Stephen Tonkin wrote:

"...I think it means that there is a bug in the program --
no doubt Bill will track it down and fix it."

That is, of course, exactly the case. Seems the parallax
stored for Polaris in Tycho-2 is 8 +/- 0 milliarcseconds,
putting the star exactly 125 parsecs away, no plus-or-minus
to it. The code for "inferring" details about stars didn't
know quite what to make of this. In theory, it would have
decided to show the result with infinite precision. Instead,
it generated garbage. Shouldn't be too tough to fix... most
likely by causing Guide to realize that such a precision is
impossible; with the parallax given in milliarcseconds,
for example, it's gotta be at least +/- 1 milliarcsec.

-- Bill