White Paper 1: O Foundation’s Responses to Oriya Script Root Zone Label Generation Ruleset (LGR) proposal 2018

This document is prepared by O Foundation (often acronymed as “OFDN”) based on a submission¹ dated October 8, 2018 to the “Proposal for an Oriya Script Root Zone Label Generation Ruleset”² (LGR) – the latter henceforth to be called the Proposal.

A follow up white paper was published on 22 November 2018 which can be accessed here in HTML and here as PDF.

Context:

Root Zone Label Generation Ruleset (LGR) was released by the the Internet Corporation for Assigned Names and Numbers (ICANN) in August 2017³. ICANN is a nonprofit which maintens and sets the process of multiple namespaces and numerical spaces databases of the Internet so that the world wide web remains a stable and secure place. Root Zone is the top-most layer of the domain name hierarchy.

The Proposal mentions that —

“Oriya script seems to be a variant of Devanāgarī”. This is a glaring mistake.

It is not just an insult to a classical language and a script [Odia] that existed more than 1500 years ago in eastern India around the time Sanskrit and Tamil flourished in the other parts of the subcontinent, but it also undermines the struggle that went into making Odisha the first state in India to be formed in 1936 on the basis of linguistic identity.

Additionally, it feels like a large portion of the content is copied from Wikipedia without verification from external sources. Linguist Liang Hai⁴ has already flagged the Gujarati part where the Odia script is termed as a variant of the Devanagari script that has use of a Mahajani merchant community. It is important to note that linguists like Debi Prasanna Pattanayak have mentioned of Odia script not being a variant of Devanagari but having similarity with the Brahmi script.

It is disheartening to see that researchers are referring to Wikipedia rather than referring to primary and secondary sources. Wikipedia itself refers to tertiary and above sources and at times contains factual inconsistencies.

Section 3.4 of the Proposal:

It is important to note that “ଵ” (U+0B35) is a non-existent character in the modern Odia alphabet. It was invented by a fringe group during sometime in the last few decades but never gained traction among the general public. It was later inserted in the Unicode chart for Odia based on unreliable sources but still remains an alien symbol for majority of the native speakers of the language.

The following documented evidences corroborates with the historical non-existence of any such character in Odia:

Notable Dictionaries or lexicons published over the last two centuries (including the most exhaustive and largest Odia lexicon “Purnachandra Bhashakosha”).
Odia literature has special literary genres known as Chautisa and Champu that were written from 7th century AD up to 19th century AD. These are poems where each stanza begins with a letter of the alphabet and continues sequentially. These also do not use this character making it of doubtful antiquity.
Scholarly studies on Odia linguistics, literature and grammar as well primary sources such as stone inscriptions and palm leaf manuscripts.
Odia language primers and textbooks including those published by governmental agencies (eg: Dept of Language, Literature and Culture of Government of Odisha, Odisha State Museum).
Popular print or electronic media including newspapers, journals and television.

Section 3.8 of the Proposal:

Another dubious claim with no explanation or logic whatsoever is the addition of nukta to “କ” (U+0B15),“ଖ” (U+0B16), “ଗ” (U+0B17), “ଚ” (U+0B1A), “ଜ” (U+0B1C), and “ଫ” (U+0B2B). There is no reference or any published resources to show the need or historical use of these. Simply put, the researchers should have gone beyond Wikipedia to find if the historical use actually exists as these characters ignore the efforts of those who standardized the language and the script on the basis of which the Indian state of Odisha was formed in 1936. If a script would evolve, it would evolve based on a dialog between the experts and the larger community. Insertion of nukta to these characters are probably done in this document in an autocratic manner without any large community consensus, historical reference and arguably to substantiate a new trend one of the researchers for Odia is promoting on social media. This should not be treated as an allegation but a cautionary flag as these serious flaws will tarnish the hard work of ICANN.

Nukta in “ଚ” (U+0B1A), however, is visible in the Karani script which is another historical variation/predecessor of the current Odia script but has to be treated as a different script. The reason for nukta in “ଚ” (U+0B1A) was for a different purpose and bringing it back for another purpose can be a gross manipulation.

We strongly support Hai’s point about “ଡ଼” and “ଢ଼” as those two characters are regarded as treasure troves of Odia script. The very name of the language “ଓଡ଼ିଆ” and the geographical place — the state of Odisha (ଓଡ଼ିଶା) — contain these characters. “ଡ଼” and “ଢ଼” are supposed to be treated as characters rather than variations of “ଡ” and “ଢ” as the use case for the former two are more than the latter.

This response was drafted immediately as the deadline was October 8, 2018. A complete and more elaborate response with necessary sources will be published and shared.

Published by O Foundation (OFDN) under a Creative Commons Share-Alike 4.0 International license. OFDN team members Prateek Pattanaik, Nasim Ali, Sailesh Patnaik, Jnanaranjan Sahu, Chinmayee Mishra, Mrutyunjaya Kar, and Subhashish Panigrahi.