19 Feb 2024 | By Mark W. Datysgeld
Introduction
The Greek language has existed for well over two thousand years. Its alphabet pioneered the use of dedicated vowel symbols alongside consonants, and the intellectual legacy of that writing system continues to resonate in literature, philosophy, and science. The standardized 24-letter Ionian alphabet served as the basis for the Latin script in which these very words are being written. Now, this cornerstone of Hellenic identity extends into the digital era in part through Greek script IDNs, providing the opportunity for further preservation of this cultural heritage.
Historical context
The Greek alphabet traces its origins back to the 9th or 8th century BCE, being widely acknowledged as the first alphabet to represent both vowels and consonants. This innovation was an incremental improvement upon the Phoenician mercantile script, which lacked distinct vowel notation. The Greek alphabet allowed for accurate phonetic transcription, laying the groundwork for the systematic archival and exchange of knowledge—a process that contributed to significant advancements in philosophy and science (Horrocks, 2014). The word “alphabet” itself derives from its first two letters: Alpha (α, U+03B1) and Beta (β, U+03B2).
Over centuries, various forms of the Greek alphabet emerged, each reflecting regional dialects and cultural nuances. Yet, by the end of the 4th century BCE, the standardized 24-letter Ionian alphabet had prevailed across the Greek-speaking world. The Greek language has proven remarkably resilient, as despite centuries marked by conquests, cultural shifts, and periods of foreign domination—from Roman and Byzantine rule to Ottoman influence—the Greek language and its script have remained a continuous thread of identity and expression (Parker & Steele, 2021).
Standardized 24-letter Ionian alphabet used for Greek |
Α α, Β β, Γ γ, Δ δ, Ε ε, Ζ ζ, Η η, Θ θ, Ι ι, Κ κ, Λ λ, Μ μ, Ν ν, Ξ ξ, Ο ο, Π π, Ρ ρ, Σ σ ς, Τ τ, Υ υ, Φ φ, Χ χ, Ψ ψ, Ω ω |
This standardization was instrumental in preserving and propagating the classical works that continue to shape contemporary Western (and, to some extent, Eastern) thought, including those of Homer, Plato, and Aristotle. The enduring nature of the Ionian system has cemented it as a defining component of Greek cultural identity. A persistent scholarly tradition has ensured the survival and evolution of Greek, despite its natural development and amalgamation with neighboring languages (Davy, Ioannou, Panayotou, 1996).
As for Cyprus, evidence indicates that the Greek alphabet was already employed on the island by the 8th century BCE, with local administrations and elites later embracing the Ionian system. Cyprus emerged as a cultural conduit between the Greek mainland and the Eastern Mediterranean, with the tradition of Hellenic literacy persisting under Roman and Byzantine rule. Despite various political and social changes, Greek remains the dominant language in Cyprus, and its script continues to underpin educational systems, media, and cultural expressions today (Davy, Ioannou, & Panayotou, 1996).
Technical considerations
As outlined in RFC 5892, Greek domain names are allowed to use the complete Greek Unicode block, including characters modified by diacritics, such as ά (Greek Small Letter Alpha with Tonos, U+03AC). The exception is the Greek Lower Numeral Sign/Keraia (U+0375), which is deemed invalid (not PVALID) due to its numeral function. For example, when the Keraia appears next to Ρ (Greek Capital Letter Rho, U+03A1), “͵Ρ” indicates “one hundred thousand”, although Keraia can be employed to express other large numbers.
Another complication requiring careful consideration is the letter Sigma, of which there are two distinct forms: U+03C3 (Greek Small Letter Sigma) which is used in initial and medial positions within words, while U+03C2 (Greek Small Letter Final Sigma) is exclusively used at words endings. This positional variation is essential for the correct representation of Greek orthography, demanding precise mapping for normalization and encoding purposes (Greek Case Study Team, 2011).
Finally, the Greek alphabet essentially forms the basis of the Latin alphabet. This creates a complex situation similar to that found in other scripts with shared ancestry, like Cyrillic, requiring extensive measures to avoid homograph issues. Because many Latin letters originate from Greek, visually confusable pairs are common. To mitigate this risk, the Greek ccTLD policies enforce strict script-homogeneity rules that prohibit mixing characters from the Greek block with those from Latin (or other) blocks (Greek Case Study Team, 2011).
IDN ccTLD implementation
The string chosen by the Greek government to represent the IDN of its ccTLD, “.gr”, was “.ελ” (U+03B5 + U+03BB). This is the shortened form of “Ελλάδα” or “Elláda”, which is the contemporary name for Greece. This choice aligns with how several other countries decided on their IDN ccTLDs, using local naming convention rather than matching the ASCII form provided by the ISO table in a different script, which would have resulted in “.γρ” (U+03B3 + U+03C1).
After some contention with the Internet Assigned Numbers Authority (IANA, then an entity under control of the U.S.), an evaluation—of a level of detail and minutiae which seems excessive, at least when analyzed outside of the context of whatever pressures surrounded the question at the time—reached the conclusion that ΕΛ (Serif Uppercase, U+0395 + U+039B) had an “error rate above threshold” in relation to the Latin script for EA (in ASCII), and therefore was confusingly similar, creating potential security risks in the usage of “.ελ”.
To this, George Papapavlou, then-representative of Greece in ICANN’s Governmental Advisory Committee (GAC), sent a letter stating that:
Greece has had a long and traumatic experience in the process of having its selected IDN, .ελ, introduced into the root. (…) IANA has no right to question languages or local Internet community support. Governments are in the position of expressing their national Internet communities. The “confusing similarity” criterion should not be the result of some technical algorithm but of simple common sense. In the Greek example: The capital letters version of .ελ (ΕΛ) was considered to be confusingly similar to the Latin alphabet letters EA. The possibility of such confusion for a Greek language speaker, who uses exclusively the Greek alphabet to type the whole domain name or address, to then switch into capital letters and type EA in the Latin alphabet is close to zero. After all, there is currently no .ea or .EA ccTLD. In my view, any IDN request which is technically complete, tabled by a body with obvious TLD expertise and has the support of the government concerned should be accepted by IANA. Where there is a need for technical judgment, ccTLD-designated delegates should be part of any IANA technical committee to ensure that decisions are in conformity not only with technical but also with common sense criteria regarding the alphabet concerned. |
As this unfolded, ICANN started the Variant Issues Project to produce case studies for the mitigation of the complexities of 6 scripts: Arabic, Chinese, Cyrillic, Devanagari, Greek, and Latin. A group assembled within the community as the Greek Case Study Team and produced the “Study of the Issues Present In the Registration of IDN TLDs In Greek Characters” for that project. This document outlined many of the cautions presented in the “Technical considerations” section above, as well as addressing specific cases such as “Greeklish”, which is the transliteration of Greek using ASCII.
It was not until late 2015 that the Greek ccTLD would be officially delegated under the Punycode “xn--qxam” for “.ελ”. It became available for public registration in 2018, with rules in place so that domain names need to be either entirely in Greek or entirely in Latin characters, as well as that if a domain name exists under “.gr”, then only the same registrant can claim its “.ελ” version. For the time being, Cyprus remains focused on its “.cy”, but has nevertheless claimed the rights for the IDN versions of the country’s name.
Despite the delays generated by ICANN procedures and bureaucracy that took away some of the momentum of the ccTLD’s release, the Greek IDN has found adoption among local businesses and initiatives eager to emphasize their connection with the country and demonstrate their cultural pride by means of their domain name.
Conclusion
The integration of Greek-script IDNs demonstrates the potential for powerful convergence of cultural preservation and digital innovation. As technical challenges such as Universal Acceptance and homograph management continue to be addressed by means of international collaboration, the need for the generation of greater public awareness about these resources is still very desirable, in order to allow for full linguistic expression in the digital era.
By Mark W. Datysgeld