ISO 9 provides lossless one to one mapping of Cyrillic characters to Latin characters. Many post-Soviet countries, including Belarus and Russian Federation had adopted it as ГОСТ 7.79-2000.

I wonder why it was not adopted in Ukraine. Instead we have постано́ва (regulation) by Ukrainian government which contains somewhat controversial (and lossy) mapping. It doesn't even mention previously active (or still active?) standard for transliteration – ГОСТ 16876-71.

While researching this I've stumbled upon second draft of reworked transliteration standard. The most interesting part in it is the answer of Ministry of internal affairs in which they point to ICAO requirements for machine readable travel documents, that is (page 18):

To achieve global interoperability, the primary and secondary identifiers in the MRZ shall be printed using upper-case OCR-B characters, illustrated in Figure 4, without diacritical marks, and conform to the number of character positions available. As such, names in the MRZ are represented differently from those in the VIZ. The issuing State or organization shall transliterate national characters using only the allowed OCR-B characters and/or truncate, as specified in the form factor specific Parts 4 to 7 of Doc 9303. Transliteration tables for the most commonly used Latin, Cyrillic and Arabic families of languages are provided in Section 6.

Those characters are A-Z, 0-9 and <.

It seems to me that authors of the regulation tried to achieve 100% compatibility with ICAO requirements. If so, then the question is why the ICAO regulations should supersede our national standard? Not to mention that countries that adopted ISO 9 should've confirm to ICAO anyway.

  • 1
    They are not Latin characters. The point of ISO 9 is lost on me. With the advent of Unicode, the point may be completely lost for anyone.
    – Sassa NF
    Commented Feb 14, 2017 at 22:53
  • @SassaNF I don't think the problem lies only on a machine-to-machine level, human factor should be also considered. Imagine someone (hypothetically) searching for Оболонь on LSE. That is not feasible for many users, they'd search for Obolon instead (or Obolonʼ ?) Commented Feb 15, 2017 at 9:33
  • Not that they'd know how to enter or 02B9
    – Sassa NF
    Commented Feb 15, 2017 at 11:59

1 Answer 1


ISO 9 is not popular due to use of diacritics (not available in ukrainian keyboard layouts). Additionally ГОСТ 7.79 Система B (ISO 9 variant) has some references to ukrainian specifics, but it failed to be abopted in Ukraine. Most probable reason is poor phonetical equivalent for some of the characters:

  • y` for и - y in transliteration represent sound j (й), and grave accent represent soft sign (ь).
  • cz, с for ц - z in transliteration represent letter z (з) on its own.
  • shh for щ - sh in transliteration reprezent sh (ш) and extra h has no phonetic meaning.

FYI, for ГОСТ 16876-71 - there was an attempt to have its ukrainian variant, but it failed.

You should keep in mind contexts, where transliteration is used in Ukraine:

  • The primary one is formal documents, i.e. passports. ICAO requirements are essential іn passport. Additionally transliteration rules are only a baselaine since owner of the passport has the right to introduce corrections to her transliterated name with significant reasons, such as: transliterations in earlier documents, national transcribing tradition, etc.
  • Other transliteration cases (for instance geographical names) are of smaller significance to average person (including governamet officials, who adopted regulation), and thus passport is the most essential transliteration usecase being cared for.
  • To sum up, the main reason for ISO-9 failure is because it is letter-to-letter transliteration and we want to have phoneme-to-phoneme system, correct? However, isn't ISO-9 exists mainly because of ambiguity of the second approach? What is Kharkiv to an anglophone is Charkiw to a frankophone. Commented Feb 16, 2017 at 9:00
  • @NashBridges Ukrainian alphabet is 33 letters plus apostrophe. English alphabet has 26, you need to cover the difference somehow. Diacritics are not usable, thus multiletter transliteration should be used. Phonetic readablity is "nice to have". Ambiguity is problematic, but all schemas in use are ambiguous so far. I do hope that there will transliteration schema adopted that match the properties above.
    – myroslav
    Commented Feb 16, 2017 at 9:56
  • "Diacritics are not usable" because one expects to type them? But why do we have to do that? The goal isn't to move us towards using Latin in everyday usage. I thought that instead one simply takes a text in Ukrainian, passes it to a software and get a result (or vice versa). I may miss something though. Commented Feb 16, 2017 at 10:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.