Totally agree with you. Lowest common denominators are such basic functionality that people would end up binding to a particular cloud interface anyway. That said, it should if nothing else, raise the profile of the above problem to others who may not be on this list. I think I'm going to rename my project to libcloud, too ;) Cheers, -Adrian jclouds On Sat, Jul 25, 2009 at 10:22 AM, Sam Johnston<samj@samj.net> wrote:

It seems though that they're not the only 'libcloud' in town - there's another libcloud effort underway at RedHat, complete with some interesting discussions that map across to what we're doing (e.g. EC2->REST).

My problem with the adapter approach is that there are significant impedance mismatches between APIs. They're a good way to get something going now but we should try to get these initiatives supporting OCCI as soon as possible to smooth the migration path... ultimately clients should be able to talk a common language rather than try their luck with translators.

Sam

On Sat, Jul 25, 2009 at 12:16 AM, Sam Johnston <samj@samj.net> wrote:

...

FYI, this is interesting stuff.

Sam

---------- Forwarded message ---------- From: Alex Polvi <alex@polvi.net> Date: Fri, Jul 24, 2009 at 11:32 PM Subject: libcloud: a unified (python) interface to the cloud

Hello,

I hope this finds you well! I wanted to share a project that I have been working on called libcloud: http://libcloud.org/

It creates a standard *client* library, written in pure python, for working with a variety of clouds. The project is still young, but if you are interested in getting on the leading edge of some cloud portability hacking, now is your chance!

Site: http://libcloud.org/ Mailing list: http://groups.google.com/group/libcloud Code: http://github.com/cloudkick/libcloud

Thanks,

-Alex

-- twitter.com/polvi

_______________________________________________ occi-wg mailing list occi-wg@ogf.org http://www.ogf.org/mailman/listinfo/occi-wg