----- Original Message -----
From: "Tom Rauch" <w8ji@contesting.com>
To: <jimjarvis@ieee.org>; <towertalk@contesting.com>
Sent: Saturday, March 19, 2005 4:09 AM
Subject: Re: [TowerTalk] Porcupines on commercial towers and stuff
> > I observed that I'd counted 138 antennas on top, when I
> > was last there. That's without porcupines. They were
> > un-moved.
> >
> > My assumption would be that the porcupines pre-date the
> > knowledge that they do no good.
>
> The fact government or private installations buy things and
> install them doesn't mean they work as claimed. Governments
> are also slow to learn anything.
I concur with Tom,here.
It's not necessarily that the govt is slow to learn. Here are two examples
where there's a perfectly good reason to buy something that turns out
useless.
1) The govt may buy the item for the purposes of testing or evaluating the
claims. This is certainly true in the case of the lightning static
dissipators. They bought them, installed them, did the tests, decided they
didn't work, wrote the report and that was the end of it. They may or may
not spend the money to uninstall them (if they do no harm, why bother). Of
course the mfr of the device can now legitimately claim (as they do) "as
purchased by NASA and the FAA" or even "tested by NASA", while leaving out
the important extra words "...and found worthless". If you read the
literature of the companies selling these things, they are VERY careful to
not make claims that they have been tested to work. They describe a theory
of operation, and cite lots of customer names, and even some anecdotal
stories of apparent success.
2) The govt spends a fair amount of money on developing things that don't
turn out in the end; called R&D. Perhaps someone has a bright idea, and has
a promising almost-working prototype, but it's not clear whether it is
feasible in a production environment. The govt may order up a bunch of
these widgets, just because if it does work, it's going to be useful, and
the overall cost is low, and there's no other way to find out, other than by
making them and using them. DARPA describes this as "high risk, high
reward". The SBIR program is similar..start with a 10 page concept, .
invest 50-100K in some research and a paper report to take the idea a ways
and figure out whether it's likely to work and how to test it, then invest
500K-1M in a Phase II to do some tests, then hopefully the small company
goes into production in Phase III.
There's a lot of things being done these days in the defense department and
in homeland security that fit in the "it might work, and if it does, there's
high payoff, so let's throw a few million at it". The problem of improvised
explosive devices (IEDs) is prompting a lot of this... it's a very hard
problem, it is very hard on morale, in the context of 100 billion dollars
already sunk, investing a few tens of million in a bunch of really
speculative ideas, one of which might work, might not be a bad strategy.
(although some of the techniques I've seen proposed are pretty darn
unrealistic, but hard to "prove" can't work, short of trying and failing..
this is particularly true for the "demonstrated in the lab at 1 meter, we're
going to ruggedize it and make it work in the field at 100 meters" kind of
proposal.)
And, of course, there's the plain old sucker problem. People buying things
for government and business aren't always experts in what they're buying,
and it might cost more or take unavailable resources to find out. There
isn't infinite time and labor force available when planning a project. In
the case of static dissipators, it might take a while, since there's no easy
way to tell they don't work. If the light poles were properly installed,
then the lightning strike won't do any harm, even if it does hit.
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|