Kim Jones wrote:
> On 06/12/2008, at 6:18 PM, A. Wolf wrote:
>>> I guess what I am on about is a bit closer to the 80s idea of "chaos"
>>> - something that is inherently unpredictable; at least if you adopt
>>> the stance of always launching your prediction from a single  
>>> present -
>>> the one you happen to find yourself in.
>> I think you mean randomness, not chaos.  Chaos theory deals with
>> deterministic systems that vary widely in result based on small  
>> changes in
>> initial starting conditions; these systems are 100% predictable.
> Don't believe in randomness. "Random" means we don't understand what  
> determines it. Like "Junk DNA"; it's only junk up until we work out  
> what it's really for.
> You can predict with 100% accuracy that systems with varying initial  
> conditions will bifurcate and become chaotic once driven beyond a  
> certain point. What you cannot say is what is determining the order in  
> the chaos once it arrives. That's closer to what I mean.
> 2 men start to dig a hole. They are instructed to make it reach a  
> depth of 5 feet. One of them murders the other with his shovel. Nobody  
> predicted that would happen. 
How do you know that?  Maybe it was quite predictable.

> We can 'determine' the reasons for this  
> event only AFTER it happens, even though it was determined by  
> something that we might have noticed prior to the event if only we had  
> been able to. All action can be seen as logically determined in  
> hindsight. 
You must not have heard of quantum mechanics.  And how can you know the 
causes seen in hindsight are correct.  Modeling the past is hard too.  
In what sense do we know why the terrorists flew planes in the WTC?

> Before something happens is where we would like to be more  
> on top of things. Intelligence exists on terrorism but usually this  
> usually fails to determine our actions to prevent terrorist acts,  
> interestingly enough. If you look closely at what I am saying, it is  
> the rather messy human consciousness part of the equation that SPOILS  
> the mathematical modelling in most cases.  

Sure.  Humans and even most animals are extremely complicated.  Even 
things like weather and viscous flow over an aircraft are to complicated 
to model except approximately over a limited range.

> I am saying that a full  
> perceptual scan of the situation often takes us way beyond what the  
> data suggests. Just like the Mumbai massacre - intelligence WAS  
> available that suggested it could happen. Yet the massacre was  
> "allowed" to happen, because nobody could see the looming pattern in  
> the data until after it happened by which time it was bleeding obvious  
> to one and all. It seems that in many situations the sheer volume of  
> information available is as much a part of the problem as the decision- 
> making process. How do we decide? Data alone are incapable of making  
> decisions. You still need a wet, messy human brain with a perceptually  
> skilled mind to do that

And apparently that doesn't work all that well either.
>>> Isn't this kind of like an act of  faith?
>> No.  Faith isn't based on evidence.  When we use math to model  
>> things in
>> reality, we do so empirically.  If a distribution doesn't fit after  
>> testing
>> it, we don't use it to model that set of data, for example.  How we  
>> use math
>> functionally is different from math itself.
> Sure, but correlations occur between sets of data and there is a  
> tendency for them to look like causations. Most scientists and data  
> collectors are trained to say "correlations do not necessarily imply  
> causations" which is great advice up until the correlation turns out  
> to BE a causation

You just seem to on a rant against everyone who thinks, calculates, 
predicts, etc.
>>> If we could perfectly model where things are heading then
>>> please tell me why all the BTSOAPs of the dismal science of the
>>> economics world could not arrange a more stable financial future for
>>> us than the one we are currently moving into?
>> The problem with (for example) economic forecasts is not that the
>> mathematics is flawed; the math is fine.  It's the data collection  
>> that's
>> flawed.
> I would say that it is the premises - the starting assumptions that  
> are probably flawed, not the collection method, unless by collection  
> method we include the starting assumptions which are unavoidably  
> guiding the triage of the data. 
It depends on the problem.  Calculating the flow over an aircraft isn't 
hard because of data collection, it's hard because there are no closed 
form solutions to the Navier-Stokes equations and because the airplane 
has a complex shape. Calculations about what humans will do is hard both 
because knowing what's in their brains is hard and because there are no 
good models

> The biggest flawed assumption of all  
> is the belief that data collection of itself will give rise to ideas  
> and concepts. Data do not do that. We choose our starting points and  
> assumptions in all cases. Often we aren't even aware of these because  
> - well, they are assumptions and nobody really questions assumptions  
> much. 
On what planet?

> Our way of looking at the data is not itself present in any way  
> in the data. I'm saying that if reality conforms to our model of it,  
> then the danger is we are looking at only a part of the situation. The  
> issue in question is the WHOLE situation, not just that part we CHOOSE  
> to look at. How can we ever be ccertain that we are in fact looking at  
> the whole situation. 
Of course we are never looking at *the whole situation* (the 
universe?).  We know we are looking at enough of the situation when our 
models make successful approximate predictions and when we understand 
where the errors come from.


You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To post to this group, send email to [EMAIL PROTECTED]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at

Reply via email to