[Python-3000] Conservative Defaults (was: Re: Support for PEP 3131)

[Stephen J. Turnbull]

BJörn Lindqvist writes:

 > > On the other hand, if "anything from *any* script" becomes the
 > > default, even on a single widespread distribution, then the community
 > > starts to splinter in a new way.  It starts to separate between people
 > > who distribute source code (generally ASCII) and people who are
 > > effectively distributing binaries (not for human end-users to read).
 > 
 > That is FUD.

Not entirely.  XEmacs has found it appropriate to divide its
approximation to a standard library into "no-MULE" and "MULE-required"
groups of packages (~= Python modules).  GNU Emacs did not, and
suffered a lot of internal dissension for their decision to impose
MULE on all users.  Interestingly, they use no non-ASCII identifiers
that I know of.  (edict.el is not included in GNU Emacs due to an
assignment refusenik among the principal authors.)  The technology has
advanced dramatically since then, but there is real precedent for
balkanization.

The phrase "effectively distributing binaries (not for human end-users
to read)" is over the top, though.  Of course they're for end users to
read, they still are Python source, etc.

 > Answer what question? If people will use the feature? Ofcourse they
 > won't if it isn't default.

I assure you, my students will if it is available to my knowledge.

 > If you set different native languages won't you get the exact same
 > problems that codepages caused and that unicode was invented to solve?

No.  There is no confusion of character identity.  This is a perfectly
legitimate way to support Unicode, as long the subset of Unicode that
is allowed is properly declared.  It does not violate the principles
of Unicode in any way.