*) Added a MemoryLimit variable. Octopus will stop if this variable is set and if the total memory depasses its value
*) Octopus was not compiling in 4D due to array mismatches

I guess I made a disaster in your opt-control code (that now will not 
wortk in single precision). The problem are the minimizing routines that 
do not have interfaces in single precision more. One should add it...

of octopus 2.0.

I generated it through the command:

cvs2cl.pl --fsf -l "-d >2005-06-01"

The command cvs2cl.pl seems to be pretty standard (at least it is in the
debian official repository). Basically it uses the output from the cvs log
command -- so all the comments that we do when committing code.
The option "-d >2005-06-01" selects logs more recent than 2005-06-01.
Otherwise the file would get too large. The choice was completely
arbitrary.

Note than commiting a revision to the ChangeLog file, the way to do it
is in my opinion to commit it isolated (just the ChangeLog file), and
putting no log in the commit message (it is sort of redundant to log
the ChangeLog). Of course this first one is the exception.

Second step at rewritting the output systems (only td left)

funny thing with strings (hard to explain... fortran...)

Started rewritting the output layer.

Added split-operator propagator (do *not* use with velocity gauge)

Added share directory to put pseudopotentials

Cleaned up non-linear core corrections (faster, smaller)

Bug fix related to the reading of OutputPotential...
Added utility to calculate harmonic spectrum from accelaration file
 (currently only works for local potentials)