6.6 Why do the faint stars and non-Messier deep sky objects disappear when I am scrolling the display?
When you update the display's central position, KStars must recompute the pixel coordinates of every object in its database, which involves some pretty heavy trigonometry. When scrolling the display (either with the arrow keys or by dragging with the mouse), the display becomes slow and jerky, because the computer is having trouble keeping up. By excluding many of the objects, the computer can keep up and the scrolling becomes fast and smooth!
6.7 I don't understand all the terms used in KStars. Where can I learn more about the astronomy behind the program?
We have started the AstroInfo project to address this problem. AstroInfo is the beginning of an interactive encyclopedia of astronomy for which KStars will be the user interface and demo engine. Right now, it's just a collection of HTML pages on various astronomy-related topics, with a particular focus on the terms that KStars uses. These articles are presented later in this document. AstroInfo is supposed to be a community effort, like GNUpedia or Everything2. If you'd like to contribute an article to AstroInfo, please let us know!
KStars is pretty accurate, but it is not (yet) as accurate as it can possibly be. The problem with high-precision calculations is that you start having to deal with a large number of complicating factors. If you aren't a professional astronomer, you'll probably never have a problem with its accuracy. There are two occasions where it may be noticeable: eclipses, and rise/set times. As discussed below, the position of the Moon is extremely hard to predict with high precision, so you probably cannot use KStars to predict eclipses at this time.
Here is a list of some of the factors we have not yet corrected for:
The positions are not corrected for atmospheric refraction. Light bends slightly as it passes through our atmosphere, which alters the apparent positions of objects in the sky. The effect is larger for objects near the horizon. For example, you may find that the rise/set time predicted by KStars is off by a few minutes without this correction. We plan to correct for refraction in a future version of KStars.
The positions of the planets are not corrected for the ‘Figure of the Earth’. The planets are shown as they would be seen from the center of the Earth. Since the Earth is about 6000 km in radius, the direction to a planet (and especially to the moon, which is very nearby) will be slightly different from different places on Earth. We plan to make this correction in a future version of KStars.
Planet positions are only accurate for dates within 4000 years or so of the current epoch. The planet positions are predicted using a fourier-like analysis of their orbits, as observed over the past few centuries. We learned in school that planets follow simple elliptical orbits around the Sun, but this isn't strictly true. It would be true only if there was only one planet in the Solar system, and if the Sun and the planet were both point masses. As it is, the planets are constantly tugging on each other, perturbing the orbits slightly, and tidal effects also induce precessional wobbling. In fact, recent analysis suggests that the planets' orbits may not even be stable in the long term (i.e., millions or billions of years). As a rule of thumb, you can expect the position of a planet to be accurate to a few arcseconds (modulo the refraction and figure-of-earth errors) between the dates -2000 and 6000.
The moon and Pluto are exceptions to this. Pluto's position is perhaps 10 times less accurate, although its position for dates near the present epoch are good to an arcsecond. The moon's position is probably the least accurate. This is because it is quite perturbed by the Earth, and because it is so close, even minute effects that would be indetectable in more distant bodies are easily apparent in the moon.
The discussion of very remote dates is currently somehwat moot because the QDate class we use for storing dates does not allow dates prior to October 1752 (when the currently-standard Gregorian calendar was adopted). However, you can enter future dates up to around 8000 AD. We'd like to implement our own date class that allows for more remote dates in the past.
Yes, definitely! Introduce yourself on our mailing list: <kstars-devel@lists.sourceforge.net>. If you want to help with the coding, you may want to look over the To-Do list on the webpage for some ideas of what's needed, and talk to current developers about what we're working on.
If you aren't into coding, we can still use your help with i18n, docs, AstroInfo articles, bug reports, and feature requests.