More on Protecting Source Code

[Frithiof Andreas Jensen]

"David LeBlanc" <whisper at oz.net> wrote in message
news:mailman.1032232999.22661.python-list at python.org...

> * Why invest a substantial amount of time and money developing in a
language
> that makes it trivial to gain access to the work product?
>

Why indeed - why do you bother?

Why is existing Copyright law not sufficient for you?

> beyond those that
> are fairly closely licensed and/or have substantial parts of the app
written
> in C/C++.

Do that then - there is nothing intrinsic in the Python licensing that
restricts what you can do with it; I seem to recall that you can pretty much
do whatever you bloody like, including wrapping Python in your own product
and selling it!

> A better solution is to make a .pyc file approximately as hard as a binary
> .exe file to decompile - however that could be done.

And according to the generous license awarded to you by the very people
whoms work you are critisizing, *you* can indeed do that; you are in fact
*encouraged* to make you own enhancements to Python and release them.

Simple Darwinism will show whether others feels the same way.

One could, in theory, package the application with a custom build of the
Python interpreter so that each interpreter uses it's own unique set of
bytecodes. Performance will likely suck and you will have a real issue with
upgrades, extension modules and maintenance of the many unique applications
shipped - but again this is your choice to make.

Personally, I do not care - I like Python exactly the way it is, including
the "openess", since that is an essential part of simplifying the
development task. If I was really, really worried about some "IP-leakage" I
would stick the core algorithm in an extension library - that would not stop
someone from just ripping the software though.