a different question: can you earn a living with *just* python?

[sjdevnull at yahoo.com]

Paul Rubin wrote:
> "sjdevnull at yahoo.com" <sjdevnull at yahoo.com> writes:
> > > > * C
> > > > * A static functional language (ML, Haskell, etc)
> > > > * Lisp or scheme Scheme
> > > > * A static class-oriented language (Java, C++, etc)
> > > > * A dynamic OO language (Python, ruby, smalltalk, etc)
> > > >
> > > > and at least a brief look at, say, Forth and Prolog.
> > >
> > > Interesting list.  Of those, I've done tons of C, just enough lisp to get
> > > the feel of it, lots of C++, and of course Python.  I've never done any
> > > functional stuff.
> >
> > You should.  It's very enlightening.
>
> Very interesting post and list.  I think I'd add at least one assembly
> language.

Yes, definitely.

>  I hate to say it but I think I'd remove Python.  As much as
> Python has helped me get useful and practical things done, from a
> learning point of it, as much as the developers deny it, I'd say it's
> basically an OO Lisp dialect with syntax sugar.  I found it completely
> natural and pleasant to program in almost immediately, because I'd
> already been using Lisp and Java.  I haven't used Smalltalk or Ruby so
> can't comment.

OO programming in dynamic languages can be _very_ different from in
non-OO languages.  I'm hesitant to put CLOS on the list as a lot of the
mind-expanding value of Lisp can/should be achieved without it and I'd
rather not confuse a person learning a couple of languages by having
them try to keep OO-styled Lisp and traditional-style Listp straight.

Hence I think Smalltalk or python/ruby are better choices for learning
dynamic OO programming.

> I wonder why you chose ML over Haskell in a few other posts.

Personal bias, my university was more ML-centric than Haskell.  At
least I didn't show an even bigger bias by putting Dylan on the list.
:-)

Either ML or Haskell is a fine choice for learning strongly statically
typed functional programming.