🚀 go-pugleaf

Reposting my reply to reach a wider distribution.

Dan


In article <1992Nov10.194812.101645@watson.ibm.com>, platt@watson.ibm.com (Daniel E. Platt) writes:
|> In article <2191@eagle.ukc.ac.uk>, aw3@ukc.ac.uk (A.Wilson) writes:
|> |> In article <1992Nov10.033543.21908@murdoch.acc.Virginia.EDU> crb7q@kelvin.seas.Virginia.EDU (Cameron Randale Bass) writes:
|> |> >
|> |> >     Since my first posting made a less-than-resounding thud, like the 
|> |> >     plodding pedantic humourless engineer I am, I figured I'd try again
|> |> >     (and since I have a couple of minutes to kill before the next run ends).
|> |> >
|> |> >     Several people have complained lately about the subject matter
|> |> >     around this neck of the woods, so I wonder if some of those people
|> |> >     will oblige me by posting what they believe to be compelling
|> |> >     mysteries or pressing problems in their particular field.
|> 
|> <lots of neat worthwhile problems in cosmology and particle physics pointed out
|> by CDB deleted...>
|> 
|> |> >
|> |> >     I'll save the rest of my rather voluminous mental 'list'.
|> |> >
|> |> >     Now surely many of the people reading here have a physical problem that is
|> |> >     compelling enough to discuss; compelling enough to wonder aloud what the
|> |> >     ultimate solution might be.  Surely large numbers of us have a list
|> |> >     of compelling physical problems that we are dying to share with the
|> |> >     rest of the sci.physics world?
|> |> >
|> |> >     Now scout's honor, I personally promise not to comment in any but a very
|> |> >     very positive manner if anyone wants to bring up anything, anything
|> |> >     at all (even chaos).  I cannot guarantee anyone else, but if you stick 
|> |> >     your neck out, I promise that the bleeding will ultimately be minimal.
|> |> >     And you'll have the side benefit of greatly increased typing speed
|> |> >     like the rest of us loudmouths.
|> 
|> Chaos is one of the issues.  One of the first monte-carlo computer simulations
|> was Fermi, Pasta, and Ulam's simulation of the ring of non-linear springs.
|> They figured that the non-linear terms would scatter the energy evenly throughout
|> all the modes of vibration... only it didn't.  This turned out to be an example
|> of the Kolmogorov, Arnold, Moser (KAM) theorem, in which it became clear that
|> non-linearities does not guarantee that the measure of phase space not sampled
|> by a non-linear process is zero.  What is the mechanism by which ergodic behavior
|> emerges?  Choas studies the approach to divergent or random behavior, as well as
|> sensitivity to initial conditions (homeomorphism with baker transformations,
|> for instance, point out lots of details), related to mixing, seem to show large areas
|> of behavior that are not ergodic even though they are chaotic.  Then there's the
|> question of why individual trajectories seem to come to thermodynamic equilibrium in
|> times so short compared to the time required to sample all of phase space.  While
|> mixing theory is supposed to deal with some of these questions, by pointing out that
|> losing correlations over some short relaxation time, it is primarily described in
|> terms of the relationship between different trajectories filling phase space in a
|> short time due to sensitivity to initial conditions (chaos) in order to lose the
|> correlations.  Then there's the problem of quantum chaos and the relationship to
|> quantum ergodic theory -- what's the quantum analogue of the classical problem?
|> 
|> There's a lot to the chaos problem - particularly in the context of foundations of
|> statistical mechanics that seems to be open questions worth looking at.
|> 
|> |> >
|> |> >     I like a bit of wonder and mystery in the universe, it gets so boring
|> |> >     when a common attitude is that everything is fundamentally solved.
|> |> >     And actually, I'm just looking for things to add to *my* list. 
|> |> >
|> |> >                                   dale bass
|> |> >
|> |> This highlights a point I raised earlier (no offence) that the great
|> |> unsolved problems in physics are by no means restricted to cosmology, 
|> |> GUT's and solar neutrinos, wlthough these are interesting to those
|> |> concerned, and to the rest of us no doubt in a more general way.
|> |> There must be other areas of physics with equally interesting and
|> |> controversial problems, some of them possibly with quite profound
|> |> implications for our understanding of nature, yet never seem to get
|> |> the same publicity as the old "Physicists think there may be a black
|> |> hole in the centre of Washington..." stories. I would be interested to
|> |> hear about the great problems in other fields of endeavour apart from HEP
|> |> and cosmology! Any comments?
|> |>
|> 
|> Without getting too mundane... having suggested an area that has lots of 
|> opportunities, there's also the question of technological opportunity
|> (back to engineering -- appropriate for CD Bass's post).
|> 
|> We have chip designs based on 'devices' (diods, transistors, etc) built
|> on the behavior of depletion layers which depends on the way charge
|> carriers introduced by dopants (holes and electrons) behave near the
|> Fermi surface along bands, which define the effective dynamic masses
|> of the carriers, etc.  Most of that is fairly easy and straight forward
|> to build and design, using simple rules, such as the law of mass-action,
|> that depend implicitly on the curvature (mass tensor) of the bands.
|> However, as chips are made smaller and smaller, the assumptions that went
|> into constructing bands, understanding how the Fermi surface works, how
|> the law of mass-action behaves, etc, becomes much more complicated.
|> The electronic behavior of simple atoms and molecules is doable (called
|> chemistry); techniques in solids with fair regularity is getting to a
|> place where its sort of doable (variational techniques in density functional 
|> theory have been applied to the behavior of diffusion of defects in
|> silicon lattices... there's still work going on in that area), and for a
|> long time have been in a place where design was 'easy' to do.  However,
|> that middle area, where the size of 'devices' are getting small enough
|> for band structures and interactions to be strongly influenced by the
|> small sizes of the devices (mesoscale electronics).  I expect that this
|> is one of those practical barrier whose solution will bring some $'s for 
|> a while to the company that actually brings a solution to market... if 
|> it is doable.
|> 
|> Another interesting problem -- after claiming molecules are doable -- is
|> the way protiens fold, how other peptides interact with proteins,
|> why some proteins can be easily converted to cancer-producing states by
|> replacing a residue, etc... even the problem of protein folding is
|> very difficult to do.  Many simulations have failed miserably to do
|> much of anything successfully on this problem -- yet it is just that
|> kind of problem that will lead to new solutions to drug designs etc.
|> Apart from practical interest, it is of real interest to try to
|> understand this problem in order to understand the feedback loop on
|> protein replication with DNA/RNA and protein function... what came
|> first, the chicken or the egg? -- or at least what kind of description
|> of the processes of how it came at all might be constructable?
|> 
|> Dan
|> 
|> -- 
|> -------------------------------------------------------------------------------
|> Daniel E. Platt                                      platt@watson.ibm.com
|> The views expressed here do not necessarily reflect those of my employer.
|> -------------------------------------------------------------------------------


-- 
-------------------------------------------------------------------------------
Daniel E. Platt                                      platt@watson.ibm.com
The views expressed here do not necessarily reflect those of my employer.
-------------------------------------------------------------------------------