Understanding through Discussion


Welcome! You are not logged in. [ Login ]
EvC Forum active members: 81 (8950 total)
28 online now:
dwise1, PaulK, Thugpreacha (AdminPhat) (3 members, 25 visitors)
Newest Member: Mikee
Post Volume: Total: 867,381 Year: 22,417/19,786 Month: 980/1,834 Week: 50/430 Day: 5/45 Hour: 0/1


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   When Inconsistent Irrationality is Better
Stile
Member
Posts: 3863
From: Ontario, Canada
Joined: 12-02-2004


Message 1 of 12 (501283)
03-05-2009 2:23 PM


I'm sure we can all agree that when truth and honesty are our priorities, then objective, verifiable evidence is the best path to take when exploring our environment.

...when time permits :)

So, what about if time does not permit? Is it possible for there to be a certain time-constrained scenario where the absolutely verifiable perfection that is objective evidence just "takes too long" in order to make significant progress?

Take, for example, the Travelling Salesman problem.
This is a computational problem where the issue is to solve the shortest round-trip distance between multiple cities.

Using absolutely rational, consistent information, we could use a a brute-force method to check all possible paths. Then simply pick the shortest route. This would guarantee us the shortest path. The problem is that this can take a long time. Even with simply a few hundred cities, the computational power required to check all paths would take years to compute.

Using irrational, inconsistent information, we could use approximate methods with some randomization to "look for" shorter paths and end up cutting our time for a 99% sure solution to a mere fraction of what a brute-force method would demand.

So, this mean to me that there are situations where inconsistency and irrationality can be "better" than consistency and rationality. Generally, if time is a factor and we'ed like to make "good enough" progress and leave the absolute verification for later.

Of course, this won't do for science. But it certainly could do for honest explorers. That is, if we deem the risks to be acceptable, we can save time and attempt to "jump to conclusions" in an irrational, random, inconsistent way. Then, when we land on something that "looks pretty good", we can take the time to add in a more rigorous approach to "make sure" we're not making any mistakes. Such an approach is faster then brute-force, consistent, rational methods. And, as long as the the after-thought rigorous check is insisted upon, it's just as valid in the end.

My point is to say that consistency and rationality is not ALWAYS the best approach. It is not the best approach if time is a factor. If anyone is holding themselves to a standard where they want to be consistent, and rational for EVERYTHING at ALL TIMES, I am submitting that you are... wasting time :)

As long as we can identify when we are being irrational, it can be to our advantage to put it to good use. Or course, it's "safer" to just always be rational and consistent. Then you don't have to worry about if you're being irrational, or if the situation calls for it. However, for anyone who is capable of identifying and controlling such things, you certainly can save time by using irrationality correctly, in the right situations. Just watch for thin ice :)

Not sure where to put this. I'm hoping for "Is it Science?" But I understand if its non-connection to the direct EvC debate is too large and it needs to be put into Coffee House.


Replies to this message:
 Message 3 by shalamabobbi, posted 03-06-2009 5:26 PM Stile has responded

  
Admin
Director
Posts: 12653
From: EvC Forum
Joined: 06-14-2002
Member Rating: 3.0


Message 2 of 12 (501414)
03-06-2009 6:58 AM


Thread moved here from the Proposed New Topics forum.

  
shalamabobbi
Member (Idle past 1194 days)
Posts: 397
Joined: 01-10-2009


Message 3 of 12 (501555)
03-06-2009 5:26 PM
Reply to: Message 1 by Stile
03-05-2009 2:23 PM


please expoud..
Hi Stile,
So, this means to me that there are situations where inconsistency and irrationality can be "better" than consistency and rationality.

Why shoot yourself in the foot with this statement. The fact that there is not enough time to make the calculation means that it is not the shortest 'path' to follow. If there is any continuity and smoothness to the data we can restrict our search to the direction of the steepest gradient and at least find a local maxima. This is the subject of optimization theory.

Maybe you should provide more examples of what you are trying to convey. Something from the real world where you follow a path dictated by something other than a rational approach. In any case I would think even if a subjective approach were followed you'd want to be consistent. I'm not sure I understand your point yet.

Edited by shalamabobbi, : No reason given.


This message is a reply to:
 Message 1 by Stile, posted 03-05-2009 2:23 PM Stile has responded

Replies to this message:
 Message 4 by Stile, posted 03-09-2009 9:54 AM shalamabobbi has responded

  
Stile
Member
Posts: 3863
From: Ontario, Canada
Joined: 12-02-2004


Message 4 of 12 (501998)
03-09-2009 9:54 AM
Reply to: Message 3 by shalamabobbi
03-06-2009 5:26 PM


Re: please expoud..
shalamabobbi writes:

This is the subject of optimization theory.

Yes. That's the whole point. Optimization theory is not strictly rigourous, it involves certain guess work and certain random areas that lead to a within 95% (or whatever) probable best solution.

Guess work and randomness are the arena of irrationality.

The fact that there is not enough time to make the calculation means that it is not the shortest 'path' to follow.

Just because we are justified in using irrationality in order to take the shortest path (if we deem time to be a significant factor) doesn't somehow make that path rational.

That's exacty what I'm saying. Irrationality has it's place. Sometime's it's best to use it. But it's dangerous to start calling a method that is based on irrationality to be "rational". Call it justified, call it "the fastest way", or call it "better" (as I did).

But when you start calling the irrational "rational," then you start down a dangerous path of possibly mixing the two together unkowningly in other areas.

Exploring is a very nice real-world example: Being the first to sail through unchartered waters and rocky shoals to reach a new island.

A rational, rigorous approach would include mapping out the terrain to make sure your ship (and therefore your life) are not lost.

An irrational approach would take a "best guess" at the rocks/currents/water ahead and get to the island. The risk for loss of both ship and life are there. The movement ahead before checking it to see if you can actually make it is irrational. That is, it may strictly be impossible... 100% chance of losing ship and life... you don't know if you don't check.

Irrationality certainly can be better, but it's not good "brain housekeeping" to say it's nothing worth identifying. Well, if "remaining in control" is a priority, anyway.


This message is a reply to:
 Message 3 by shalamabobbi, posted 03-06-2009 5:26 PM shalamabobbi has responded

Replies to this message:
 Message 5 by shalamabobbi, posted 03-09-2009 5:51 PM Stile has responded
 Message 6 by Straggler, posted 03-09-2009 6:38 PM Stile has responded

  
shalamabobbi
Member (Idle past 1194 days)
Posts: 397
Joined: 01-10-2009


Message 5 of 12 (502113)
03-09-2009 5:51 PM
Reply to: Message 4 by Stile
03-09-2009 9:54 AM


Re: please expoud..
That's exacty what I'm saying. Irrationality has it's place. Sometime's it's best to use it. But it's dangerous to start calling a method that is based on irrationality to be "rational". Call it justified, call it "the fastest way", or call it "better" (as I did).

In a discussion such as you are starting, the biggest problem is definitions to avoid talking past each other.
Irrational is not quite what I think you are after. Maybe a new word and definition could be coined by you for where it is you are wishing to take this after establishing your premise.

"not rational" "a-rational" etc

Otherwise you will start a discussion in semantics. The more precise the definition the better you will be communicating your thoughts.

An irrational approach would take a "best guess" at the rocks/currents/water ahead and get to the island. The risk for loss of both ship and life are there. The movement ahead before checking it to see if you can actually make it is irrational.

I would say as a counter to your use of the words rational/irrational that the best guess is a rational approach within the constraints of time allowed to make shore. They would explore the shoal for the best passage within the time constraints allowed.

Maybe what you are saying is that in certain arenas science does not provide a definitive answer. I don't think that is debated by anyone. Chaos theory establishes that fact in certain situations. QM establishes it also for certain kinds of knowledge.

You may have to move toward your conclusion before encountering an issue upon which there may exist some debatable topic

if "remaining in control" is a priority, anyway.

Control or prediction is the goal of scientific knowledge, but in most areas the limitations imposed on that possibility are also understood where they are fundamental in nature.

Irrationality certainly can be better

Again a semantic issue. I think you are saying that a complete solution is not always possible. This does not make the evaluation irrational however.

This message is a reply to:
 Message 4 by Stile, posted 03-09-2009 9:54 AM Stile has responded

Replies to this message:
 Message 7 by Stile, posted 03-10-2009 7:39 AM shalamabobbi has not yet responded

  
Straggler
Member
Posts: 10285
From: London England
Joined: 09-30-2006


Message 6 of 12 (502121)
03-09-2009 6:38 PM
Reply to: Message 4 by Stile
03-09-2009 9:54 AM


Re: please expoud..
I think your (mis)use of the terms "rational" and "irrational" is going to take you down all sorts of misleading paths here..

For example scientists use Newtons equations of motion and gravity in all sorts of circumstances. They know that the calculations would be marginally more accurate but much much much more complicated if undertaken using the equations of GR and SR.

Wouldn't it be irrational for them to do this if Newtons calculations provide perfectly adequate and practically workable answers?


This message is a reply to:
 Message 4 by Stile, posted 03-09-2009 9:54 AM Stile has responded

Replies to this message:
 Message 8 by Stile, posted 03-10-2009 7:42 AM Straggler has not yet responded

  
Stile
Member
Posts: 3863
From: Ontario, Canada
Joined: 12-02-2004


Message 7 of 12 (502169)
03-10-2009 7:39 AM
Reply to: Message 5 by shalamabobbi
03-09-2009 5:51 PM


Not a fascinating new idea
I would say as a counter to your use of the words rational/irrational that the best guess is a rational approach within the constraints of time allowed to make shore. They would explore the shoal for the best passage within the time constraints allowed.

My main point was to use the words "rigorous, objective and verifiable" and to say that they are not always the best (fastest) route. That is, when time is a factor. The title's use of irrationality and inconsistency is just an eye-catcher.

Do you have a problem with me saying that the sailor's exploration of the shoal for the best passage "within the time constraints allowed" is not "rigorous, objective, or verifiable?"

Because it isn't. It's just the sailor's best guess.

The point is to say that alway being objective and verifiable is time consuming. And, if time is an issue, then it can be better to abandon this strictly formal methodology.

"Can" is an important word here, because the consequences may also be very negative (in the case where the shoals may actually be 100% impassable, say).

Maybe what you are saying is that in certain arenas science does not provide a definitive answer. I don't think that is debated by anyone.

My post here isn't meant to be some mind-blowing, new knowledge. It's just a post on an internet forum. It's not exactly new stuff. It's just a place to put thoughts. What may be common knowledge to you may not be so obvious to someone else, though. I don't believe there is anything at all that is "not debated by anyone." It certainly may not be debated by you, but you also cannot speak for everyone.

I didn't intend this to be some profound new idea, simply a statement of the obvious, really.

Again a semantic issue. I think you are saying that a complete solution is not always possible. This does not make the evaluation irrational however.

No, it does not force it to be irrational. However, one can choose to make it so. And in so doing, it's quite possible for the situation to come out "better" (reach a solution faster). Which is what I'm saying.

I'm not saying that a complete solution is not always possible.
I'm saying that a non-objective, non-rigorous, non-verifiable approach has the potential to reach an acceptable solution much, much faster than a strictly formal objective, rigorous, verifiable method.

It's not really that big of a deal.
But, on this board, it sometimes comes off that "the rigorous scientific model" is held up on some sort of golden, un-surpassable pedestal. The point of this thread is to show that this is not true. It is only placed on such a dais when truth and knowledge are the extreme priorities. Sometimes they are not priorities... sometimes exploration at any cost is a priority... sometimes other things.


This message is a reply to:
 Message 5 by shalamabobbi, posted 03-09-2009 5:51 PM shalamabobbi has not yet responded

  
Stile
Member
Posts: 3863
From: Ontario, Canada
Joined: 12-02-2004


Message 8 of 12 (502170)
03-10-2009 7:42 AM
Reply to: Message 6 by Straggler
03-09-2009 6:38 PM


Re: please expoud..
Straggler writes:

I think your (mis)use of the terms "rational" and "irrational" is going to take you down all sorts of misleading paths here..

You are correct. They were intended for advertising purposes. The main point is more like what I've described in my post to shalamabobbi.

Stile writes:

It's not really that big of a deal.
But, on this board, it sometimes comes off that "the rigorous scientific model" is held up on some sort of golden, un-surpassable pedestal. The point of this thread is to show that this is not true. It is only placed on such a dais when truth and knowledge are the extreme priorities. Sometimes they are not priorities... sometimes exploration at any cost is a priority... sometimes other things.

Message 7

This message is a reply to:
 Message 6 by Straggler, posted 03-09-2009 6:38 PM Straggler has not yet responded

  
Sarawak
Member (Idle past 3823 days)
Posts: 47
Joined: 03-07-2009


Message 9 of 12 (502228)
03-10-2009 12:07 PM


This problem has always fascinated me, but it leaves out very practical considerations. The simple fact is that some stops/customers demand higher priority than others, so the shortest distance isn't the right criterion for determining the route, but the customer demand.

Since shortest distance is so tough, how about dumping a sort of subjective criterion of "priority" into the question?


Replies to this message:
 Message 10 by Stile, posted 03-10-2009 12:43 PM Sarawak has responded

  
Stile
Member
Posts: 3863
From: Ontario, Canada
Joined: 12-02-2004


Message 10 of 12 (502238)
03-10-2009 12:43 PM
Reply to: Message 9 by Sarawak
03-10-2009 12:07 PM


Useful for actual travelling
Sarawak writes:

This (Travelling Salesman) problem has always fascinated me, but it leaves out very practical considerations.

Yes, it does. But that's because it's not really meant to be used "just for travelling." It's more of a tool used for testing, creating and improving all sorts of computational optimization techniques.

...how about dumping a sort of subjective criterion of "priority" into the question?

I agree that such a thing would make the solution easier to attain for a real-world travelling problem. However, since the Travelling Salesman tool can be used to solve other real-world problems as well, such as the manufacture of microchips or genome sequencing. Adding a destination-priority onto those sorts of tasks doesn't really help at all. The purpose in those tasks is simply to "go to" all the "destinations" as fast as possible, with none taking a priority.

The main point, however, is that regardless of the specific improvements there do exist certain situations that can be overcome easier when not using strictly rigorous, objective, verifiable methods.


This message is a reply to:
 Message 9 by Sarawak, posted 03-10-2009 12:07 PM Sarawak has responded

Replies to this message:
 Message 11 by Sarawak, posted 03-10-2009 1:20 PM Stile has responded

  
Sarawak
Member (Idle past 3823 days)
Posts: 47
Joined: 03-07-2009


Message 11 of 12 (502250)
03-10-2009 1:20 PM
Reply to: Message 10 by Stile
03-10-2009 12:43 PM


Re: Useful for actual travelling
Actually my point was that putting in priorities makes the solution to the problem much more difficult, since the nature of priorities is quite subjective and therefore very difficult to solve with mathematical models.

But priorities are also quite practical.


This message is a reply to:
 Message 10 by Stile, posted 03-10-2009 12:43 PM Stile has responded

Replies to this message:
 Message 12 by Stile, posted 03-10-2009 1:24 PM Sarawak has not yet responded

  
Stile
Member
Posts: 3863
From: Ontario, Canada
Joined: 12-02-2004


Message 12 of 12 (502252)
03-10-2009 1:24 PM
Reply to: Message 11 by Sarawak
03-10-2009 1:20 PM


Re: Useful for actual travelling
Sarawak writes:

Actually my point was that putting in priorities makes the solution to the problem much more difficult, since the nature of priorities is quite subjective and therefore very difficult to solve with mathematical models.

Oh, I see. Yes, that is difficult.

I assumed that the priorities would be values entered by a human. Therefore, they are simply "known values" as far as the computer program is concerned. That would make things easier. But you're right, if the priorities themselves are to be analyzed/developed by the program then obtaining accuracy gets much more complicated.


This message is a reply to:
 Message 11 by Sarawak, posted 03-10-2009 1:20 PM Sarawak has not yet responded

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2018 by EvC Forum, All Rights Reserved

™ Version 4.0 Beta
Innovative software from Qwixotic © 2019