[Kstars-devel] more naked-eye stuff

[Jason Harris]

Hi Milton,

Glad to see your reply.

mph wrote:
> 1) Alternate constellation names using the same (IAU) constellation lines 
> (French, German, Chinese, etc.) [...]
> 
> 2) Alternate star names.  [...]
> 
Regarding these alternate naming schemes, is the idea to simply provide
localization of star/constellation names for non-English speakers?  If
so, then this is already handled by the current translation procedures.
  Each translation team gets the list of star and constellation names as
strings that can be translated.

If, however, the idea is to provide alternate sets of star names that
can be used by speakers of any language, then we can actually use the 
existing translations to get started.  For example, I know the Chinese 
team has already provided constellation names in their native language.

By the way, I think these alternate name sets should be an "extra"
downloadable through the "Get New Stuff" tool, rather than included with
the standard distribution; any objections?

As for the file format, for constellation names it can be identical to
cnames.dat (perhaps with a header line containing the name of the set).
For star names, something very simple like "gname : Alternate Star
Name", where "gname" is the abbreviated genetive name as in the
hipNNN.dat files.

> 3) Alternate constellation shapes (clines.dat) paired with alternate 
> constellation names.  We might need to reorganize the data somewhat to be 
> able to pair clines.dat files with cnames.dat files. Presumably these would 
> be paired according to a particular culture and potentially linked to 
> culture-specific star names, though not necessarily.  This would take the 
> most time to create (data), not only because of the number of points to be 
> specified, but also because I have found it hard to find text sources that 
> specify stars.  I know the tail of Scorpio is Maui's fishhook, for example, 
> but none of my books have drawings suggesting exactly which stars are 
> included.  Similarly, I have references to the Hindu nakshatra containing 
> Spica, but I don't know what other stars are in it.  I'm always interested in 
> such sources if anyone on the list has some references...
> 
Yes, the source for these figures is going to be a big hurdle,
especially since we'll want reasonably complete sky coverage (rather
than just a handful of constellations).  There must be *something* on
the www though.

> I had a few thoughts.  First I thought, "what if we could specify any star 
> (including those without gnames) in the clines routines?"  
Yes, we should be able to use any star.  However, the current set of
genetive names is incomplete, so if there is an unnamed star which needs
to be used as a constellation vertex, I think it's likely we'll be able
to identify a genetive name for it.

> Then, we could 
> "draw" crude figures and it would be up to the artist to put them on the 
> curved sky.  Extending that thought, "what if we allowed 'dark stars' to 
> allow more flexibility in creating the pictures?"  And then, "what if we 
> could specify splines to connect the dots?"  I don't know how practical these 
> ideas are.  
Yes, this is exactly the kind of thing I was thinking of: defining each
constellation drawing in terms of curves on the plane of RA and Dec.
That way, as the sky is distorted in the view, the drawing will
automatically distort in the same way.  In addition to splines, we'd
want to be able to change the line thickness, and perhaps add shading too.

> It does put a lot of burden on the artist, who would almost need 
> a tool set to help them create the data set.  
A tool would be nice.  Something that showed the constellation lines
(and perhaps stars) on a "flat" RA/Dec plane, and allowed you to
add/edit spline curves with the mouse.  This would be a significant
development effort, though.  Since the potential user base for the tool
is very small, it might not be worth it.

I wonder if we could adapt something like xfig to the purpose.  The xfig
file format is quite simple, so it should be easy to create a template
xfig file that contains the constellation lines and stars, using a perl
script or something.  The artist could then use xfig to add/edit
splines, and we could extract these from the xfig file with a script.

> Anyway, as Jason mentioned, 
> this would be the messiest task to program of all the ones mentioned above, 
> so I'm not going to get into any rush.  There's plenty of research I see 
> ahead to create just lists of star names....
> 
Sounds good.  As I said, the existing kstars translations might be a
good place to start.

regards,
Jason