- From: <marja.j.phipps@gd-is.com>
- Date: Thu, 4 Oct 2001 12:23:22 -0500
- To: www-ws@w3.org
... It's possible that there is always in principle a "correct" level of description from which all other representations can be derived, but (a) I doubt it; and (b) we seem to do fine with coarse-grained representations. My conclusion from these speculations is that egocentric representations are the obvious way to go until someone comes up with an important application of neutral representations, or shows that for some central special case (such as web services) you can derive the egocentric representations for all participants from a neutral representation. ... Why do the egocentric representations need to be derived from the neutral representation? It would be easier to abstract the coarse-grained neutral representations from the more fine-grained egocentric representations. Perhaps I am missing your point, but I believe that multiple abstractions will need to exist in almost any scenario (for the sake of re-usability and extensibility, if not for the process itself), and that multiple perspectives are nothing more than scoping the view to an appropriate subset of the abstractions. A domain which requires neutral representation is the military, where process participants form a hierarchy. For example, when "executing a mission", the commander will monitor the process(es) from a coarse-grained perspective. Each of the officers "carrying out parts of the mission", however, will need to monitor their more-detailed respective process(es) from the perspective of their role in "executing the mission". ... and on down to the actual warfighters ... If a participant "encounters a problem", he/she has the option of reasoning over the solution space from their perspective or 'drilling-down' to a more fine-grained view of the problem space. I.e. A given participant's perspective may need to dynamically change. Obviously, the egocentric representations have to exist and they need to map into the neutral representation. Marja
Received on Thursday, 4 October 2001 13:24:54 UTC