[Kstars-devel] Star catalogs (was Popup menu usability)

[Akarsh Simha]

> > > One thing I dislike in current catalog handling is cheating about
> > > spectral class.
> >
> > It is not cheating. It is inferring from the available data.
> >
> My point is that such inferring is bad. Maybe cheating is not right word.
> It wasn't measured. It's kind of "We think that it's correct", not "We 
> measured it (and think that is done correctly)". Moreover Akarsh says that in 
> some cases it was guesswork. One cannot rely on such data. B-V itself would 
> be more useful. 

Lack of data! I was talking to Médéric yesterday and we were
discussing the possibility of using SDSS for more accurate data. But
again, that will cover better data for a region of the sky.

[On a side note, our star data is much better than our NGC/IC data. I
think we should really do something to fix the default NGC / IC
catalog. I'm considering merging the NGC stuff from SAC into our
catalog]

> > > Tycho2 doesn't provide spectral class information (GSC and USNO
> > > too). But in details dialog it has spectral class like 'F?'. Derived from
> > > B-V presumably. Not a nice thing to do.
> >
> > Do you have any better idea? We won't get the spectra of 100 million stars
> > anytime soon. Even gaia will provide spectra only for a tiny fraction of
> > them. The colour is a relatively good indicator of the spectral type. Far
> > from perfect but still the best in our case.
> >
> In fact yes. Do not provide unavailable data at all. (-:

Considering the amount of discrepancies astronomical data have, I
think our guess for spectral type should be okay. We require the
spectral type to draw the color. It might be a good idea to have
spectral classes like F? as they are likely to be correct (as far as
my knowledge of astronomy goes). Maybe we should not guess B and V
magnitudes when they are not there using the fact that they are most
likely to differ by 1.6 - but I think these are very rare cases.

Regards
Akarsh