[Python-3000] How will unicode get used?

[Adam Olsen]

On 9/20/06, Guido van Rossum <guido at python.org> wrote:
> On 9/20/06, Adam Olsen <rhamph at gmail.com> wrote:
> > On 9/20/06, Guido van Rossum <guido at python.org> wrote:
> > > On 9/20/06, Adam Olsen <rhamph at gmail.com> wrote:
> > > > Wow, you really did mean code units.  In that case I'm very tempted to
> > > > support UTF-8, with byte indexing (which is what code units are in its
> > > > case).  It's ugly, but it technically works fine, and it's the de
> > > > facto standard on Linux.  No more ugly than UTF-16 code units IMO,
> > > > just more obvious.
> > >
> > > Who charged you with designing the string implementation?
> >
> > Last I checked, the point of mailing lists such as these was to allow
> > input from the community at large.
> >
> > In any case, my reaction was simply because I misunderstood your intentions.
>
> I was specifically reacting to your use of the phrasing "I'm very
> tempted to support UTF-8"; this wording suggests that it would be your
> choice to make. I could have pointed out the obvious (that equating
> the difficulty of using UTF-8 with that of using UTF-16 doesn't make
> it so) but I figured the other readers are also tired of your attempts
> to move this into an entirely different direction, and based on a
> thorough lack of understanding of the status quo no less.

It was poor wording then.  I never intended to imply that it was my
choice.  Instead, I was referring to the input I have as a member of
the community.

I am not attempting to move this in a different direction.  I (and
apparently several other people) thought it always was a different
direction.  It is obvious now that it wasn't your intent to use code
points, and I can accept that code units are the best (most efficient)
choice.

-- 
Adam Olsen, aka Rhamphoryncus