Text 578, 173 rader
Skriven 2004-10-30 16:55:00 av Guy Hoelzer (1:278/230)
Ärende: Re: the why question
============================





in article clrbf7$em2$1@darwin.ediacara.org, Tim Tyler at tim@tt1lock.org
wrote on 10/28/04 10:49 AM:

> Guy Hoelzer <hoelzer@unr.edu> wrote or quoted:
>> in article cllqu1$1u0c$1@darwin.ediacara.org, Tim Tyler at tim@tt1lock.org
>>> Guy Hoelzer <hoelzer@unr.edu> wrote or quoted:
>>>> in article clc03g$24sv$1@darwin.ediacara.org, Tim Tyler at tim@tt1lock.org
>>>>> Guy Hoelzer <hoelzer@unr.edu> wrote or quoted:
>>>>>> in article cl65so$b0p$1@darwin.ediacara.org, Catherine Woodgold at
> 
>>>>>>> If you mean "maximizing, under the constraint of
>>>>>>> following all the other physical laws" -- is that really
>>>>>>> any different from just plain "following all the other
>>>>>>> physical laws"?  What exactly is your claim?
>>>>>> 
>>>>>> My claim is that this IS a physical law.  There is an ongoing debate
>>>>>> about whether this claim constitutes a fourth law of thermodynamics, or
>>>>>> whether it can be accommodated by a minor modification of the way we
>>>>>> articulate the second law.  I am in favor of the latter.
>>>>> 
>>>>> Several points here:
>>>>> 
>>>>> I wouldn't even describe the second law of thermodynamics as a physical
>>>>> law.  It's not a "law" - since it permits exceptions
>>>> 
>>>> Our traditional understanding of a static 2nd law explicitly applies to to
>>>> closed systems, and there are no exceptions permitted or observed, AFAIK.
>>> 
>>> Violations of the second law are commonplace on small scales - e.g. see:
>>> 
>>> ``One of the most important principles of physics, that disorder, or
>>>   entropy, always increases, has been shown to be untrue.
>>> 
>>>   This result has profound consequences for any chemical or physical
>>>   process that occurs over short times and in small regions
>>>  
>>>   Scientists at the Australian National University (ANU) have carried out
>>>   an experiment involving lasers and microscopic beads that disobeys the
>>>   so-called Second Law of Thermodynamics [...]''
>> 
>> This is not inconsistent with my claim, because there cannot be such a thing
>> as a truly closed system inside of the universe.  Closure can only be
>> approximated at small scales.  In response to the interpretation of these
>> empirical results I can only predict that the apparent failure of the 2nd
>> law is due to either the cryptic openness of their system or a
>> misunderstanding of what went on inside.  If this prediction is false, then
>> so is my current paradigm.
> 
> Closure is no big deal - as with fluid dynamics, you can define a boundary
> and measure entropy-flux through that boundary - if you are dealing with
> an analog of the second law in an open system.
> 
> There's no problem with the experiment.  The problem is with the second
> law.  The second law expresses a statistical tendency - not an iron rule.
> It's a rule that applies when the number of states in the system is large.
> Nobody should have expected it to apply on small scales in the first
> place.
> 
>>>> So I do see it as a universal and physical law, albeit one that
>>>> appropriately leads to statistical sorts of models. [...]
>>> 
>>> It's no more a physical law than evolution is.  Evolution and
>>> the second law are *consquences* of the physical laws (and the
>>> initial conditions) - but they are not themselevs necessary to
>>> explain what happens in the world.  I.e. The physicist's "theory
>>> of everything" would make no mention of either concept.
>> 
>> I expect otherwise.  Do you know of any instances where someone has tried,
>> let alone succeeded, in deriving the second law from putatively more
>> fundamental physical laws?  Would you agree that your view of the 2nd law as
>> consequence is an unfounded (unproven? untested?) speculation at this point?
> 
> The second law is statistical.  It was derived from statistics (plus
> some Newtonian physics) by Boltzmann:
> 
> ``Toward the end of the nineteenth century, Ludwig Boltzmann showed that
>   that second law of thermodynamics is a statistical statement about the
>   behavior of particles. He proved that the molecules of a system tend to
>   approach their equilibrium distribution when started off away from
>   equilibrium. That equilibrium is characterized by a certain quantity H,
>   which is essentially negative entropy, approaching a minimum. In short,
>   Boltzmann basically derived the second law by assuming that matter was
>   composed of particulate bodies-atoms and molecules-and applying
>   Newtonian particle mechanics along with principles of statistics.
> 
>  - http://www.csicop.org/sb/2002-09/reality-check.html

I appreciate Boltzmann's statistical approach to the second law.  It might
even approximate reality better than Clausius' articulation (not
statistical), but it is not completely consistent with it.  Boltzmann said,
for example, that entropy in a closed system CAN decrease, it is just VERY
unlikely; while Clausius never allowed for a universal decrease in entropy.
 
>>>>> Care to say what you have in mind for a modification of the second law?
>>>> 
>>>> I think that I've posted this before on sbe.  I would offer the following
>>>> conjecture as a universal physical law:  "Thermodynamics causes an
>>>> exploration of paths (e.g., mechanisms, processes) affecting the rate of
>>>> entropy gain within closes systems (I suspect that the whole universe is
>>>> the
>>>> only truly closed system), and favors those that maximize this rate."
>>>> Favoring these paths is equivalent to the structure of the universe taking
>>>> the path of least resistance as it "falls" toward higher entropy levels.
>>> 
>>> In some respects, that seems weaker than the second law - in that it
>>> doesn't explicitly say that entropy can't decrease.
>> 
>> The static second law is completely subsumed by the one I stated in today's
>> universe due to the scale and particulate nature of the universe, and
>> because it implies that the absence of bias favoring decreasing universal
>> entropy. [...]
> 
> To "completely subsume" the second law - IMO - it would need rewording to
> explicitly state that entropy can't decrease.

It would not need to do this to be consistent with Boltzmann's view,
although I agree that it would to be consistent with Clausius.
 
>>> I agree with the spirit of this sort of idea - though as we've
>>> discussed, I have some issues about whether such maximisation
>>> will necessarily happen in the short term - since organisms
>>> can act to conserve resources in the short term under some
>>> circumstances - as a squirrel burying a nut demonstrates.
>> 
>> To reiterate my previous response, if you think that this is a counter
>> example to my argument, then you have not fully appreciated the argument.
>> Everything that the squirrel does increases universal entropy.  The law I
>> and many others are advocating as conjecture does not imply anything about
>> those activities that do not occur, such as eating every nut as soon as
>> possible.
> 
> In that case, it seems very unclear what it does imply.

Note that it shares this feature (being about what does happen as opposed to
what doesn't happen) with the views of Clausius.
 
> The squirrel deliberately refrains from maximising entropy in the short
> term - something which it is evidently capable of doing.

Your claim here is not true, given the definition of "maximization" that I
espoused in a previous post.  The squirrel is increasing the extent of
universal entropy at every instant, and thus over every time scale, of its
existence.  There is NOTHING it can do to decrease universal entropy, and
the only way it can stop increasing universal entropy is to die.  I still
don't see this example as inconsistent with the conjectured, dynamical
version of the second law.  Since you seem unwilling to accept, even
operationally, the meaning of the term "maximization" as I defined it, it
might help to avoid confusion in our discussion if you replaced "maximize"
with "increase" in my statement of the second law.  I still argue that my
usage is consistent with that of many (if not most) theoretical physicists,
and I think it is a more appropriate term in this context; but you would not
loose much of the general point to use "increase" instead.
 
> To understand why I see no alternative to considering the
> timescale over which the function being maximised is measured.

I don't understand this sentence.  Can you elaborate?

Cheers,

Guy
---
ū RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info@bbsworld.com

---
 * RIMEGate(tm)V10.2á˙* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 10/30/04 4:55:54 PM
 * Origin: MoonDog BBS, Brooklyn,NY, 718 692-2498, 1:278/230 (1:278/230)