Doug Ewell dewell at
Sun Apr 2 07:58:05 CEST 2006

Frank Ellermann <nobody at xyzzy dot claranet dot de> wrote:

>> But the $64,000 question is... would it have happened if this
>> group had a liaison officer with ISO 3166 ;-)
> The CS disaster ?  Many folks including the IAB Chair thought
> that it was wrong, she wrote an open letter about it to the
> 3166 MA.  It was about a years later that I finally understood
> that they are simply forced to recycle alpha-2 region codes at
> some point in time.

They are only forced to recycle code elements to the extent they feel 
compelled to achieve mnemonicity.

Out of 676 possible alpha-2 code elements, 345 of them -- more than 
HALF -- are not allocated in any way.  They are neither assigned, 
reserved, private-use, escape codes, nor "not used at present stage." 
They are *completely free*.  This is over 100 more than the total number 
of "officially assigned" code elements.

Almost all ISO 3166-1 code elements are at least somewhat mnemonic for 
the country name they represent, in some language, and *that* is what 
causes the perceived shortage.  There are a lot of country names that 
start with A, B, C, G, M, S, and T.  When another one comes along, the 
mnemonic code space may be tight, but there are scores of available code 

Remember that every code element ever assigned to a country since 1974 
is either assigned or reserved today, and there are still 345 that have 
never been used.

> Five years or fifty doesn't change this.

There may not be a theoretical difference between 5 and 50 years, but 
there is arguably a practical difference.  There is a noticeably better 
chance for a formerly used name (and thus a formerly used code element) 
to go out of common usage, and for data to be replaced or retagged, 
after 50 years than after 5 years (or 10, the actual elapsed time in the 
case of CS).  Tragically, in the U.S. there are some people who still 
think Czechoslovakia is a country.

Doug Ewell
Fullerton, California, USA 

More information about the Ietf-languages mailing list