[extropy-chat] Peak Oil - Meta

[Ian Goddard]

--- Hal Finney <hal at finney.org> wrote:

> > [*] http://users.erols.com/igoddard/wacoflir.htm 
>
> Well, you deserve all the credit for that.  The 
> truth is that my theory was wrong.  You came up 
> with the idea about material on the ground
> fluttering in the wind to make things flash, 
> which explained all the data.


 That idea addressed the major fact the official
report overlooked: some flashes appear at given spots
in rapid succession naively resembling machinegun
fire, a central axiom of the gunfire theory. It was on
a friend's large 3-foot TV screen that I saw a
flapping piece of something (also be seen in photos)
attached to a fallen wall panel just where the most
noted successive flashes appear. The flag also
constantly flapped that day.

 That gets to my point regarding the way to find the
truth of fringe claims. That's it and it's as good for
*any* claim. If a claim says, "There exists some x
such that x has the property A," then go out and find
an x as best as you can and see if it's an A. But that
may not always be practical for everyone, so your
proposed project is probably worthwhile.



> Today I wouldn't even try to get into it.  This 
> kind of fringe stuff is too far off the radar to 
> even be credible enough to pay attention to.

 But IR professionals are split on the gunshot theory
(see my list at link above), which makes it a good
example here. The government stuck with those who said
they were reflections. So there's no way the average
observer could really know whose right without looking
at the source data themselves, which I provide at my
site and which makes this case an ideal example of
needing to examine source data. 

 A CATO report presents what may be the consensus on
the Waco FLIR, which says more research is necessary:
http://www.cato.org/pubs/pas/pa395.pdf . For me, being
able to line up the flashes with inanimate objects,
and only that, makes the case closed.


> I'm sure that was already the case for most people 
> even back then. And after all, they were right, at 
> a considerably lower cost than I went through, and 
> enormously lower than what you had to go through to 
> get to essentially the same conclusion.

 Right, but most who believe the experts who say the
flashes are not gunshots are doing the same thing as
those believing the experts who say gunshots... which
is to base their belief on that of others. Second-hand
belief. What I'm saying is that it's better to base
belief on a first-hand review of source data. But
obviously one cannot afford to do that on all matters,
so your idea has a necessary niche.

 But still, when the govt hired Vector Data to examine
the FLIR, a report avoiding the enormous expense of a
flash-by-flash analysis would not be acceptable. If
they'd just taken a vote among FLIR experts (which
might be a draw), that would not be acceptable.
Ultimately someone has to actually check the source
data, and that's not necessary cheap. 

 But to your point, perhaps some way of automated
interrogation of http://scholar.google.com might prove
useful in your project. That's basically my intuitive
response to the question, "What's most likely worth
believing regarding topic x." ~Ian

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com