BLOCKQUOTEThe other important goal, however, is that other languages are poorly represented
in the DNS, and non-English countries will represent the fastest-growing parts of
the Internet, especially Asian countries. That is a strong reason to favor non-English
proposals, provided that businesses and organizations are ready and able to come
forth with viable proposals for them. /BLOCKQUOTEI disagree. East Asian
languages are largely confined to single countries, causing significant overlap between
new gTLDs and already-existing ccTLDs. The biggest exception (Mandarin Chinese)
has the least consistent Romanization, making it the poorest candidate for use in
gTLDs.
As long as ICANN is only going to begin with only a few new gTLDs, the new
mnemonics should be limited to languages with Roman alphabets, significant multinational
distribution, large numbers of nonnative speakers, and large numbers of speakers
with access to the resources to use the Internet. That is, English, French,
German, and Spanish (with a special emphasis on English, but ideally the new mnemonics
should have similar associations/meanings in all four of those languages plus as
many others as possible).
When things move to the level of having several dozen
or more gTLDs, gTLDs of less multinational utility will make sense. In the
meantime, the new gTLDs should be comprehensible to virtually all residents of the
Americas, Australia, New Zealand, and Western Europe; most of the persons in Japan
and South Korea born since 1945; and the well-educated and wealthy in South Asia
and Africa. Which is to say, a significant majority of all those using the
Internet and all those likely do do so in the next five years.