Embedding Python at the OS level - build problems [success]

[Ryan Paul]

On Wed, 26 May 2004 14:13:07 -0400, Phil Frost wrote:

> Success! They said it can't be done. "Python is too big", they said.
> "Too high level for an OS". And yet, it's done :)
> 

congratulations! truly a unique and innovative endeavor. I can't wait to
see how you use it. I'm quite curious about how you have bound python to
the operating system: i'm particularly interested in what specifically
this lets you do, that you couldnt otherwise do on a platform where python
is just a command-line accessible application. On your web site you say
that your OS is very modular (this conjures up thoughts about
microkernels) Projects like the Hurd facilitate construction of userspace
kernel extensions; can your OS allow for something similar, but with
Python? It would be neat to see a compact and efficient OS designed to
natively support things like a python-extendable VFS, or kernel level net
services that support scripted extensions.

I've experimented with the hurd, qnx, and a few others, and all of them
support graceful and efficient userspace kernel extension- something that
linux's monolithic kernel regrettably precludes. I think that pluggable
micro-kernels are a wonderful technology, and I love studying
implementations- any information you can provide about the model of your
OS, particularly the embedded python components, would be greatly
appreciated. I havent learned assembly yet, so looking at your code
probably wouldnt be very useful for me, and your web site doesnt have the
kind of technical details i'm interested in. If you have some time, i'd
really like to hear more about your project!

thanks,
--SegPhault