Proposed modified records for 'cpe' and 'son'

Doug Ewell doug at
Wed Dec 31 18:14:50 CET 2008

CE Whitehead <cewcathar at hotmail dot com> wrote:

> Whatever you do is fine, so long as you have backwards compatibility 
> (that is, 'cpe-Latn' will be backwards compatible with 'cpe' ??  I 
> should know whether it will be backwards compatible or not by now).

RFC 4647 explains the concept of matching.

> However I do note that many English dialects are unwritten but if 
> content in these were to be tagged, the suppress-script would still be 
> Latin (& thus unwritten content would have to be tagged 'Zxxx' I 
> think).

I don't agree with John that the presence of unwritten languages within 
the collection is an argument either for or against Suppress-Script. 
S-S says that *if* the content is written, it is almost certainly 
written in such-and-so script, so the script subtag should ordinarily be 
suppressed.  But it doesn't say that the content must be written, or 

Elsewhere in RFC 4646 it is made clear that the absence of a script 
subtag, for whatever reason, does not imply that the text must be 
written, and that unwritten content is not required to be tagged as 
such.  Script subtags are never required, and Suppress-Script is a 

> My guess is that most often people who try to write these down will do 
> so using a Roman alphabet ;

That's fine.  Removing the S-S doesn't mean we are making some kind of 
authoritative statement that people write these languages with other 

> if left audio then these can be tagged with a script code of 'Zxxx'.

Or not.  There's no requirement that unwritten text must be tagged as 

>> because the languages that are in the collection are not enumerated 
>> anywhere by ISO 639.
> (But some are registered:  Tok Pisin is, as is Haitian Creole; I can't 
> find Guadaloupe Creole however . . . )

That wasn't his point.  ISO 639 doesn't tell of which of its 7,600 
languages belong to collection X, so it's impossible to determine 
whether the S-S should apply to all of them.

Suppress-Script is by far the most overdebated topic on this list.

Doug Ewell  *  Thornton, Colorado, USA  *  RFC 4645  *  UTN #14  ˆ

More information about the Ietf-languages mailing list