[alma-config] Re: CONFI

[Min Yun]

John,

Although exactly what Leonia's algorithm does in practices is not
entirely certain, the first thing it does must be the break-up
of periodicity as you noted.  What is nicer in effect, this break-up
progess also re-distributes the visibilities so that any obvious
gaps in the uv-plane are filled.  Again, exactly how this is achieved
is not entirely clear to me.  Leonia wants to discuss his algorithm
in detail as one of the talks for the Tucson meeting, so we can
learn more about the merit of his algorithm and any potential problems.
Bryan, for example, is worried that this algorithm is pushing the
sidelobes outside the optimized region, say 20 ro 40 synthesized beams,
and in the process produce enormously large sidelobes outside this
region.  Whether this is a problem or not is not obvious to
me.  If it is, perhaps Leonia's algorithm can have another subcycle
that breaks that up as well.  I would of course encourage anyone else
to come up with an alternative algorithm that can drive the
configuration design to more favorable parameter spaces.




					-- Min