Design-by-Committee

[Andrew Dalke]

Michael Hudson wrote:
>I'm not Thomas (obviously) but I can easily think of places you might
>use Python where you don't need unicode.  Scientific computing and
>sysadmin-style duct tape are just two that spring to mind almost
>immediately.

I suppose if you are strictly refering to numeric scientific
computing, then that is true.  I work with a lot of biology
databases and I want them to migrate to Unicode to reduce the
ugliness of not being able to spell things correctly.  That
includes people's names (Monta~no), chemical compounds (H2O),
peptide names (<beta>-defensins), measurements ("wavelength
<lambda> = 366 nm" or "2 <micro>M concentration " or "37<degree>"),
atomic weights (U238), species name ("bacteriophage <phi>12345"),
and many more.

(Some of these, like subscripts, are not properly part of
Unicode, although I think some of the most common are.)

                    Andrew
                    dalke at acm.org