Python for large projects

Dave Brueck dave at pythonapocrypha.com
Tue Mar 23 10:44:23 EST 2004


Piet wrote:
> >>>>> Jacek Generowicz <jacek.generowicz at cern.ch> (JG) wrote:
>
> JG> I feel honour-bound to point out that citing static typing (explicit
> JG> static typing, in particular) as a means of creating more correct
> JG> programs, to be one of the greatest contemporary myths of software
> JG> engineering.
>
> JG> Static typing makes it easier for compilers to produce more efficient
> JG> code. That is the only advantage that static typing offers.
>
> JG> I am of the opinion that (explicit) static typing contributes to the
> JG> bugginess of programs.
>
> I would like to know if there is scientific evidence (rather than
> feelings/opinions) about the pro or con of this statement.

There probably is scientific evidence of this, but nobody has paid the price to
gather it yet. :)

Despite that, there is quite a  bit of anectdotal evidence to generally support
the ideas Jacek put forth. While not as "hard" as something derived from an
official study, it can still be useful in the decision-making process. It's
interesting to note that for the opposing view there doesn't seem to be much in
the way of scientific OR anectdotal evidence.

I'm not familiar enough with some of the more modern languages that use
explicit static typing to know if they offer other advantages beyond helping
the compiler be more efficient, but from experience I don't hesitate to say
that the explicit type systems of Pascal/C++/Java seem to do more harm than
good (meaning that if they provide some benefit, overall it's still a net
negative due to the extra burden they place on the developers). For large
projects in particular they seem to get in the way more and more as the project
increases in size.

-Dave





More information about the Python-list mailing list