Log in — the desi way

Domain Names In Indian Languages To Become A Reality
Pune: The web is set to reach new heights in India as use of several Internet Domain Names (IDN) got a thumbs up at an international meet held at New Delhi recently.

Internet Corporation on Assigned Numeric Names (ICANN), the apex body, which controls domain names, discussed the subject at ICANN’s 31st public meeting in the capital on Friday.

Mahesh Kulkarni, the programme head of Punebased Centre for Development of Advanced Computing (C-DAC), gave a presentation on ‘development of Internationalised Domain Names (IDN) in local languages’.

“There are two sides to this issue—policywise and technical. We have recommended to the Department of Information Technology and ICANN to bring some policy changes to enable IDNs in local languages. The work is going on for two years but it has reached an advanced stage now,” said Kulkarni. India has 22 official languages and domain names in 13 languages have already been tested. But the issue now is allowing top level domain names (.com, .gov, .in, etc) in Indian languages.

Kulkarni said, “Given the complexities of our languages, the issue has to be resolved systematically. The main risk in allowing IDNs in Indian languages will be from phishing sites (who indulge in internet frauds using duplicate domain names). Keeping this in view, we have made some recommendations.”

According to him, if the suggestions are accepted, it could pave the way for email addresses in local languages.
----------
Risks and the solutions

There are 22 languages written in 11 scripts. Many languages are written in more than one script — Konkani is written in Roman, Devnagari, Malyalam and Kannada, while 54 languages (including dialects) are written in Devnagari.
Solution: Deciding one script for any given language, on which a decision is yet to be taken.

There are many characters common to more than one languages. Making it easier for fishing sites to pose as original websites.
Solution: Reserving specific names for important sites and not allowing similar characters for domain names. A variant table of characters for each language has been prepared so that character set of one language is not copied to other characters.

Many languages have similar sounding words (homophones) and similarly spelled words (homographs) making the access difficult as well as prone to risk.
Solution: Unicode sets for each character has been finalised and only homographs are focused, restricting uncommon characters from using in domain names.

There is a risk that some obscene words or phrases might be used by some elements to register the names.
Solution: A list has been prepared of such words which will not be allowed to use, a la European countries.