On reflection there are clearly some holes in my understanding of the CONOPS.
But it also seems to me time to think about what we have done so far, and
what we still need to do. So this is an attempt to apply some of the
principles we have been building for supporting Cdrs to the problem in
hand.
So these are associations, observations, what have you:-
Original Objective. To support the J-8 mission in management of JM&S
Necessary action:- To provide J-8 decision makers with information well
matched to their decision trees on JM&S.
Necesaary action:- To collect and label said information so as to optimise
determination of truth values deciding between binary alternatives
in said decision trees.
Observation:- These requirements propagate down the command hierarchy
with J-8 at top and concerned with JM&S.
Necessary Action:- We must understand the decision trees of all
personnel in said command hierarchy well enough to optimise
information provided to J-8, in sense aforesaid.
Achievements so far:-
Tom has found a good labelling scheme, and explained it to
folks at J-8. How do we know it's good? Inertial navigation
by seat of pants.
Next Necessary Action:-
The fact that Tom could do inertial navigation tells us he
knows a lot about the needed decision trees. Next action -
To articulate that knowledge, and put it down on paper. NOT
EASY.
Resulting Possible Further Action:-
Having got that knowledge, we have a handle on who needs to
know what and why. This is the functional requirement for
access to parts of the data dump.
Conclusion:-
Even if we distribute whole data dump on CDROMs we need guidebooks
for people traversing it. These can be written only when we know
what they have to decide in order to feed optimal information
up the chain. Conversely, the guidebooks are what have always
been called guidance by Cdr. That is to say, we should be
acting as staff to J-8 in deciding what guidance they need to
send down the tree so as to optimise their decisions.
Conclusion:- The guidebooks determine what we MUST distribute for the
system to function, i.e. WHAT PEOPLE NEED TO KNOW. Since the
data dump is UNCLASS we can also distribute WHAT THEY MIGHT
NEED TO KNOW IN UNFORESEEN CIRCUMSTANCES, perhaps the whole
data dump, unless it is misused in support of inter-service/
intra-service quarrels. Note that some uses in this arena
are not only justified but needed. A needs to be able to say
"I wonder if B needs this special item, or whether we can
adapt something I use to his needs. I can clearly tolerate
these kinds of changes without impairing my operational
effectiveness, but those give me grief, what does B think?"
What must NOT happen, and I see it too often in many places
is A never having even walked ten paces in B's shoes, saying
"Oh hell, B can't possibly need what he says - the ATO is
a perfectly good view of the battlefield." Cf. Peter's
visit to Benning, and the question of common night vision
equipment.
Note to readers. I'm circulating this to some people outside because
it is relevant both to my work with DFS, and to technology transfer
into the non govt. sector.
John Gabriel 12/10/92 10:20 hrs CST.