Re: Redesign of QEMU startup & initial configuration

[Daniel P . Berrangé]

On Tue, Dec 14, 2021 at 03:42:52PM +0100, Mark Burton wrote:
> I think we’re talking at cross purposes, and probably we agree (not sure). 
> I’ll top quote to try and explain my point of view.
> 
> I think there are two discussions being mixed:
> 1/ A discussion about improving the CLI. (Having a new one, etc etc)
> 2/ A discussion about supporting a low level, and complete, API that can be 
> used by “management layers” (QAPI).
> 
> I think this also gets mixed up with the discussion on migrating the CLI to 
> use the low level API.
> 
> I want to focus on the low level API. 
> 
> I dont see why we’re discussing a ‘high level’ thing when, for now, we have 
> to support the CLI, and we have work to do on QAPI.

We're discussing both because we're setting out what our end goal is
to be, and that end goal should be expected to cover both use cases.

> If somebody wants to build a new CLI, with a new ‘high level’
> interface, using QAPI - let them!

This is too weak of a statement, as it implies that a replacement
high level interface is optional and not important for the overall
project. I don't believe that to be the case, so I'm saying that
our design & impl plan has to demonstrate how we intend to cover
both deliverables or use cases. We can't simply ignore the high
level API saying it is someone else's problem to worry about.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|