On Sat, 31 Jan 2004, Paul Bennett wrote:

> On 28 Jan 2004 at 10:14, Tristan McLeay wrote:
> > On Wed, 28 Jan 2004, Andreas Johansson wrote:
> >
> > > Quoting Tristan McLeay <[log in to unmask]>:
> > >
> > > > Given all the problems with /a_2/-for-diphthong, I think it almost
> > > > make more sense to remove it as an option in CXS.
> > >
> > > I rather agree.
> >
> > I don't like changing the standard once it's been made
> Feh. A standard is only immutable once it's accepted and in regular
> use, and has stopped naturally fluctuating. Up until that point it's
> just a plaything.

The problem, though, is that I don't know how many people are using the
CXS image on their webpage. I know of at least two (Andreas and Pablo
Flores) in addition to my two, but google only shows up one of mine with a
google image search for 'cxs ipa'. Perhaps version numbers would help, but
then we'd all have to say 'CXS 2', and it's hard enough convincing people
to call it 'CXS' and not 'X-Sampa' (when referring to CXS and not X-Sampa)
that that'll just confuse people even more...

However, if the changes are used and minor enough, I might add them to the
chart to keep the chart in line with usage, but if they aren't, I'm not
going to start them (Andreas: this is your cue to start a conlang that
distinguishes /4`/ and /r`/!). The only artificial change on the chart was
the creation of [&\] for X-Sampa [&], because otherwise there'd be a whole
(I'd never seen anyone using a low front rounded vowel while talking
(proto-)CXS). On the other hand, Andreas could just change his chart and
try and get it to become the dominant one :)

> Version numbers would help, until the first schism,

CXS _itself_ is a schism :)