[aklug] Re: Xmonad (again) / Functional Programming

* Christopher Howard <christopher.howard@frigidcode.com> [110417 23:39]:
> On 04/17/2011 07:18 PM, Joshua J. Kugler wrote:
> > On Friday 15 April 2011, Christopher Howard elucidated thus:
> >> Anyway, I'll try to lay off blabbing about Xmonad for a while. Right
> >> now I'm going through a Haskell tutorial... don't know if I'll end up
> >> being a Haskell programmer, but the little I've learned so far has
> >> been quite interesting. I'm hoping that some of the ideas from this
> >> "functional" programming paradigm will help me overcome some of the
> >> challenges I face whenever I try to start a large programming
> >> project. Seems like I often get mired down in dealing with all the
> >> complexity in the interaction between the objects in the program as
> >> well as the order of how issues should be resolved.
> >
> > Here's my take on OO/functional/procdural from few years back. My views
> > haven't changed much, if any.
> >
> > http://jjncj.com/blog/archives/127-Objects-have-failed...maybe...kind-of.html
  And I quote from that:
  """
  Objects have "failed" in that they've tried to be the be-all,
  end-all solution to all problems, while not being powerful enough
  to be the be-all, end-al
  """
  You don't need the be-all, end-all to deliver the product. The
  results are what count. I use objects in python only when
  I deem them necessary.
> >
> > j
> >
>
> Does Python support tail call optimization (that is, tail-recursive
> functions without a growing stack)? Pretty important, I believe, for
> serious functional programming. Apparently my Ruby 1.8 install doesn't
> have it, but I think it is available in Ruby 1.9.

  No. But there is a stackless python variant. I don't know if that
  would make TRF more workable. Scheme doesn't use the stack, but
  local GOTOs, if I recall correctly.
  
  BTW: There are some vigorous arguments in some sectors in which
  some argue that tail recursion is not necessary for functional
  programming. Newlisp does not use TRF and neither does rebol,
  although early versions of rebol (back in the 90's) did.
  Furthermore, I don't believe that CL (common lisp) does that much
  with TRF. And many lispers would argue that CL is the end-all,
  be-all when it commons to functional programming.
  
  I think that is hair-splitting. Again, it is the results that
  count, and in many cases what really matters is:
  can the coder deliver code that
  1)works
  and
  2)can be efficiently maintained by someone else?

  MTCW

-- 
Tim 
tim at johnsons-web dot com or akwebsoft dot com
http://www.akwebsoft.com
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.