On Friday, May 30, 2014 10:08:04 PM UTC+5:30, Mark Lawrence wrote:
> On 30/05/2014 17:15, Rustom Mody wrote:
> > On Friday, May 30, 2014 8:36:54 PM UTC+5:30, jmf wrote:
> > It is now about time that we stop taking ASCII seriously!!
> This can't happen in the Python world until there is a sensible approach
> to unicode. Ah, but wait a minute, the ball was set rolling with Python
> 3.0. Then came PEP 393 and the Flexible String Representation in Python
> 3.3 and some strings came down in size by a factor of 75% and in most
> cases it was faster. Just what do some people want in life, jam on it?
I dont see that these two are related¹
You are talking about the infrastructure needed for writing unicode apps.
The language need not have non-ASCII lexemes for that
I am talking about something quite different.
Think for example of a German wanting to write "Gödel"
According to some conventions (s)he can write Goedel
But if that is forced just because of ASCII/US-104/what-have-u it would
Likewise I am talking about the fact that x≠y is prettier than x != y.²
In earlier times the former was not an option.
Today the latter is drawn from an effectively random subset of unicode
only for historical reasons and not anything technologically current.
¹ Ok very very distantly related maybe in the sense that since python is a
key part of modern linux system admin, and getting out of ASCII-jail needs
the infrastructure to work smoothly in the wider unicode world.
² And probably 100s of other such egs, some random sample of which I have