Wednesday, April 12, 2006
Anticipatory Web discussionà [home]
Anticipatory Web Challenge Problem à [link]
Community Centric Service Methodology Glass Bead Games
In response to [6]
Paul,
In your note, you say:
I am conjecturing that two problems are left to the
"pragmatics" of specific
deployments.
1) The development of a conceptualization of the
business (or transaction
space) by human (non-IT) stakeholders.
2) The aggregation of a number of such
conceptualizations into a common
model of a ecosystem (of business entities and interactions)
I equate these problems to the following challenges:
1.
Establish a model of managed resources (e.g. network, servers,
applications, systems) and their relationships, with common semantics across
the different resource types (or be able to map between existing ones). Use
this model to gather and exchange the necessary information between components
of an autonomic system.
2.
Use the basic vocabulary in (1) to construct higher-level
concepts, like desired changes to the resources; desired conditions to be
maintained; required actions to deploy resources; detected situations and what
to do about them. Use these to construct autonomic managers that can monitor
and perform actions on the resources.
3.
Figure out how all of the above can be directed/influenced so
that it achieves the overall objectives of its human stakeholders - whether
this is via formalizing those objectives in a way that can be used by the
autonomic managers, or by actively involving humans in monitoring and decision-making.
If that's what you meant, yes, those challenges exist and need to be solved. There may be different solutions in disjoint domains, but where there is a need to build a set of integratable components that can be adapted to a range of scenarios (e.g. multi-vendor management scenario in an IT organization).
I don't see that they can be left to the pragmatics of specific deployments.
Regards,
Christine