<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<TITLE>Message</TITLE>
<META content="MSHTML 6.00.5346.5" name=GENERATOR></HEAD>
<BODY><!-- Converted from text/plain format -->
<P><FONT size=2>A brief word from the ignorant who started this whole thing by
sending out the original message on behalf of ISO 639 Ras-JAC:<BR><BR>ISO 639
registers <U>English</U> reference name(s), <U>French</U> reference name(s), and
"indigenous" reference name(s) of languages. However, the objects of
standardization are (1) the item as such, i.e. which language it is, and (2) the
language identifier. The spelling of names in English and French has not been a
primary concern (nor will it be, even after this discussion).</FONT></P>
<P><FONT size=2>The registration of indigenous name is not always absolutely
relyable in our tables. A number of spellings and transcription systems have
been used, and there are a number of inconsistencies. As a linguist, I don't
like that. But the spelling <U>is not</U> the object of standardization, and it
isn't possible to allocate the time needed to do that particular part perfectly.
Unfortunately.</FONT></P>
<P><FONT size=2>The English reference names are <U>English</U>. They don't
attempt, and shouldn't attempt, to represent indigenous forms, character sets or
scripts. As far as I know, the English language doesn't have e.g. clicks. And in
an <U>English</U> word like "N'Ko" there aren't really a whole lot of dumb and
smart apostrophes to chose between. There is just an apostrophe. That is
what there is in the English reference name column of ISO 639.</FONT></P>
<P><FONT size=2>The language "<FONT size=2>Gwich'in" was mentioned a while ago
in the discussion. It is correct that on the web site (Library of Congress) the
character ´ (acute accent over "air") is used. But in the standard itself (ISO
639-2:1998) there is a plain and simple English apostrophe. May be the form
"Gwich´in" on the web site is "wrong"; but it isn't intended to be
<U>different</U> from "Gwich'in". If this is a problem from retrieval point of
view, it is the search engine that is too "smart".</FONT></FONT></P>
<P><FONT size=2><FONT size=2>I am not at all opposed to enhancing the quality of
the representation of names in ISO 639. We should aim for a system where names
in all kinds of languages are available, and where the indigenous names have the
highest quality. But we don't get there by increasing the number of
different apostrophes in the <U>English</U> language.</FONT></FONT></P>
<P><FONT size=2>Regards,<BR>Håvard</FONT></P>
<P><FONT size=2>-------------------------<BR>Håvard Hjulstad
<A href="mailto:havard@hjulstad.com">mailto:havard@hjulstad.com</A> &
<A href="mailto:hhj@standard.no">mailto:hhj@standard.no</A>
<BR>-------------------------<BR>all outgoing mail is scanned using Norton
AntiVirus<BR></P></FONT></BODY></HTML>