I18N (was: Re: (dictionary or not) Re: [dtn-security] 00 version of the Bundle Security Protocol Spec.)

[Stephen Farrell <[e-mail]>]

Scott Burleigh wrote:

> Michael Demmer wrote:
> 
>>> Aha!  Ok, it can work that way so long as there're no I18N
>>> problems there (there aren't, right?).
>>
>> Well -- we need to decide on what to do re I8N. My inclination is to
>> have the protocol just be in ASCII, and put the burden on the
>> application interface to go to/from some multibyte encoding when
>> interfacing with the application.
>>  
>>
> I agree: the external representation locally exposed to applications and 
> users is where the internationalization is really needed.  On the wire I 
> don't think it's useful.

I disageee - endpoint IDs are v. likely to want to include
non-ASCII characters since they'll practically be used
to name things.

Furthermore the way we're farming out identifier scheme
standardisation (via IANA or RFCs or whatever) just
increases the liklihood that I18N will be needed.

Read stringprep to see how bad it gets! Talk to Paul Hoffmann
or someone else that knows if you want to hear to pain that
iDNS invovled.

Stephen.