opaque ids vs. natural keys

[Mark Canonical Ramm-Christensen]

> My understanding of the sprint in Oakland is that we'd not tackle
> nesting at this point, which means in theory we'd just need a simple
> attribute saying "contained": "lxc" for the unit itself. This seems a
> few order of magnitudes simpler than changing primary keys like that,
> but I may be missing what this is all about.
>

My understanding is almost the same, we do not need to tackle nesting at
this point --  but here is the small but critical difference -- my notes
suggest that we were specifically asked to make sure we create a namespace
that *will* support addressing nested containers.

And my understanding of the work in progress with the LXC instance broker
puts us in a good place to support nesting, so I don't want to
un-necessarily make nesting support harder.

--Mark Ramm
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/juju-dev/attachments/20130530/1f2586ac/attachment.html>