In my posting to the e-forum
http://colab.cim3.net/forum//soa-forum/2008-01/msg00004.html
I suggested:
“According to the second school viewpoint, any computing system is not Rosen-complex. In theory, Rosen-complexity can occur in hybrid computing/natural-systems if the architecture is asking for human choices at specific times. The CoreTalk architecture does this through a sophisticated interaction within delineated communities using agreements and contracts. “
The point is that formal systems, such as classical mathematics OR well-specified SOA, must suffer when there is some type of novelty in the current event structure. Crisis often has novelty and often the response to crisis requires a real time re-specification of the meaning of data structures.
This ability to allow real-time re-specification of the meaning of data is a challenge to the first school. Within the viewpoint of the second school, the re-specification must come within the context that emerges the moment the event is manifesting. The human must be able to add "Rosen-complexity" to the understanding of the event that is manifesting. This understanding can focus on the particulars of an event as experienced within that event, whereas universals (formal constructions pre-developed) may in fact not well represent meaning in context.
This understanding of the novelty, the uniqueness of the particular, of a event is easy for humans, because our brain systems and biology has evolved precisely so that an adaptation of understanding can occur in real-time.
We, humans, recognize the nature of the present moment, unlike what might be claimed to occur with artificial intelligence or in engineered knowledge management systems (ie in cognitive engineering).
If the SOA architectures now being designed and developed by contractors to e-Governance, do not account for the concerns related to service discovery and the orchestration of an aggregation of semantic primitives, then these architectures will predictably fail in times of crisis.
From specific theory, delineated in the second school paradigm, the failure will be traced to concurrency issues as many (thousands) of operations lock up transaction engines because of failures to complete pre-anticipated sequences of events (cascade failures). The human brain deals with this using both non-locality related to quantum-neuro physical mechanisms (see work by Hameroff and Penrose on self orchestrated collapse involved in those physical mechanisms supporting everyday human awareness) .
So far there are no SOA architectures being considered by e-gov program managers that deal with the nature of collective intelligence. The reasons for this absence can be traced to the first school viewpoint that a pre-specification of data meaning is both necessary and sufficient.
To quote from wiki
http://en.wikipedia.org/wiki/Transactional_memory
"In 2005, Tim Harris, Simon Marlow, Simon Peyton Jones, and Maurice Herlihy described an STM system built on Concurrent Haskell that enables arbitrary atomic operations to be composed into larger atomic operations, a useful concept impossible with lock-based programming. To quote the authors:
Perhaps the most fundamental objection [...] is that lock-based programs do not compose: correct fragments may fail when combined. For example, consider a hash table with thread-safe insert and delete operations. Now suppose that we want to delete one item A from table t1, and insert it into table t2; but the intermediate state (in which neither table contains the item) must not be visible to other threads. Unless the implementer of the hash table anticipates this need, there is simply no way to satisfy this requirement. [...] In short, operations that are individually correct (insert, delete) cannot be composed into larger correct operations.
—Tim Harris et al, "Composable Memory Transactions", Section 2: Background, pg.2"
<end quote>
The point being made by others and myself is that aggregation, or the composition, of atomic operations into services, is not ALWAYS a fixed sum process involving a well-specified finite state machine.
When this composition is not open to human choices, the composite human-computing system has a specific type of inflexibility.