Re: [Evolution-hackers] misc. evo. requirements ...

[Andrew Cowie <andrew operationaldynamics com au>]

On Sat, 2003-06-28 at 08:22, Ettore Perazzoli wrote:
> In my current vision of how things will work activating the shell will
> be very quick (just a round trip to bonobo-activation to get the list of
> available components, without the need to activate them until they are
> needed).

Wouldn't that cause long pauses when you DO need them? If you can
achieve getting the shell up quickly (that would be great) then wouldn't
it make sense to load the other things in the background (separate
threads?)

Chances are the user won't notice or care if its chugging away a bit
behind the scenes so long as the interface is up. If they're too fast
and want a component that's not loaded yet, then fine, they'll have to
wait, but isn't that better than the alternative of long waits all over
the place as one by one deferred components load?

AfC

-- 
Andrew Frederick Cowie
Operational Dynamics Consulting Pty Ltd

Australia +61 2 9977 6866  North America +1 646 270 5376

andrew operationaldynamics com au