Well, you said this:
"Why is that relevant? No one's claiming the models are identical, just that regular, repeating phenomenon are predictable."
And the answer "why relevant" is in the fact that the software is not based on repetition,
which makes less relevant the part about "repeating phenomenon are predictable".
Not really, you just have to make a formula that predicts repeating phenomenon.
One simple example: precession of Mercury is not quite periodical.
Attempts to predict it correctly were failing before General Relativity.
Even if it was, you think people wouldn't understand how such formula works?
Also, let me remind you:
"Planetarium software is application software which allows a user to simulate the daytime and
nighttime sky on a computer. Such applications can be as rudimentary as displaying a star chart
or map for a specific time and location, or as complex as rendering photorealistic views of the sky.
While some planetarium software is meant to be used exclusively on a personal computer, some
applications can be used to interface with and control telescopes or planetarium projectors.
Optional features may include inserting the orbital elements of comets and other newly discovered bodies for display."
I wonder how to insert history of observation for newly discovered celestial bodies...
This is a snake we have in my country that I have seen
. I observed it attacking me when I got too close. Now, I travel (in a straight line) to Australia, and suddenly I discover this entity I have never seen before, looking somewhat like
(albeit, probably not in this position). Now, I have never seen this newly discovered creature, but still I am able to predict, based on the interaction of my home-grown snake, that if I would come to it it might attack me.
Using data on known celestial bodies you can predict what newly discover celestial bodies might do.