Recent editorial missed the mark

Published 9:00 pm Monday, December 5, 2005

The Nov. 18 editorial titled “Internet should speak more than just English” misses the mark. Web addresses right now are restricted to the Latin alphabet as embodied by the ANSI character set. The United Nations wants to open that up to use the UNICODE set, which supports all the languages of the world. No one is being “provincial or selfish” in wanting to keep this restriction. There are valid security issues at stake here.

The most obvious would be the malicious chaos introduced by alternate characters. There are many letters in different alphabets that look like the Latin “o,” for instance. Imagine clicking on a link for “http://www.google.com” because you think Google is a safe site to visit, only to be taken to a porn site because one or both of the “o” characters are Cyrillic letters, not Latin. Or worse: You’re taken to a site owned by someone who maliciously infects your machine with a virus or tricks you into divulging your bank account password.

This is a very important technical distinction! Although they may look identical, a Cyrillic “o” is not the same character as the Latin “o.” There are many letters in many different alphabets that have this problem, and addresses using one are not the same as addresses using another. If we open domain names to all alphabets, this kind of misdirection will become commonplace and undetectable to the naked eye.

When politics starts meddling in technology issues, we all suffer.

Ron Logan

Everett