SUO: Re: Lifecycle Integration Schema
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
LIS. Discussion Note 41
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
JA = Jon Awbrey
MW = Matthew West
JA: I have voted NO the first question and YES on the second question,
but I am still open to having my mind changed on the first issue.
My concerns with the first proposal are these:
MW: And there was me thinking that it was the first motion
that was uncontroversial!
JA: The spirit of the motion is all mom and apple pie, as we
say over here, but between the spirit and the letter many
a head has been seen to roll. My fear and trembling, etc.
is over seeing yet another robot's rules of order approach
to questions of inquiry and engineering. Orderly method,
where we find it, is a boon to inquiry, but the brand of
language game that lawyers can beat engineers at 99 times
out of 100 is more often than not used to quash any attempt
to address the real problems that exist in plain sight.
MW: This sounds like "We might fail so let's not try".
I would rather say "Don't try unless you are prepared
to fail". Of course it could all go wrong, but I've
worked in contentious environments and have positive
experience of applying processes that encourage good
outcomes where previously there was underachievement,
so I have reason not to be so pessimistic.
JA: Not in the least. That might be a fair comment if we hadn't both been
working here for the last 3+ years, and paying attention to what keeps
happening on a recurring basis. I am saying "We have already fallen --
now how exactly are we going to get up?" Experience tells me what can
happen, and it's not pessimism to draw on that experience in order to
make different mistakes, at least, the next time around.
JA: So I have a real need to ask specific questions about the design
of your procedures, just like I have a real need to ask specific
questions, based on my real experiences, good and bad, with what
actually happens in reality, about the design and the operating
characteristics of the new-fangled furnace that I'll be buying
this week.
JA: Let me try to rephrase the question this way:
1. How do we know when a problem is really solved?
MW: A problem is identified when an issue is formally raised (which might be
after some informal discussion). The issue is formally solved when both
the person raising the issue and the person charged with resolving it
agree that it is solved. They may be wrong of course, but then the
issue will recur later.
Why will it recur later?
JA: I, too, have positive experiences with methods that work.
And that's precisely why I have always tried to approach
our work here in the spirit of the literature-reading and
problem-solving seminars in which I spent the greater part
of a decade as a multiphase alternating cycle grad student
in comp sci, math, psych, and stats. I do not recommend it
as a way to pick up a degree in a hurry, but it was the best
way there was to learn the subjects in question. Most of my
math professors and some of my psych professors understood the
the fact that their D.Phil.s and Ph.D.s gave them but the first
elements of a small fraction of their subjects, and it was usual
to spend the rest of their lives trying to teach each other a bit
more of the subject, and to work together to try and tackle a few
more of the vast array of problems that remain yet to be conquered.
JA: Given that experience as an idea of how things ought to work,
the question for me is whether the design of your procedures
will give this fallen effort that kind of a lift up, or not.
MW: I would say that it supports it, but does not enforce it.
It would give you a place to put issues where they would be
visible, at least until resolved. I would consider that the
number and seriousness of issues should weigh on those who are
voting to advance a document in the process, and certainly what
is acceptable as a standard.
JA: I am suspicious of bureaucratic procedures that achieve
an illusion of consensus about a given issue or a pretence
of credibility for a given document, when no such consensus
or credibility really exists. We have had our fill of such
procedures already, and so I am naturally somewhat wary of
seeing anymore of this kind of sleight-of-hand.
MW: It is a common misapprehension of standards that they "enforce"
some thing or other. In fact their sole purpose is to prevent
you from having to reinvent something mundane everytime you
want to use it. In this case the mundane thing is how to
raise the quality of something by noting defects and
systematically trying to remove them, against some
stated quality criteria.
JA: Again, I stand up and salute this glorious flag.
But that is not the problem that I'm worried about.
The original furnace in our 20 year old house having
seen the last days of its lifecycle of faithful service,
I'd like to think that all of the various certifications
brochured before our eyes by the heating and cooling guys,
from ACRIS to AFUE to NATE to SEER to UL, and many whose
logos these old eyes are far too farseeing to unravel at
close range, actually mean something.
JA: What I hope for in the semantics of these logos is that some body
of technically competent engineers and humanely-minded ergo-nomics
professionals have all done their homework, both when they went to
school in schools of engineering and ergo-nomics, and when it came
to deciding whether to stamp their logo on these 8.5 x 11 glossies.
And beyond all that, I hope that there was some overarching spirit
of public and human interest in the performance of this competence.
JA: The devil or the god, as they say,
depending on how you translate it,
is in the details.
MW: The process would tend to support those
that stubbornly point out reality.
Yet another cliche involving wishful thinking.
I wish I was so good at this wishful thinking.
The question is: How will the programme design
support those who stubbornly point out realities,
since we all know that this doesn't always happen
the way that we wish and hope and promise it will?
JA: That is the long run hope, of course, and I share it.
But just saying that a given design of procedure will
do what you say is not enough. I have seen a kind of
short-run process that does just the opposite of what
you say, ignoring what are obvious realities to just
about everybody who is not already hypnotized by the
procedural ritual. But it's become some kind of sin
here to mention the fact that there is knowledge out
there that many communities of knowledge consider to
be quite "standard" already, that if our start-work
documents do not contain it, and if our pet logical
formalisms have trouble dealing with it, then they
will remain non-starters for the rest of the world,
whatever local "status" we might dub them with.
MW: Well the process has a political dimension to it.
So people will get what they deserve in the end.
So the Challenger and Columbia crews
got what they deserved in the end?
JA: So it comes down to the next question:
JA: 2. How can you tell when people have done their homework?
MW: Well personally it is because people can explain what they
are doing clearly and concisely, and do not rely on taking
arbitrary decisions (decisions are the enemy of discovery)
to justify their position.
Easy for you to say -- but at least you are engaged in explaining.
JA: Here we are talking about (1) procedures for solving
the actual problems that reality throws at us, and
(2) procedures for solving the associated problems
that arise in discussions among communities of
people who have radically diverse viewpoints
with respect to the nature and implications
for action of the initial problems, up to
and including whether problems of these
descriptions really exist.
JA: From my perspective, there is already a method for converging on
a state of congruence with both our society and our reality, and
that is known as the method of scientific inquiry, as exemplified
by all of the ways that problems are actually solved and phenomena
are really explained in the research endeavors of all our various
communities of inquiry. So I am reluctant to settle for anything
that is proposed to serve as a second-rate substitute for that.
MW: I have no objection to scientific enquiry, and I think
the Quality Process that is at the heart of what I am
proposing is quite compatible with it. It just provides
a little structure which amongst other things would mean
that you knew where you were.
JA: We all have good intenstions at heart. We set up procedures that are
hopefully designed to actualize our intentions and maintain all sorts
of set-points within the domains of our essential operating variables.
I worry about the thermostat. Has it been designed, hopefully or not
so hopefully, to inform us when this control system is out of control?
By what measures should we recognize that the idiot lights are rather
more idiotic than is nominally acceptable? These are real questions
of design-to-task and design-for-testing.
MW: It is out of control, or not controlling at least,
when there are large numbers of issues that are
rejected.
JA: Yes, that's one clue.
JA: Furthermore, I do not how any such work programme can be rendered
enforceable among people who do not already see the need of it,
except by means of coercion of various sorts, the main problem
here being that, once the power of coercion toward assent is
instituted, it almost always ends up being abused in ways
that corrupt the good ends intended.
MW: It can't of course be enforced. The best we get is what we deserve.
If people work and operate this process diligently I have seen it
produce great results, when people try to cheat the process for
the illusion of progress, I have seen poor results. It is not
a silver bullet.
JA: I think that we deserve better than we are currently getting.
How do we address the gap between the suopper and the desert?
MW: The best safeguard it contains is the occassional formal votes
amongst peers. If people take their obligations for the votes
seriously, then progress should only be allowed when supported
by evidence.
JA: The safety of this safeguard, and whether what should be will be,
depends on how well informed are those who vote on the questions.
MW: It would be helpful if at least people
understood what they did not know.
JA: In my experience, it takes work to understand the things
that I wasn't born knowing about, and it takes work to
understand the things that I was born knowing about.
JA: How would your work programme facilitate that?
MW: No magic here. I know no cure for human frailty.
JA: I did not say "magic". I am a little too old for that.
I did not even say "guarantee". Too old for that, too.
When we finally blazon the IEEE logo on SUO, somewhere in
the 24th Century, they'll probably want a limited warranty.
So I have to stress: How will your design "facilitate" that.
MW: It is the visibility of the issues that is the critical thing,
and that the raiser has to agree that it is resolved.
JA: Look, Matthew, these are not academic or rhetorical questions.
Here in the US we are agonizingly working our way through the
aftermath of the Columbia disaster, the aftermath of which is
astoundingly like the aftermath of the Challenger disaster.
And all the after-the-math commission reports point to
a culture of institutionalized ignoramming of what all
the techies were telling all the de-tachees all along.
The question is, when are the organizational systems
that "enculturate" people this way going to wake up
and do the pre-math, so we don't have to keep doing
the aftermath? That is the question. Call it #3.
But it really should have been #1.
MW: It all comes down to visibility.
That is the secret weapon of
this process.
How will your programme make visible
a culture of winking at real data,
a culture of deliberate agnosia?
Jon Awbrey
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o