- From: Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>
- Date: Mon, 7 Nov 2011 23:59:04 +0000
- To: Satya Sahoo <satya.sahoo@case.edu>
- Cc: Luc Moreau <L.Moreau@ecs.soton.ac.uk>, Provenance Working Group WG <public-prov-wg@w3.org>
On Mon, Nov 7, 2011 at 23:00, Satya Sahoo <satya.sahoo@case.edu> wrote: > Any software agent will be always be a process. It is up to the application > to model it as an agent (Email client, virus etc.) or a process (as in > activity) - similar to "one person's metadata is another person's data"... A plan can also be intended for humans. If you want to model me as a process, that's fine, but perhaps a bit unintuitive at first. As I'm about to unpack and assemble my IKEA book case, I might collect the provenance for this. In this case the assembled piece of furniture has gone through various process executions such as "unpacking", "aligning", "screwing", etc. - these are all executed by me (the agent, in the role of "assembler"), somewhat according to the assembly instructions (the plan). However I might find that I don't agree with the plan, so it's not just "interpreted". On a higher-level picture you can imagine an Assembly process execution which controls all these smaller operations (lacking any other way to express subprocesses), and uses both myself (role="worker"), the assembly instructions (role="instructions") and the packed box of furniture parts (role="parts"). -- Stian Soiland-Reyes, myGrid team School of Computer Science The University of Manchester
Received on Tuesday, 8 November 2011 00:00:04 UTC