[Esip-preserve] A Suggestion for Rama and John Moses in Their Presentation

[Bruce Barkstrom]

I think it would help the group's discussion to include material
on when in a project's life particular items appear.  For now,
it would probably suffice to say whether it appears
- at or before Conceptual Design Review
- between Conceptual Design Review and Preliminary Design Review
- between Preliminary Design Review and Critical Design Review
- between Critical Design Review and Shipping Readiness Review
- post-launch
For example, calibration data from pre-launch testing usually
appears between Critical Design Review and Shipping Readiness Review,
since you can't really calibrate an instrument before it's assembled
and been through initial testing to make sure it works.  Blueprints
will usually have been produced shortly after Critical Design Review.
If we can have this kind of grounding, I think it would make the
discussion more meaningful.

Second, it would help to know if the items in your lists are
one-off kinds of things or have some structure.  Again, for example,
blueprints for instruments will be done in a hierarchy whose
inventory control usually uses an instrument-developer schema.
In terms of understanding how the blueprints are organized, it
helps to know that an instrument is built from assemblies that
are constructed from sub-assemblies, and so on.  Ditto for
electrical circuit diagrams.  There are plans and procedures
that are linked together by subject (meaning, for example,
that there are calibration plans and calibration procedures,
as well as test plans and test procedures).  Merely having
a list of items is less helpful than having a structured
set of objects.

Personally, I think the structuring of the information is very
important - and not just a "detail" to be brushed over on
grounds that we need an overview that provides soundbite
understanding.  To put it a bit differently, searching through
an unorganized list takes a time proportional to the number
of items in the list; using a hierarchical organization should
go as something like log(N), where N is that number.  The
difference in time that people have to spend in searching
with these two methods is likely to be noticeable.

Bruce B.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.esipfed.org/pipermail/esip-preserve/attachments/20101210/15fb04a1/attachment.html>