Notes from Spec Work

[Rex Brooks <rex.brooks@ncoic.org>]

  Hi Everyone,

I just uploaded the first full Working Draft (WD01) version. I decided 
to wrap it up before I send this message so I can include it all in one 
message, of which I suggest you save a copy for reference during the 
face to face meetings. The WD01 does not yet have a complete Data 
Dictionary. However, I left placeholders for the remaining 129 (+ or -). 
I only got 21 entries done for that, but Tim has the original of that 
from which I am working, and I will keep a copy of that with me, which I 
can update as we go along, and make sure that the correct entries are 
used in the document of record. I tried to get this to you before 11:00 
a.m. Eastern Time, but I'll just have to be satisfied with getting it 
out to you before noon Eastern Time.

Needless to say there is much that needs to be written yet, but I've 
pulled together enough the specification document so that we can work 
with it as a whole even if it isn't complete yet. I thought that was 
important for setting the stage for us to keep our eyes on the objective 
of getting the first review draft done as soon after the f2f as is 
feasible with doing the best we can.

As usual, I would suggest we take the position of trying not to let the 
perfect be the enemy of the good enough. This is a complex little 
bundle. I have already noted that we need to include "Requests" for each 
situation report message type, and you will see that I have allowed for 
that. We can still do it whichever way we choose, but at least it is not
missing" here at the start.

I wanted to call your attention to a few items that we will only get 
into IF we go through the Notes I have collected while working on 
aligning the EA model with the Data Dictionary and with preparing an 
initial WorkingDraft (WD01) of the Spec for our Face-to-Face meeting. (I 
will upload WD01 a little later today) Since going through that document 
is not a given, and not something I think we need to do in depth 
immediately, I don't want these three issues to get lost in the mountain 
of stuff we have to go through.

First, I suggest we add four new elements to give each of the predefined 
situation report messages their own root element. We already have 
FieldObervation so we only need SituationInformation, 
ResponseResourcesSummary, ManagementReportingSummary and 
CasualtyAndIllnessSummary. This also makes the diagrams for these 
situation report type messages work better because otherwise there is no 
element through which the message is "realized." Regardless of the UML 
term for it, since each message needs a root element, we have to do 
something to make that work. I have noted this in my "Notes" for the 
redundantly minded.

The second item is how we refer to the Situation Report Root element 
throughout the Data Dictionary. Currently we have 'SitRep "Root" element 
/ container' and I think we should just use the actual name of the root 
element of the whole specification, which I suggest should be 
"EDXLSituationReportRoot" or  "EDXLSituationReportReferenceType" to use 
the same naming as that used in EDLX-RM either of which is both shorter 
and more specific and leaves less room for anyone to get confused about 
what we are referring to.

The third item is closely related to the second one above. I suggest 
that whenever we refer to an instance of a message from the spec we use 
a completely generic uncapitalized series of three terms: situation 
report message. This is also helpful whenever we want to refer to a 
specific kind of situation report message, for instance: "... we use 
textual descriptions for the situation report message type titled 
FieldObservation."

I would like to avoid using "/" between any two terms if we can avoid it 
because it immediately invites the reader to ask "what do you really 
mean by that?"

Regardless, while I would prefer these items to handled as I've noted, 
I'm not about to be roadkill for it. We've got waaaay too much to do to 
get wrapped around any axles.

Cheers,
Rex