[Spry-dev] A crisis of faith

[Mark Hahn]

I have been using C# a lot since I have been implementing Spry (was PyCS) in
it.  To my surprise, I have also been using it for quick hacks where I used
to use Python.  I find it as easy to program in as Python.

I have had a religious belief in dynamic languages for some time.  I
originally did Prothon with prototypes instead of classes because prototypes
are inherently more dynamic.  My dissapointment with prototypes and the
availability of the .NET library to dynamic languages caused me to switch to
developing Spry.  Spry is basicly a more dynamic version of IronPython.  The
way it is designed it will allow many of the dynamic advantages of
prototypes even though it is class-based.

But now I find to my surprise that C# is good for "script-like" programming.
I think the reason is that the availability of the large library is what
made Python so good for scripting (I use scripting to mean rapid
programming, no matter what the goal), not the dynamic features of the
language.

So my faith in dynamic languages is slipping badly.  I have stepped back
from the programming lately and taken a hard look.  Was Dylan a failure at
Apple because there is no real advantage to dynamic languages?

One advantage of Python is mitigated in C# because of the excellent
development tool Visual Studio.  The excessive amount of code required in C#
for all the namespace and variable declarations is less a problem when the
tool makes typing easy.  I find myself typing as few keystrokes to produce
the lengthy C# code as I did in Python.

Of course the real savings of coding in Python is the large existing library
to make coding easy.  But now the .NET library actually exceeds the scope of
the Python library.

Can someone please tell me why I should continue believing in dynamic
languages?