Hi Andy,

My goof; when I updated the Web page to describe the current
"in-testing" version, I didn't correct the text at the top that
basically says, "there is no 'in-testing' version. If you go
back to

home.gwi.net/~pluto/devel/find_orb.htm

it should (correctly) describe the "in-testing" version. Which
does have some very nice new features (especially the uncertainty
data), and I'd recommend using it. In fact, if you're using .rwo
files, I _highly_ recommend it, to get around some known problems
in the "stable" version caused by a format change.

As far as I know, the "in-testing" version is actually quite
stable... at least, nobody has told me otherwise. Bug reports, either
on this or in the officially "stable" version, are always welcome.

I've been doing a fair bit with Find_Orb since then, and hope to be
able to post some interesting improvements in a bit. The computing of
uncertainty data works Just Fine (still not totally happy with how it's
displayed to the user). I've been working on some code to make it
easier to link widely separated arcs, so you can take an arc of a few
nights from 1990 and link it to a few nights in 2011 (this has been
nearly impossible before now). And handling three-parameter (A1, A2,
A3) comet solutions, and a variant for objects like 2009 BD where the
non-gravitational forces are just inverse-square, and computing circular
orbits, and so on...

Problem is, I also made improvements in the initial orbit determination
routine. This still needs some work. My eventual plan is that Find_Orb
ought to just load up astrometry and find a good orbit, even for some
rather unusual cases (SOHO comets, for example). But it's not quite
ready yet.

-- Bill