Python Productivity over C++

[Frank V. Castellucci]

Aahz Maruch wrote:
> 
> In article <8hqbvr$fgr$1 at sunnews.cern.ch>,
> Wim Lavrijsen <wlav at hpatl26.cern.ch> wrote:
> >
> >First off, any claim of an order of magnitude increase in productivity
> >is pure nonsense. See Brooks for an explanation or Boehms and McConnell
> >for hard data.
> 
> That's half-true.  Lines of code per unit time is roughly constant per
> programmer (though I think Python breaks even that rule).  Of more
> interest is the extent to which one language has greater expressiveness
> than another for a given domain.  So if Python programs are generally
> one-tenth the size of C/C++ programs, there *is* an order of magnitude
> increase in productivity.

I would be careful with that bat. The more libraries you use, the more
editors that provide expansion and completion, the more generator type
tools available increase the LOC but reduce the time spent at the
keyboard. For any language.

Productivity is not measured by LOC alone, it needs to be measured by
QLOC (Quality lines of code). This is easily tracked with most revision
control systems. If developer A spends ten minutes putting out crap
code, then developer be must come along and spend even more time
debugging and correcting those "productive" lines of code.
 
> Where I think Python really kicks butt is in maintenance, though.
> Reading Python code is significantly easier than any other language I've
> seen; rewriting Python code is also significantly easier.

Agreed.

-- 
Frank V. Castellucci