--LwW0XdcUbUexiWVK
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sat, Nov 08, 2003 at 12:31:09PM +0000, Malcolm Parsons wrote:
>=20
> I've attached debian bug report #219666, and an email from Joseph Myers
> asking questions about the correct way too fix it.
>=20
> What does NetBSD think is the correct fix for this bug?

I think the correct thing is to leave things alone, for two reasons.

1) NetBSD has chosen to use American English as its language, so the=20
current use of, "english" by this program is fine. This patch is=20
attempting to add an optional localization to the tool, but it is doing it=
=20
in a manner that will not scale. If we want to support different=20
translations, then we really should cope with the fact that there probably=
=20
are more than just two.

2) The use of, '-e', to enable the alternate behavior is inappropriate. =20
"English" is the name of a language with dialects, it is not a dialect. =20
Thus it is inappropriate to use it to choose between different dialects of
itself. "Queen's English" would work (for a few more decades), but I think
a better choice would be something like, '-b', for "British."


If we really want to add this, then I'd say use '-b' for British. The=20
problem though is what happens when a different English dialect has a=20
different number need? The only thing I see that'd scale would be some=20
sort of tie-in with localization.

Take care,

Bill

--LwW0XdcUbUexiWVK
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQE/s/5IWz+3JHUci9cRAupoAJ9uFYHO7q80MbAu6HUxlUehjBACKwCcDHyg
FpRkCMB25d/8gtqsqf4mAfc=
=fdBF
-----END PGP SIGNATURE-----

--LwW0XdcUbUexiWVK--