[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [soa-rm-ra] collected concept definitions per action
|
Frank, We create definitions that are specializations of accepted
meaning so as to have a consistent connection between concepts important to our
domain. However, we are not free to rewrite the dictionary because common
usage hasn’t caught up with our perceived needs. If we go against
our reader’s expectations of language, we only create confusion. Ken --------------------------------------------------------------------------- Dr. Kenneth Laskey MITRE Corporation, M/S
H305
phone: 703-983-7934 7515 Colshire
Drive
fax: 703-983-1379 McLean VA 22102-7508 From: Francis McCabe [mailto:fmccabe@gmail.com] Ken: The origin of the concept of
intent within the RAF is different to this. We are using it to capture an
actor's internal processes in performing actions. Nothing to do with declaring
to the world. I think that there are v. few
dictionaries that do a reasonable job on the definition of action. On Sep 28, 2010, at 1:18 PM, Ken
Laskey wrote:
Frank, Most of what I’ve put
together is just a cut and paste of what is in the document. I did revise
the definitions of intent and commitment because your changes in response to my
comments missed the mark; more on that below. If some of the model labels
I used don’t catch the proper essence, we can change the labels or
possibly that indicates some needed changes in the words. Now to Intent and Commitment. Intent is not internal
planning and orienting. The online definition is “Something
that is intended; an aim or purpose.” Synonyms are intention, intent, purpose,
goal, end, aim, object, objective. There is no planning or
orienting. This was an attempt to capture what
are internal processes. Action is the application of intent. The application of
what? Intent is important because it has a private/public aspect, much like
trust. In order to lift a pencil, my arm
has to move. Before the arm moves, my brain 'decides' to lift the pencil and
starts to plan the muscle movements. Before deciding whether to lift the
pencil, my brain has to be 'oriented' so that lifting a pencil is a reasonable
thing for it to decide on. In computer architecture terms,
orienting is analogous to ensuring that certain variables (including the
program counter) have reasonable values prior to actually performing
actions (invoking functions).
The online definition of
commitment is “A pledge to do.” The participant
doesn’t ensure the commitment is satisfied but they state what they
intend. Any dictionary definition that
limits commitment to actions is deficient. I am committed to the well being of
my family; this is can can be stated without elaborating on the actions needed
to ensure the desired state.
Additionally, you note
“rights are part of a social structure. Permissions are part of
policies.” Yes, we may currently discuss rights as connected
to words about social structure and we may discuss permissions as connected to
words about policy, but either the definitions should make clear how the two
are different or we should use the same term/concept in both places. This
applies equally to the other definitions I suggest are redundant. I agree wrt rights and
permissions. This is slightly tricky as there does seem to be a distinction,
and there are existing uses of the terms, but the distinctions are not always
obvious. I prefer connecting rights, responsibilities etc to social structures
because that seems to give a clear distinction and also seems to honor the
intuitions.
Again, my *intent* was
a reasonably complete summary in one eyeful of what is already there. Is this a publicly declared intent,
or one that you were pursuing without telling the whole story on :)
Ken --------------------------------------------------------------------------- Dr. Kenneth Laskey MITRE Corporation, M/S
H305
phone: 703-983-7934 7515 Colshire
Drive
fax: 703-983-1379 McLean VA 22102-7508 From: Francis McCabe
[mailto:fmccabe@gmail.com] Some immediate thoughts: 1. In general, we are only
interested in measurable things. So, there is no sense in talking about
unmeasurable state. We had a lot of discussions about
state in the RM days. We came to the conclusion that state is characterized by
facts but is probably not the same as facts. 2. We have generally avoided
cardinality unless needed. Cardinality is not needed for private state. 3. The real world effect should
refer to the world, not to what is known about the world. 4. Intent DOES NOT need to be
stated. Generally it is NOT stated. Indeed, most people rely on evaluations of
intent (much like trust) as opposed to stated intent. We should keep closer to
the original definition: Intent is the internal planning and orienting
of an actor to
achieve an objective 5. Commitment is not directly
connected to intent. e.g., I promise that it will be sunny tomorrow. Should not
be unnecessarily connected to shared anything. Original definition: A commitment is a social
fact about the future that a participant is responsible for ensuring is
satisfied. 6. The concept of permission is a
standard part of the policy ontology. Not inherently connected to social
structure. A right is a permission that is connected to a social structure. 7. Achievement should not be part
of obligation - it implies that the actor is not currently in the desired
state. Similarly for permission. 8. I am not against prohibition,
except that it is a form of negative permission/obligation. 9. goals are internal, purposes are
external. I do not have a purpose but I do have goals. 10. rights are part of a social
structure. Permissions are part of policies. 11. Ditto for obligation and
responsibility On Sep 28, 2010, at 10:48 AM, Ken
Laskey wrote:
<section 3 collected
definitions.docx> |
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]