Could Python supplant Java?

FISH joeking at merseymail.com
Thu Aug 22 17:16:34 CEST 2002


Peter Hansen <peter at engcorp.com> wrote in message news:<3D642A6D.A2F1E280 at engcorp.com>...
[snipped...]
> Assuming it is really only 1%, wouldn't you say that if the cost
> is 10%, it might not be worth it?  Or maybe it's more like 25%, for
> the extra typing and the extra thought required to lay it all out in
> advance.  Is it still worth that 1%?
> 
> I'd much rather save all the time I do by using Python, and get 
> much more done, including writing excellent unit tests (white box)
> and acceptance tests (black box), and take the time I saved to 
> solve 100% of my bugs (I have very few using TDD with Python) 
> rather than do a lot of extra make-work typing just to let the 
> compiler point out a few trivial errors to me, making up 1% of 
> the total...
> 
> -Peter


You haven't factored in the cost of tracking down the bugs when
they occur.  Sure, they may not be the most common of bugs, but
mis-casting or mis-typed data can often be really buggers to
track down, depending upon the source of the data that fed the
variables.  Especially when your language tries its hardest to 
adapt whatever's in the variable at the time into something 
suitable for the operation you are performing - meaning the 
error can be carried forward into other parts of the code, and 
the ultimate point of failure (the point at which the software 
actually gives up and dies) can be miles away from the actual 
error itself.

Unless you write lots and lots of checking code for every 
critical variable assignment... of course.  What was that you
said about saving time, money and effort...?  :-)


-FISH-   ><>



More information about the Python-list mailing list