects by physical location. Classification as a NAFTA country might
> certainly be of value in e-business application, but it certainly is not
> a geographical category.

No, but it is a key grouping as far as e-business is concerned. Economic
groups are to me, the key to the role of the Regional context, specifically
where special Customs rules apply.

> And I don't even know where the US MINOR OUTLYING ISLANDS? are (and
> apparently, neither does the author given the big question mark), but I
> would lay odds that these islands are war-booty somewhere in the
> Pacific, and probably have no manufacturing base that makes NAFTA
> classification relevant.

I had presumed it was things like the Virgin Islands, (but not the British
Virgin Islands) and perhaps Hawaii. The question mark was there to indicate
my lack of knowledge of its relationship with the block it has for the time
being been placed in.

> And why would PR - PUERTO RICO, a wholly
> owned subsidiary of the U.S., be classified in the Caribbean, then, when
> for all commerce purposes (customs, taxes, postage, currency, banking),
> it is practically the same as doing business domestically within the
> United States?

If it is treated economically as a NAFTA member it should be moved there,
but if not it remains in the same geographical catchment area as Cuba and
the Bahamas.

>And is TR - TURKEY really Middle Eastern?  - A difficult
> judgment call since it straddles Europe and Asia.

This was a difficult one. It was placed in Asia by two considerations: it is
not yet a member of the EU and the bulk of the country is within Asia.

> Why should Europe and
> Asia *even* be classified as separate continents?  Unless you were
> trying to forcibly lump together a bunch of Muslim countries with Israel
> just to cause trouble, and though that may or may not be of any
> relevance to commerce, why is there a "Middle East" category at all?

Because insurures treat shipments to most Middle East countries as being in
a different risk category than the rest of Asia! Ideally we should have a
separate Arab Economic Region entry covering countries currently listed as
being in the Middle East or Africa, but without reference documents it was
impossible to do this at the Tokyo meeting. (I'll happily hand this task out
to you:-)  )

> And what's with the "Eastern Europe" ghetto?  That's not a geographical
> location - it's an occupied nation mindset, where Poland, Latvia,
> Estonia, Lithuania, Croatia, Slovenia, Hungary, the Czech Republic and
> Slovakia certainly don't belong - they're overwhelmingly Roman Catholic
> culturally if not in fact (with a smattering of misc. Protestants), and
> use the Latin alphabet - as sure a definition of "Western" as there's
> ever been.

As stated above, this is a catch-all for those countries that do not
currently belong to one of the established economic blocks.

>Further, it's interesting to note that all countries
> physically on the continent of Africa have been  categorized as one
> undifferentiated mass, perpetuating the stereotype of the "Dark"
> continent, as if we can't be bothered to tell any of those people apart.

As stated above, the reason is that they are treated as a single block for
quarantine purposes.

> It's a miracle that the ISO has even managed to name and code these
> entities.   Trying to further classify "countries" like this is a
> Herculean effort, sure to result in inaccuracy and insult.  So why even
> bother? I'm not even talking political correctness here: it's of no
> value in e-commerce.  It's a waste of time.  And it doesn't even serve
> as a useful pedagogical example.

I disagree with your statement that it is of no value to e-commerce.
Increasingly different trading rules need to be applied in different
economic areas. The way America is failing to tackle global warming this fac
tor is likely to become increasingly important in the future, not less so.
What is important is that we get the starter set of classifications correct
as regards to economic regions, as soon as we can identify someone with
sufficient knowledge and time to do this.

> The core ISO 3166 list, without classification, should be suitable for
> context setting;  additional lists could be made from subsets of ISO
> 3166 which identify useful agglomerations such as lists of NAFTA
> countries, EU countries, or NATO countries, etc. etc.

Agreed, except that you also need some way of grouping countries that do not
fit into existing groups (for example, the EU gives special treatment to
Eastern European countries in certain areas, and European car insurance
covers most countries in Eastern Europe as well as EU and EFTA countries).
That is what we were trying to achieve with the initial classification
scheme.

Martin Bryan