The real problem with Python 3 - no business case for conversion (was "I strongly dislike Python 3")

[Kevin Walzer]

On 7/2/10 3:07 PM, John Nagle wrote:

>
> That's the real issue, not parentheses on the "print" statement.
> Where's the business case for moving to Python 3? It's not faster.
> It doesn't do anything you can't do in Python 2.6. There's no
> "killer app" for it. End of life for Python 2.x is many years away;
> most server Linux distros aren't even shipping with 2.6 yet. How can a
> business justify spending money on conversion to Python 3?

That's decision for each business to make. My guess is that many 
businesses won't upgrade for some time, until the major 
libraries/modules support Python 3. I don't plan to move to Python 3 for 
at least a couple of years.
>
> Python 3 is a nice cleanup of some legacy syntax issues. But
> that's just not enough. Perl 6 is a nice cleanup of Perl 5, and
> look how that went. Ten years on, it's not even mainstream, let
> alone dominant.

The Perl analogy isn't really useful here. Perl 6 is somewhere between 
the HURD and Duke Nukem Forever in terms of being viable. Even the Perl 
website says, "If you are looking for production ready code please use 
Perl 5." That's one reason why Perl 5 development has recently undergone 
a resurgence.

Python 3, by contrast, is production-ready in itself; libraries are 
gradually moving to support it, and Python 2 has a definite end-of-life 
release in 2.7, with an extended maintenance period for 2.7. The Python 
developers are providing a much stronger and clearer path forward for 
Python 3. The transition period may last five years, but the path is clear.

As a Mac developer, I'm sympathetic to your frustration. A few years ago 
Apple deprecated one of its major API's (Carbon), on which my own 
development depended, and there was a lot of uncertainty about major 
libraries that use Carbon being updated. This is normal in any 
transition period. Eventually, the major libraries I depend on were 
updated by their developers (i.e. ported to the Cocoa API), I was able 
to migrate my own applications to the updated libraries, and life went on.

I think the same thing will happen with Python. It's useful to note the 
libraries that are not yet ported to support Python 3, and to share best 
practices for moving forward. Past a certain point, however, I don't see 
much point in attacking the existence of Python 3 or questioning the 
need to move toward Python 3. It's here, it's the way forward, and 
that's not going to change. Might as well accept it.

--Kevin


-- 
Kevin Walzer
Code by Kevin
http://www.codebykevin.com