I would like to discuss the driver SPI. It is a bit too EC2ish (I think):
the images call is
images(credentials,ids)
where ids can be either a ownerid or an image id.
For the RHEVM driver, I need to know which I am looking at. Can we either make this a hashmap or a
images (credentials, owner=null, id=null)
so that I can do different things based on the content? I think the former is better because you can then make calls such as
images (creds, :id => "Foo") or images (creds, :owner => "Bar")
And it will be clear on both sides.
-- bk