Re: [Ltru] Minor proofreading nits again

["Martin J. Dürst" <[e-mail]>]

On 2011/07/14 16:18, Jukka K. Korpela wrote:

> Moreover, "on language" is somewhat mild, as transliteration may depend
> both on the original language of a name (or other text) and on the
> language environment where the transliteration is used - and only the
> latter is discussed in the example. For example, Cyrillic letters can be
> transliterated according to different principles depending on whether
> they are Russian, Ukrainian, Khantuan, or something else.
>
> I would say "but also on source language and in the language context
> where the transliteration is used". And maybe the example could be
> replaced by a more difficult one - like a Greek name that has multiple
> translations, depending on whether it is treated as a classical name or
> a modern name, on the language of the context, and on the specific
> transliteration scheme used.

There are certainly cases where there's more than the source and target 
language and script involved. But on the other hand, there are also 
cases where there's not really a target language.

An example would be what can currently be denoted by ja-Latn-hepburn. My 
understanding is that such cases are also supposed to be covered by -t. 
How would such cases look? How much more time and effort (than for a 
variant subtag) would be required for registration.

Regards,   Martin.