Script codes in RFC 3066, 4 issues

Peter_Constable at Peter_Constable at
Wed Apr 9 12:31:23 CEST 2003

Mark Davis wrote on 04/09/2003 09:06:42 AM:

> 1. One small quibble:
> > A writing system is encoded by
> > an ISO 15924 script code.  An orthography is encoded by an ISO 3166-1
> The term writing system is often contrasted with script. There is no need
> identify them; it is simpler to always use script:
> "A script is encoded by an ISO 15924 script code."

The latter statement is, of course, self-evidently appropriate, but it
misses what I think John was getting at. There is a kind of distinction
that needs to be made as illustrated in examples like Azeri in Cyrillic
script versus Azeri in Arabic script, or English in the common Latin
orthography versus English in phonetic transcription versus English in

These things to be distinguished are not scripts; they are something else.
In the papers that I wrote and to which John referred, I proposed the
terminology "writing system" for this type of category. Ken Whistler has
been concerned as this usage since it doesn't match usage familiar to
grammatologists, and so is potentially confusing. Hence, he has suggested
that my terminology should be changed thus:

instead of my "writing system", use "orthography"
instead of my "orthography", use "spelling convention"

I have no strong objection, since it's important that we can communicate
clearly, and since I place more importance on finding the best underlying
ontological model than on the terminology that is adopted.

What I think John was meaning, then, is that an orthography (or "writing
system" -- whichever term is preferred) is encoded by combining an ISO
15924 script ID with a language ID.

> There are compatibility issues with this both ISO 639 codes and
> codes are growing; a "shortest" code may change over time, which makes my
> use of a code now legal, tomorrow illegal. (E.g., a new ISO 639 code
> subsumes an Ethnologue code.)

could you provide a hypothetical example to illustrate what the concern
here is?

> 3. For compatibility, also we need that once a 3066bis code, forever a
> 3066bis code. That is, even if the Ethnologue or ISO remove/deprecate a
> code, that code is remains forever valid for use in a 3066bis subtag.

Absolutely. I would add, though, that changes will inevitably be needed as
knowledge of languages improves and as the sociolinguistic realities
change, and so relationships betweeen IDs that become deprecated and new
IDs that replace them needs to be documented. This has begun to happen with
the Ethnologue, and I would expect that information to be incorporated into
what is provided with ISO 639-3.

- Peter

Peter Constable

Non-Roman Script Initiative, SIL International
7500 W. Camp Wisdom Rd., Dallas, TX 75236, USA
Tel: +1 972 708 7485

More information about the Ietf-languages mailing list