OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

emergency-msg message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: EDXL-AT & Situational Awareness

Hi Tim, fellow SCers,

I am not assuming this is true, but recent emails make it appear 
likely that we are going to get a formal submission for an EDXL-AT, 
though I hope it is through the Infrastructure SC. This is, if it 
happens, I assume, separate from Situational Awareness (it certainly 
SHOULD be since Assets are only one type of Situational Factor 
Category). Three things must be taken into account as this process 
moves forward:

1. Absolutely clear and unambiguous definitions of ASSET and 
RESOURCE, using the work already done on EDXL-RM "MUST" be written. 
This is simply not optional, and when people start getting hinky 
about it, just let them know that they will have nice, messy free 
text DESCRIPTION fields to make all the ambiguous shaghetti they feel 
they need. They can be counter-productive on their own dime in their 
own time.

No amount of assurance that they "know" best is going to prevent some 
of us from finally digging in our heels and refusing to accept 
candidate specifications no matter how long or conscientiously any 
group or set of groups has diligently and expertly worked on them.

2. We are not accepting Candidate Specifications. Regardless, whether 
it is AT or Situational Awareness, or both we need to make it 
PAINFULLY clear to our SMEs that we are NOT looking for candidate 
specifications anymore. Please make sure they have our Requirements 
Specification Template. It remains in the RC Document Repository, I 
know because I sent it to the International Health Continuum TC this 
week for the same purposes reiterated below.

We need them to build their Use-Cases as was done for EDXL-RM, and 
then build a rigorous set of requirements from those Use-Cases 
(diagrammed in UML if possible) with CONFORMANCE language built into 
it, e.g. "MUST," "MAY" and "SHOULD."

EDXL-HAVE, in my opinion, proves beyond a shadow of a doubt that 
accepting candidate specifications by groups of SMEs is almost always 
a mistake. SMEs rarely fully understand how broad, yet rigorous, the 
specification-writing process needs to be.

We have now had to PAINFULLY deal with the consequences of being 
persuaded that the SMEs know their area best and that means they are 
the best group to write a specification for that area, instead of 
carefully working through a rigorous requirements-writing process.

3. We are working toward developing a unifying EDXL Reference 
Information Model. Unfortunately, we don't have an EDXL-RIM yet, and 
that is the only thing that would make building and submitting a 
candidate specification marginally feasible. When we do develop and 
OASIS approves EDXL-RIM, we can ask them to conform to it, if they 
think they absolutely must waste their time and ours by writing 
another candidate specification.

If at all possible, even if it means spending a lot of time in a 
crash project, we should build EDXL-RIM before or concurrently with 
any new specs. We need to get that EDXL-RIM sucker done yesterday!

Lastly to repeat and rebeat this dead horse, regardless of how well 
SMEs write a candidate specification and how well their collective 
experience in the area informs them, anything they write is not going 
to be organized in the way we have developed, nor is it likely to 
conform to the specialized techniques and processes we have developed.

I really don't want to go through this PAIN or these discussions and 
debates again, We have learned this lesson three times now. Let's not 
continue building the case for the necessity of Formal Requirements 
from Use-Cases.

Trying to take a Stitch in Time,
Rex Brooks
President, CEO
Starbourne Communications Design
GeoAddress: 1361-A Addison
Berkeley, CA 94702
Tel: 510-849-2309

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]