Japanese transliteration: ja-Latn-hepburn

[Phillips, Addison]

> 
> The question is whether the variant 'hepburn' for our purposes
> means
> only Revised Hepburn, in which case we would need a fourth and
> fifth
> variant under "ja-Latn" to denote Traditional and Modified Hepburn
> respectively, or whether 'hepburn' means any variety of Hepburn, in
> which case we would need three new variants under "ja-Latn-hepburn"
> to distinguish between the three varieties.  

The test, I think, is what the requester needs. If the requester needs to identify a specific flavor of Hepburn, to the exclusion of others, then the request should be narrow and specific. 

> This is the approach I prefer,
> because it would still be possible to say "any old Hepburn," which
> I think will be a common use case since the varieties are so similar.

I tend to agree. Unless there is a good reason to do so, narrowing a subtag limits its usefulness and promotes confusion about how to tag content. Unless the distinction is immediately important to the requester, it is probably best to save finer details for private-use or future registration. Since users typically rely on the registry alone or the imputed meaning of the subtag--because of an expectation that language tags are self-documenting--it would be unsurprising if 'hepburn' were used generically. To the extent that we can make it so, a subtag's meaning should be unsurprising.

> while currently there is no way to tag Tongyong Pinyin.
> 

There are always private use sequences: "zh-Latn-x-tongyong"

Addison

Addison Phillips
Globalization Architect -- Lab126

Internationalization is not a feature.
It is an architecture.