Re: Tracking Mechanism

[Carmelo Montanez <carmelo@nist.gov>]

> 
> 1. Should the dates not be in YYYY-MM-DD format for ease of sorting/
> searching?

            Easy enough!!!!!
> 
> 2. <test-reviewer> in the case of an "accepted" status test - should we
> have *both* reviewers names here (for audit trails etc)? Similarly for
> <review-date>. Perhaps a structure like:
> 
> <test-reviewer review-date="2001-09-01">Carmelo
> Montanez</test-reviewer>
> <test-reviewer review-date="2001-09-07">Ben Robb</test-reviewer>
> 
      Not a problem either!!

> 3. Should the status attribute be a lookup? Something like:
> 
> <revisions>
> <statusflags>
> <statusflag id="s1" displayname="accepted" />
> <statusflag id="s2" displayname="rejected" />
> ...
> </statusflags> 
> 
> <test status="s1" id="Numb001">
> <explanation>Missing "name" attribute</explanation>
> <test-reviewer review-date="2001-09-01">Carmelo
> Montanez</test-reviewer>
> <test-reviewer review-date="2001-09-07">Ben
> Robb</test-reviewer>
> <test-date>June 1, 2001</test-date>
> <test-submitter>David Martson</test-submitter>
> <comments>Problem was reported back to
> organization</comments>
> </test>
> ...
> </revisions>
> 
> The advantage of this is that in any admin tools you are able to present
> the user with dropdown lists very easily, and misspelling/ case issues
> are reduced. It also means we have a central repository of Status flags,
> which makes the process of updates to the words easier. I suppose it
> comes down to design style whether you want this level of control.
> 

        Will atempt to implment today and will post by next week.

> Of course, this then leaves us with a problem if one reviewer accepts
> the test, and another rejects it.
> 

        I think, there is not much we can do about that!!

> Perhaps putting a status flag on the reviewer?
> 
> <test-reviewer review-date="2001-09-01" status="s1">Carmelo
> Montanez</test-reviewer>
> <test-reviewer review-date="2001-09-07" status="s2">Ben
> Robb</test-reviewer>
> 
> You could then build your business rules to test to make sure that
> *both* reviewers have the same status flag...

        mmmm yeah, but still there will be cases that they will have
        different status.
> 
> I'm going round in circles - I'll have a think about it tonight.

    quite the contrary, I think you had great ideas.

Greetings,
Carmelo