Ietf-languages Digest, Vol 81, Issue 40

[Mark Crispin]

Thanks Frank for your comments.

Given what you said, I agree that it probably does not make sense to both 
with registering Nihon-shiki.  The final concern that I have is this:

Would your application be adversely impacted if it turned out that the 
ultimate usage of the Hepburn tag became "something that is more or less 
Hepburn" (with all of the variants therefore) and of the kunrei-shiki tag 
became "something that is more or less kunrei-shiki" (with all of the 
variants such as JSL, Nihon-shiki, etc.)?

My prediction is that if you expect the Hepburn tag to be used only for 
strictly compliant Hepburn (as defined by Hepburn) and the kunrei-shiki 
tag to be used only for strictly compliant ISO 3602, you will be 
disappointed.

If, on the other hand, it's alright if the usage winds up being generic, 
and that we call text Hepburn if it has "Fujitsu" and kunrei-shiki if it 
has "Huzitu", then I see no problem.

Or, put another way, for the Japanese word 失礼 ("rudeness"), can a user 
of these tags triage:
 	Hepburn:	shitsurê, shitsurē, shitsurei, shitsuree
 	kunrei-shiki:	siturê, siturē, siturei, situree
without causing trouble for you and your application?  If so, then we're 
good.

As you may have guessed, my hangup is with bothering to register the 
distinction between Nihon-shiki and kunrei-shiki yet not any of the other 
variants which commonly appear.

I agree that it is madness to try to register all the possible variants in 
order to have strict compliance.  I've noticed that even native Japanese 
who prefer kunrei-shiki don't strictly follow ISO 3602 once they are out 
of the clutches of their primary school and no longer have to do as the 
Monbushou says...

If the generic usage is alright, then I suggest we the registration say 
something to reflect this and specifically debunk any assumption of strict 
compliance.

-- Mark --

http://panda.com/mrc
Democracy is two wolves and a sheep deciding what to eat for lunch.
Liberty is a well-armed sheep contesting the vote.