Re: Ethnologue

Date: Wed Sep 13 2000 - 16:36:41 EDT

On 09/13/2000 06:37:25 AM Michael Everson wrote:

>>First, by the definitions assumed in the Ethnologue, they are all
>>considered to be distinct languages; they would be candidates for
>>literacy and literature development (if currently spoken-only), and if
>>literature were to be developed, then they would need to be distinguished
>>for IT-processing purposes such as spell-checking.
>Yes, well, it is these definitions and how they are implemented which we

But the only questions raised about the definitions and their
implementation have been based on anecdotal evidence, or on attested
instances of error in the results, which certainly exist but which do not
logically invalidate the definitions.

>>Thirdly, there are a large number of users looking for a complete set of
>>tags. This includes people in the specialised fields of linguistics and
>>anthropology, but it also includes governments, development
>>and many businesses in the IT industry.
>Peter, why has the Maintenance Agency for ISO 639 never heard from any of
>these people, if governments are looking for standardized tags and
>else is looking for tags, why haven't they contacted 639 or 1766 to inform
>us of the kinds of requirements they have?

I can't answer for all these users. For example, I have no way of knowing
why UNESCO consulted SIL on enumerating the world's languages rather than
ISO. My understanding is that some have attempted to make their needs
known, but for whatever reason have not succeeded in getting the reponse
they hoped for. Perhaps there is a perception that ISO is unresponsive
leading people not to make their requests. Perhaps the Maintenance Agency
*is*, in fact, unresponsive. That's how you've been coming across in these
discussions: rather than saying, "I recognise the need, but have some
concerns about some details, so lets investigate how we can find the best
all-around solutions," your response has been, "I am not interested in
considering the list of languages enumerated in the Ethnologue."

There may be other reasons I don't know about. As for SIL, we have not made
any request before now since (a) ISO 639 so obviously and thoroughly fell
short of what we needed that there was no indication we could expect the
kind of support we need; (b) our efforts at enumerating languages for IT
purposes have a rather longer history than does ISO 639-x, and we have up
to now met our own needs - that's why the Ethnologue exists in the first
place. We are only now beginning to broach this topic with IT standards
bodies for these reasons:

1. Others keep coming to us asking us to do this.

2. We want to engage with other agencies in developing distributed archives
of online, linguistic data, and want to conform to industry standards like
XML (and, thus, RFC 1766 or its successor) in order to ensure good
documentation and interoperability.

3. We are increasingly involved in partnership with outside agencies like
UNESCO or government agencies for whom standardisation is deemed important.

If we only needed to maintain our own data internally, it would be a whole
lot easier for us to do this ourselves. It's the interaction with those
outside SIL that is pushing us to pursue more. Whether results are achieved
through ISO 639-x or through some successor to RFC1766 doesn't really
matter a lot, as long as the users involved are happy that there is a
de-facto standard.

- Peter

Peter Constable

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

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:13 EDT