RE: Agile specifications (was: Managing Extreme Programming)

George,

> -----Original Message-----
> From: Dinwiddie, George 
[>] snip
> 
> "Standard off the shelf" for spacecraft is a bit less highly evolved
than
> it
> is for many earthly endeavors.  Because of the long lead time to
certify
> any
> new components for flight, evolotion of those components in inhibited.
> 
> A friend who still works at Goddard tells me about a British
spacecraft
> builder who's a lot more agile in their hardware development.  Their
> hardware interfaces are apparently more evolved than Ball's and allow
them
> to mix-and-match or create new modules in a much shorter timeframe and
for
> a
> lot less money.

[>]Ball's approach is to reuse the sensors and the interfaces that drive
them. The "profiles" of the sensors are tailored to the mission, but the
form factors and other items (cooling, calibration, and diagnostics) can
be reused as well. The idea is to drive toward common platforms.

My connection is on the DoD side so the civilian or commercial side is
likely different. Also in the spacecraft business there is no such thing
as "give me one like those guys had," they're always customizing them
because of sensors and comm. gear.

> > What's defined at time of the order is the detailed behavior
> > of the instruments and the onboard processing as well as the
> > data link processing onboard and on the ground.
> 
> Yes, and the design of the instruments could be better if they
depended on
> what the Principal Investigator knew 6 months prior to launch rather
than
> 2
> to 5 years prior to launch, but there's a long process to get a
project
> approved, get funding, build the hardware and software....  Spacecraft
> hardware and onboard software tends to be *very* primitive.

[>] You're using words of scientist; my connection there does DoD so the
specs are defined ahead of time. I do have neighbors that work on the
sensor side for earth sciences and the 5 years ahead is not the normal
lead time at least for crop and water sensors. A firm here in Longmont
had their bird built by LM in under 18 months and were in business
shortly after that.

> > What organizational hurdles have you come across that slow
> > down the process that could be improved? How could "agile"
> > make these improvement from your experience?
> 
> It's been several years since I've had direct contact with any NASA
> projects, and I'm not an expert at Organizational Patterns.  I don't
think
> I'm the person to advise on how it can be improved.  You don't have to
be
> an
> expert, however, to see that there's room for improvement.  Or that
many
> formalities are done for show, without doing the underlying work
that's
> intended.  I've seen "design" documents that had been reviewed and
"signed
> off" but didn't answer many important questions.  They dodged the
unknown
> issues with language.  It's ludicrous to say that these particular
> documents
> "specified" the design.

[>] there's always room I know, but like most of us sitting on the side
lines, it seems it is always harder than first thought.

> I'm reminded of a tale I heard about a long-ago medical exam.  The
proper
> answer to "Why does morphine cause a person to be sleepy?" was
"Because it
> has a dormative principle."  Of course, this is a tautology and
explains
> nothing.  It just says that morphine makes a person sleepy because
it's
> been
> found that morphine makes people sleepy.
> 
>  - George
> 
[>]Glen B. Alleman


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Flexible Keyboard is the ideal accessory for PDA users that are on the move.
http://us.click.yahoo.com/dCBVZC/WnCFAA/xGHJAA/nhFolB/TM
---------------------------------------------------------------------~->

To unsubscribe from this group, send an email to:
[email protected]

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/