NavList:

Antoine, you wrote:
"help me to unravel the "mysteries" about the way you account for planetary semi-diameters."

Just ask, Antoine. :-) Planetary SDs are treated as zero in my online lunars tools. That's what was done historically. This means that when you shoot a lunar with Jupiter, for example, if you can resolve the disk through the sextant's telescope, you should place its center on the lunar limb as nearly as possible. In practice (have you tried any?), this is easy.

You wrote:
" I am especially thinking of interested Historians - there must be some - who may not have your sharp and deep understanding of all the relative motions involved."

The clearing tool was originally designed for analyzing historical lunars and can certainly be used for those cases. I use it to analyze historical lunars all the time. The business about off-angle lunars has never arisen in any of the historical cases I've examined. There's probably a simple reason for this. The predicted lunar distances in the almanacs AUTOMATICALLY removed those possibilities. You wouldn't shoot a lunar using a star perpendicular to the Moon's motion across the sky because the almanacs did not provide the necessary geocentric lunar distances. Anyone clever enough to calculate their own geocentric lunar distances, bypassing the almanac's automatic filter, would hopefully have enough sense to select bodies in line with the Moon's motion.

You added:
"May I also add that the general word "Approximate" necessarily implies - at least in my sense - that we are to remain in the same "order of magnitude", which definitely was not the case here, even if the studied case has a quite exceptional nature."

I agree that it is an imperfect word. If words were perfect, linguistics would be computer science. :) It would be "nicer" to include various other detailed options. And that's what's going into the next version of the "Lunarian" software that I've been selling. It includes an option that lets the user decide what to do with the data: use it purely as a test of the observation process, attempt to calculate a longitude from it, or attempt to use it for a "lunar line of position". Anything else we can do with a lunar distance?

You wrote:
"I am sorry to have inadvertently waived the red flag here and that you (over-)reacted to this mention of "TT-UT" value."

Ahem. It was only a "PS".

You added:
"On the other hand, I keep mentionning the exact "TT-UT" value I'am using whenever I publish data about my computed results. This remains the ONLY WAY - and by the way, do you see any other one ? - to help other "number crunchers" (I now know of at least 3 different ones now just on NavList Forum) to meaningfully check their own results against mine."

Ya know, folks publish their calculations on lines of position, altitudes, etc. in NavList posts on a regular basis. I can't recall any of our detail-conscious calculators ever including a value for delta-T in those posts. Why is that? Well, of course, it's because that is an irrelevant detail.

But for lunars, people do become obsessive about irrelevant details... Consider that delta-T is known for any year within the past century to a fraction of a second and any year within the historical period of lunars (back to the 1760s) to within a second or at most two. If you get it wrong by even two seconds of time, what impact does it have on the clearing process? Well, as we know, a tenth of a minute of arc (6" of arc) corresponds to 12 seconds of time. So a two second difference in delta-T would yield a single arc second error in the clearing process. Anyone worrying about ephemeris data at this level for lunars is missing the forest for the trees. You don't need to post values of delta-T except in exotic cases (or cases before about 1760 or hypothetical cases after the present --after 2015 or so).

You asked:
"One last item : have you tried your On-Line Computer to compute the (very rare indeed) limit cases of Lunar occultations, and if so, what results do you get ?"

No, I haven't. It's not designed for that, but off-hand I can't think of anything that would ruin the calculations. It would be rather tough to use though. With lunars, there is a degree of linearity in the errors. So for example, if you pick a DR position and enter a specific lunar distance for a specific instant of GMT, the error in the cleared lunar will provide some indication of how you should adjust your DR position to bring that error to zero. In a few steps of trial and error, you're there. But with very short distances --and presumably especially so in the limiting case of occultations-- the trial and error process swings about rather wildly. If someone has an actual practical use for such calculations, I would consider some modifications to the online tools. But if it's just a mathematical game or experiment, I probably wouldn't spend time on it ...and I wouldn't want to spoil the fun! :-)

-FER

----------------------------------------------------------------
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to NoMail[at]fer3.com
----------------------------------------------------------------