	II. Explanation of the structure of data directories

Different directories such as Data0, Data1, etc. correspond to
different impurity distributions generated by a different random
seed.  Within each directory, there are several files which specify
the parameters to be used in the simulation and directories
corresponding to the different electric fields.  Some important
files are:

elec: specifies the electric fields to be run at.
	The format is: 
	line 1: number of electric fields
	line 2: value of an e field
	line 3: directory name in quotes and ending in a slash
	line 4: next value of e field
	line 5: next directory name
	and so on.
	Notice that if you put a small number like 1 in line 1,
	then the program will read only the first field and 
	ignore the rest of the file.

elecall: same format as elec
	The purpose is that elecall will contain all the electric fields,
	whereas elec can be highly variable because you may want to
	rerun a certain field or run a few fields for more
	iterations.  Thus programs such as plotbyc3(plots the cdw
	configurations of all the e fields in elecall) read from
	elecall instead of elec.

parameters: This file contains four parameters:
	(1) dt, the time step taken each iteration
	(2) diffmax, the maximum allowed difference at any point
	in the cdw configuration after one period.  This was
	more relevant in the FLR model than the phase slip model
	because in the FLR model the cdw configuration is strictly
	periodic above threshold.  Specifically, if after one
	period, the new configuration minus 2pi is not different 
	from the old configuration by more than diffmax at every
	point, then the program considers it to have converged
	to an equilibrium state.	
	(3) savefreq, an integer.  The frequency that the vector
	phi is to be saved out to a work file called phi3.out.
	savefreq=1 means that every time the configuration travels
	by 2pi, it is saved to phi3.out.  This is so that if the
	machine crashes in the middle of the simulation, we have a
	saved state from which to proceed. In addition, the velocity
	as calculated by 2*pi/period is saved to data.out.  As with 
	diffmax, this variable is more relevant to the FLR model.
	(4) random number seed: the random number seed to use
	to generate the impurity distribution.  Caution: the
	program (actual code in init3.f) will reject positions
	which are too close together.  Otherwise the simulation
	will run into problems of numerical instability which 
	requires one to use an incredibly small time step and
	wait for an extraordinarily long time.

timeout: specifies the length of time to run the simulation

x: the impurity distribution shared by all the fields.

windowsize: the size of the window (y-axis) which you want the
	plotting program to use when plotting the cdw configuration

In addition, there are all the different directories like E1.3, E0.7, F0.7,
F1.3, etc.

They correspond to different electric fields.
The ones beginning with E were the first data obtained,
some from an erroneous alpha factor and phase slip algorithm.
I later ran the correct algorithm on these configurations.
But since the behavior of the system can be drastically different 
depending on the starting configuration, I started the 'F' line  of
directories, whose configurations have the following history:
Start with Phi=0. Run on high field for t=5000.0. Then using this
configuration, go down to a lower field.

In short the directories starting with F, e.g. F0.7 are the ones I
was working on right before I leave.

Now within each subdirectory like F1.3, there are many files
which may begin to confuse you (and me) because I was experimenting
with several things and didn't have time to organize.
Take Data0/F1.3 for example:

The numbered files like phi3.beg0 are old copies of the original
file like phi3.beg.  I "mv" it so that I can compare old and new
results. 

The details of what .beg .end .mbeg .mend mean are boring, and 
not particularly elegant.  I was experimenting on them on the last
week, and thus created a mess.  The idea is the cdw-mon, which
monitors spring strains, writes the initial config to phi3.mbeg
and runs for a while, monitoring, and then writes the final
configuration to phi3.mend.  Please modify the programs to produce
cleaner organizations.

Current: current at different time.  The sampling rate as well as
total length of time is given in fftparam.

VEL: the average velocity over the whole run of individual points

data.out: quite useless

fft: fft of current vs. time

fftparam: the parameters for cdwcurrent, sampling interval and total
	sampling time

spectrum: power spectrum obtained by taking the square mag. of fft

spring.mon: produced by cdw-mon, monitors if the strains in springs.
	details of what they mean can be found by reading the code
	in loop-mon.f

These four files are not too elegant either.  I wanted to know
for how long the cdw configurations have actually been run since
their creation.  Thus whenever the program finishes, it would append
the time run to the end of the file.  I hadn't quite ironed out
all the bugs though.
time: time run by pcdw
time-mon: time run by cdw-mon
timeall: time run by all
timelast: time last run
