Chair, All, I've been asked by an acquaintance who's first langugae is written in the Arabic Script to comment on the applications by the .COM zone contractor which may replicate many of the existing features of the Latin Script .COM zone in one or more Non-Latin Script zones. I think the question is "should the business model associated with most of the Latin Script namespaces be imposed upon the Non-Latin Script namespaces?" This is somewhat abstract from how the question was posed to me, and some explination may be useful. The current .COM contractor has applied for about a dozen strings in scripts other than Latin which when sounded out are similar to the sound usually associated with the string "COM". My acquaintance pointed out that users of Arabic (and Farsi) domain names written in Arabic Script, will not know which namespace is the intended namespace when fully qualified domain names are communicated orally. He observed that this intentional similarity will (a) force existing .COM registrations to be replicaed in the Verisign Arabic Script zone, and where this replicated registration is not made by the original .COM registrant, will (b) facilitate abusive registrations by third parties. I assume that his observations are likely to prove correct, that duplicate, or abusive, registrations will be the prevelant initial form of registrations in the "sounds-like-COM" namespaces. How I infer that this will have the more general effect of repeating the registration practices and patterns of registrations of the .COM market in each of the Non-Latin Script domain name markets is by observing that with a market share of 90,000,000 domains in the COM zone, if only 1% of the COM registrants (and their abusers) replicate their Latin Script registrations in the Non-Latin Script namespaces Verisign has applied for, each Non-Latin "one percent of COM" will have nearly a million registrations, in scripts which currently have no registrations other than a few restricted IDN ccTLD equivalents created under the ccTLD IDN Fast Track Program. Having walked thus far out on a pier of unknown length with a murky crystal ball, I would like to suggest that there are some public interest issues. 1. "Confusingly Similar" namespaces may benefit registrants and end users if the two (or more) registries offer the substantively similar resolution outcomes. An example is the Simplified Chinese / Traditional Chinese (SC/TC) equivalence, where a substantial portion of the Han Script users associate substantively similar meaning to equivalent characters (bi-literacy). 2. "Confusingly Similar" namespaces will not benefit registrants or end users if the two (or more) registries do not offer the substantively similar resolution outcomes. 3. The registration practices and patterns of registrations of the .COM market are problematic and as a matter of public policy should not be replicated. 4. The market share of the current .COM incumbent contractor meets the test to trigger competition policy restrictions on the maintenance and expansion of that entities market share. 5. The development of mappings from persistent identifiers to transient identifiers (names of resources to addresses of resources) in scripts other than Latin must arise from public practice, not the promotion of private interest. I suggest that a North American At-Large position on a region-local vendor's exploit of its Latin Script market share that, if allowed, may determine the development of the DNS for scripts other than Latin, is appropriate subject matter for the Toronto meeting. I therefore suggest that a few minutes be allocated to the subject during the September 10th monthly call time. Eric Brunner-Williams, Unaffiliated Eugene, Oregon