I strongly agree with the importance of this benefit. In my current work at Watkins-Johnson on semiconductor processing equipment software based on ControlWORKS, one of the major causes of thrashing has been the existence of some components which do not provide for adequate hardware simulation during development and debugging. Work on those modules that do provide support has been greatly aided by close to transparent simulation of I/O ports and hardware with the only changes necessary to switch between simulation and real world being a couple of changes to a configuration file. Actually, I would like it to go furthur and allow for switching in and out of simulation individual components on the fly to enable confidence testing prior to actual hardware action. As far as I know, ControlWORKS does not provide this capability.
I really hope that such transparency is the result of your effort.
Jerome
--
Jerome E. Garcia jegarcia@adventurousmind.com
______________________________ Reply Separator _________________________________ Subject: Re: Subsumption Architectures in Squeak Author: Dave Astels astels@audesi.com at INTERNET Date: 10/27/98 10:02 PM
Sam Adams wrote:
Dave Astels wrote: <<
<snip> I've started a project to build a version of Rod Brooks Behaviour language
I would also like to know more your subsumption interests and efforts.
I've started writing up an overview. Expect it in a couple of days.
Sounds like you're more aimed at embedded control, but a software environment to test your subsumption designs could be useful, too.
The benefit of a VM is that you could work in an entirely software environment with simulated hardware. You could tie into that by subclassing the VM and providing implementation for I/O ports. The hardware simulation could be anything from numeric oriented dialogs to full 3D, realtime visualizaton.
Dave
-- Dave Astels The people who are crazy enough Software Engineer to think they can change the world, AudeSi Technologies Inc. are the ones who do. astels@audesi.com (work) 01490312@3web.net (home)
squeak-dev@lists.squeakfoundation.org