When will Python 3 be fully deployed

[John Nagle]

You'll have some answers here: http://jessenoller.com/2009/12/04/pythons-moratorium-lets-think-about-this/

I'd argue against using Python 2.6 for production work. Either use Python
2.5, which is stable, or 3.x, which is bleeding-edge. 2.6 has some of the
features of Python 3.x, but not all of them, and is neither fish nor fowl
as a result. 2.6 is really more of a sideline that was used for trying
out new features, not something suitable for production.

I think the idea is to run your 2.5 code through '2to3" and see if it
works in 3.x.

Anyway, it will be years, if ever, before Python 3.x gets any real support.
Too many major packages still aren't fully supported on it, and some popular
packages, like SGMLlib and Feedparser, are being dropped.

When a few major Linux distros ship with Python 3.x and enough of the binary
packages to run a web site, take a look at it again.

John Nagle

 

[Ned Deily]

I'd argue against using Python 2.6 for production work. Either use
Python
2.5, which is stable, or 3.x, which is bleeding-edge. 2.6 has some of the
features of Python 3.x, but not all of them, and is neither fish nor fowl
as a result. 2.6 is really more of a sideline that was used for trying
out new features, not something suitable for production.

I disagree with that advice, strongly. 2.6 not only has new features
but it has many bug fixes that have not and will not be applied to 2.5.
It is hardly a sideline.

See http://www.python.org/download/releases/2.5.4/ for the official
policy on 2.5, in particular:

"Future releases of Python 2.5 [ -- that is, should the need arise -- ]
will only contain security patches; no new features are being added, and
no 'regular' bugs will be fixed anymore."

"If you want the latest production version of Python, use Python 2.6.1
or later." [2.6.4 is the latest version].

Then see http://www.python.org/download/releases/2.6.4/

Note that Python 2.6 is considered the stable version and is "now in
bugfix-only mode; no new features are being added". Per normal python
development policy, new features are added to the next major release
cycles, now under development: Python 2.7 and Python 3.2.

 

[Gabriel Genellina]

I'd argue against using Python 2.6 for production work. Either use
Python 2.5, which is stable, or 3.x, which is bleeding-edge. 2.6 has
some of the
features of Python 3.x, but not all of them, and is neither fish nor fowl
as a result. 2.6 is really more of a sideline that was used for trying
out new features, not something suitable for production.

I think the idea is to run your 2.5 code through '2to3" and see if it
works in 3.x.

In addition to Ned Deily's previous comments, I'd like to note that 2to3
assumes the source is valid 2.6 code - you have to ensure the code runs
fine with Python 2.6 before using 2to3 to convert to 3.x

 

[Martin v. Loewis]

In addition to Ned Deily's previous comments, I'd like to note that 2to3

assumes the source is valid 2.6 code - you have to ensure the code runs
fine with Python 2.6 before using 2to3 to convert to 3.x

That's wrong - 2to3 works just fine on, say, 2.3 code that has never
been run on 2.6.

Regards,
Martin

 

[Aahz]

I disagree with that advice, strongly. 2.6 not only has new features
but it has many bug fixes that have not and will not be applied to 2.5.
It is hardly a sideline.

Ditto -- we had some webserver crashes that were fixed by upgrading from
2.4 to 2.6 (we were already using 2.6 in the client and decided that
skipping 2.5 on the server was best).

 

[Roy Smith]

Looking back over the years, after I learned Python I realized that I
never really had enjoyed programming before.

That's a sad commentary. Python is fun to use, but surely there are other
ways you can enjoy programming?

The first thing I learned how to program was an HP-9810
(http://www.hpmuseum.org/hp9810.htm). I had LOADS of fun with that. Then
I learned BASIC (using my high school's ASR-33 hookup to a HP-3000 a couple
of towns away). Lots of fun there too.

Then came Fortran. I guess I had fun with that, at least in the beginning.
I did a bunch of assembler. Some of it was fun (pdp-11, 6800), some of it
was not (pdp-10, IBM-1130). Lisp was fun for a while, but I never really
got into it.

C was fun at the beginning, but quickly became a drag. C++ was was evil
and horrible at the beginning. As opposed to now, when I'm somewhat of an
expert in it, and it's still evil and horrible.

Learning PostScript was blast! One of the true epiphanies of my
programming career was hooking a video terminal up to the RS-232 port on an
Apple LaserWriter, typing a few lines of PostScript at it, and watching a
page come out with a square drawn on it.

Everybody should learn PostScript. People think of it as just some
document printing thing, but it's a real (Turing-complete) programming
language. Not just that, but it's a fun language to learn, and lets you
explore some corners of the language design space which most people never
see. Go forth and learn PostScript!

Learning Java was about as much fun as kissing your sister.

I'm sure I've left a few out, but the point is there are plenty of ways to
have fun programming besides Python.

 

[Aahz]

That's a sad commentary. Python is fun to use, but surely there are other
ways you can enjoy programming?

Not really. I've been programming more than thirty years, and the
closest I came previously to enjoying programming was Turbo Pascal, and
even that has too much tedium and lack of brain-fit.

Before Turbo Pascal, there was BASIC on an HP-1000. Afterward came
HP-41, Ada, FORTRAN, Paradox PAL, C, Perl, and there must be some others
I'm forgetting. Thankfully, I didn't learn Java until after I'd been
programming in Python for a while. (And arguably I still haven't learned
Java despite writing a PGP encryption wrapper around BouncyCastle.)

Programming is difficult to begin with, and everything other than Python
just gets in my way.

To be fair, my quote isn't entirely honest: I never called myself a
programmer before I learned Python because I didn't really like it. It
took Python to make me realize that programming *could* be fun, or at
least not annoying enough to keep me from making a career of programming.