[via Links ]:

Earlier this year at CodeCon, Ben Laurie showed me the work that he had done on adding capabilties to Perl. Now he's put up a post with a pointer to the code and a little bit of documentation. He originally got interested in the problem because he was interested in adding capabilities to Python, but that turned out to be harder than he thought. Along the way, he formed some conclusions about Python and security:

Also, it seems the Python devlopers aren’t really interested in capabilities (nor all that interested in security, it seems, since the restricted execution mode is not maintained).

I don't think it's quite as a bad as Ben thinks, since he and I had some conversations with some Python developers and those folks were definitely interested in capability support. Of course, quite a few of them were a bit Twisted.

CaPerl is an alternative approach to adding capabilities which involves compiling a capability enhanced version of the language into the regular language. As to the rationale for doing this in Perl:

So, I did this for Perl, on the basis that if you can secure Perl you can surely secure anything.

I'm curious to see whether making the code available has any impact on the uptake of these ideas. Perhaps there will be some impetus in the Perl or Python communities to pick up on these ideas. When I saw Ben at Mind Camp, I suggested to him that perhaps the most profitable place to seed these ideas is the Ruby community, given the momentum hype of Rails, and the relative openness of the Ruby community to non-mainstream ideas.

I'd love to be proven wrong.