EURid, the .eu registry, in collaboration with UNESCO, in November of last year released the 2012 World report on Internationalized Domain Names (IDNs) deployment. It updated previous year's study, Internationalised Domain Names State of Play, which was published in June 2011 and presented at the 2011 United Nations Internet Governance Forum in Nairobi, kenya.
Today, Irina Bokova, Director-General of UNESCO has released a statement concerning the linguistic diversity on the Internet stating: "UNESCO's experience and the 2012 study of the use of internationalized domain names undertaken with EURid show that the main challenges are technical. Obstacles lie with Internet browsers that do not consistently support non-ASCII characters, with limited e-mail functionality, and with the lack of support of non-ASCII characters in popular applications, websites and mobile devices."
Below is an excerpt from the 70-page EURid-UNESCO 2012 report:
This year, the data set for this study is expanded from 53 to 88 TLDs, and includes 90% of all domain names registered as at December 2011, albeit that the data set is not complete for every parameter. The World Report includes case studies on the ccTLDs for the European Union, Russian Federation, Qatar, Saudi arabia, Egypt and the Republic of korea. Where an existing registry has launched an IDN ccTLD (for example, .sa and السعودية.) these are considered as two separate entities for the purpose of the report.
Part 1 of the World Report on IDN deployment sets out a background to IDNs and a timeline. It considers progress in supporting IDNs in email and browsers. It then reviews the IDN applications in ICANN's programmes to create new TLDs. A comparison of growth rates of IDN registrations versus general registrations is made within European registries and usage rates are compared amongst .eu and .рф IDNs and benchmarked with other TLDs. Case studies follow, on the European Union (.eu) ccTLD, and country case studies on the Russian Federation, Qatar, Saudi arabia, Egypt and the Republic of korea.
Also noteworthy is the included foreword in the report by Vint Cerf (excerpt below) on the historical adoption of simple Latin characters in the early days of the Domain Name System (DNS). Cerf writes:
"For historical reasons, the Domain Name System (DNS); and its predecessor (the so-called "host.txt" table) adopted naming conventions using simple Latin characters drawn from the letters a-Z, digits 0-9 and the hyphen ("-"). The host-host protocols developed for the original aRPaNET project were the product of research and experimentation led in very large part by English language speaking graduate students working in american universities and research laboratories. The project was focused on demonstrating the feasibility of building a homogeneous, wide area packet switching network connecting a heterogeneous collection of time-shared computers. This project led to the Internetting project that was initially carried out by researchers in the United States of america and the United kingdom, joined later with groups in Norway, Germany and Italy, along with a few visiting researchers from Japan and France. The primary focus of the Internetting project was to demonstrate the feasibility of interconnecting different classes of packet switched networks that, themselves, interconnected a wide and heterogeneous collection of timeshared computers.
The heterogeneity of interest was not in language or script but in the underlying networks and computers that were to be interconnected. moreover, the Internet inherited applications and protocols from the aRPaNET and these were largely developed by English language speakers (not all of them necessarily native speakers). The documentation of the projects was uniformly prepared in English. It should be no surprise, then, that the naming conventions of the Internet rested for many years on simple aSCII-encoded strings. The simplicity of this design and the choice to treat upper and lower case characters as equivalent for matching purposes, avoided for many years the important question of support for scripts other than Latin characters. as the Internet has spread across the globe, the absence of support for non-Latin scripts became a notable deficiency.
For technical reasons, support for non-Latin scripts was treated as a design and deployment problem whose solution was intended to minimise change to the domain name resolution infrastructure. This was debated in the Internet Engineering Task Force more than once, but the general conclusion was always that requiring a change to every resolver and domain name server, rather than changes on the client side only, would inhibit deployment and utility. This led to the development of so-called "punycode" that would map Unicode characters representing characters from many of the world's scripts into aSCII characters (and the reverse). This choice also had the salient feature of making unambiguous the question of matching domain names since the punycoded representations were unique and canonical in form. This design is not without its problems but that is where we are at present."
IDN introduction timeline – Source: EURid-UNESCO World report on Internationalised Domain Names deployment 2012 (Click to Enlarge)
The full report can be downloaded here in PDF here: EURid-UNESCO World report on Internationalised Domain Names deployment 2012
|Cybersquatting||Policy & Regulation|
|DNS Security||Registry Services|
|IP Addressing||White Space|
Minds + Machines