Michael Everson everson at evertype.com
Mon Jun 16 02:21:44 CEST 2003


>I am a little surprised, since I thought tags were supposed to identify
>existing practices,

Oxford spelling is an existing practice. Tagging English-language 
text for this spelling is not presently possible, because there is no 

>not be the first step in defining a practice (or practise)

"Practise" is a verb. American spelling advice/advise stands in the 
same noun/verb relation to British & Oxford practice/practise, though 
there is no pronuncuation difference in the practice/practise pair, 
as there is in the advice/advise pair.

>that spell-checkers can follow. Isn't that the argument by which es-americas
>died (or is dieing)?

No. I rejected es-americas because it was not demonstrated that there 
is a single entity that can be subsumed under es-americas. My 
argument was that there was a great deal of variation, for instance 
in the second-person singular and plural pronouns, in "American 
Spanish" which did not, to me, suggest any kind of unity that could 
be subsumed under a single tag.

>Do we have examples that follow OED, other than the OED itself?

Oxford University Press publishes any number of English-language 
dictionaries which follow this orthographic practice. It is a 
long-standing editorial practice. An English-Cornish dictionary for 
which I was editor and which was not published by Oxford conformed to 
Oxford spelling. :-)

>There are many other rules differences between british, american, spellings
>and probably if we look to canada, south africa, and australia a few more.

Perhaps, but one would like to see some evidence. It is certainly the 
case that "center, color, civilize" is "US English" in software and 
that "centre, colour, civilise" is "British English" in software and 
that one cannot get software which supports OED practice with either 
of those.

>I am not sure we need tags to distinguish these. Besides the spellings, there
>are usages differences, so the terms need to change as well.

The tag is proposed to support orthographic practice, not other usage.

>I am not sure that producers of software nor users of software want 
>to select localization based on a preferred combination of spelling 
>rules alone.

I want to.

>I can see that users might want to select a spell checker on that 
>basis, but I question whether they need a tag for that. It could 
>just as well be the role of configuring the checker without 
>requiring a tag at all.

I can think of no spell-checker which can be so "configured". 
Spell-check dictionaries are single compressed-wordlist files which 
cannot be changed by the end user. Yes, we can add words to 
user-dictionaries, but regarding the -ise/-ize suffix which is very 
productive there is no option which the user can choose. And since 
the suffix is productive it is a real pain to require users to have 
to add their words to such a user dictionary

>Also, what happens if OED changes its preferrences over time and as the
>language evolves?

Oxford spelling preference has been constant since *at least* 1933. 
It is a principled etymological spelling, with regard to the suffix 
derived from the Greek -izein suffix.

>I am sure some of their preferences must have changed over the 3 editions...

Prove it. :-)

>How does that affect the tag? Which rules preferences are important 
>to the tag, and which not?

The facts are as I have stated them.
Michael Everson * * Everson Typography *  * http://www.evertype.com

More information about the Ietf-languages mailing list