[Python-3000] educational aspects of Python 3000

Raymond Hettinger rhettinger at ewtllc.com
Tue Sep 12 23:58:19 CEST 2006


Ron Adam wrote:

>Maybe "input" can be depreciated in 2.x with a messages to use eval(raw_input()) 
>instead.  That would limit some of the confusion.
>
>  
>

Let me take this opportunity to articulate a principle that I hope this 
group will adopt, "Thou shalt not muck-up Py2.x in the name of Py3k."

Given that Py3k will not be backwards compatible in many ways, we may 
expect that tons of code will remain in the 2.x world and it behooves us 
not to burden that massive codebase with Py3k oriented deprecations, 
warnings, etc.  It's okay to backport compatible feature additions and I 
expect that a number of people will author third-party transition tools, 
but let's not gum-up the current, wildly successful strain of Python.  
Expect that 2.x will continue to live side-by-side with Py3k for a long 
time.  It is a bit premature to read the will and auction-off the estate ;-)

Any ideas for Py3k that are part of the natural evolution of the 2.x 
series can of course be done in parallel, but each 2.x proposal needs to 
be evaluated on its own merits.  IOW, "limiting 2.x vs 3k confusion" is 
NOT a sufficient reason to change 2.x.


Raymond






More information about the Python-3000 mailing list