Re: Definition of ``codel'' + ``execution engine'' ...
Herman Bruyninckx <[email protected]>
| Newsgroups | gmane.science.robotics.orocos.user |
|---|---|
| Message-ID | <Pine.LNX.4.44.0202081636250.19205-100000@pc25-213.mech.kuleuven.ac.be> |
On Fri, 8 Feb 2002, Anthony Mallet wrote: > Herman Bruyninckx wrote : > | But I thought it was possible for a codel to use > | a system wide lock, for example to interface some hardware? Otherwise, > | if a codel cannot do that, which component can? The device driver? > > Yes! Okay, that's a good distinction! That means that a codel is (should be) completely portable? Hence, belongs completely to the framework, and not to the ``device interface''? > | Ok, but what about `robotic tasks'? The things we want our robot to > | execute are seldom called `services' :-) But I agree that the > | codel/execution engine concept is much more general than the robotics > | context, so I use `service'. > > I think that robotic tasks are typically made of several services. Fine with me. I also prefer to begin to talk about services more and more. Herman