Dilvish wrote:It looks to me like come moderate code juggling could let star and planet choosings be done before naming, and then more of the visibly named systems could wind up with the solo naming scheme.
Sounds like a good idea to me.
Well, the current (pre-implementation) feedback seemed to be that folks liked the idea of a fair bit of grouping even when the number of stars didn't especially require it, so 400 stars being mostly "Whatever" isn't what I was shooting for. But I can rework it so that the ratio of "Whatever"s to group names is more controllable.
Ok, I've generated several galaxies to get a feeling for the thing, and these are my impressions so far:
First of all, I like the idea that at least some of the systems get these "group names" even in galaxies that aren't big enough to use up the entire starname list. They add a very nice flavor, so far I'm in agreement with the others
However I think that the amount of systems with group names should be far less, they should only make up a fraction of all the systems (if possible, very large galaxies can't avoid that of course). Several reasons:
- If the majority of systems has group names, that actually takes a bit from the special flavor they add, for the very reason that they aren't special, but common it that case. Of course, that's a very subjective impression, others may feel different about that.
- You get too much system names that are both similar and long. Especially in larger galaxies I'm staring at a wall of Alpha, Beta, Gamma whatsoevers. It's of course far better than a wall of autogenerated catalogue names, but compared to having most systems with "normal" individual names it's still notably worse.
- Prefixed system names being significantly longer also leads to name clutter on the galaxy map when the majority of the systems has group names. In my couple of tests I got much more overlapping, both of names overlapping other names, and names overlapping system icons, to a degree that was more than just a minor annoyance. You have to zoom in quite close to reduce the overlapping clutter so it becomes sufficiently bearable.
So I'd suggest maybe 10, max 20% of the system names should be group names, unless, as I already mentioned, the galaxy is too large. In that case ramping up the amount of group names accordingly can't be avoided. As it was mentioned elsewhere, with very large galaxies every naming scheme we might come up with can't avoid such drawbacks (unless we provide a name list with 10,000 unique names...
).
Dilvish wrote:Geoff the Medio wrote:Then maybe start the empty space numbering from the Omega end of the prefixes? Stars get Alpha, Beta, Gamma, Delta, ... and empty space gets Omega, Psi, Chi, Phi, ...
the way I was starting to plan on reworking it, I would simply place the empty systems at the end of a group, so the initially visibly named systems would all have the lower order prefixes. Any systems gaining a star via nebulae might then still wind up creating an apparent gap in the naming, but I think that's a pretty minor issue I wouldn't worry about.
That sounds like a very good solution to the problem. But that approach hasn't been in your 2nd version of the patch, has it? Because I when I tested with that version of your patch, I've seen no such ordering taking place.
Finally some nitpicking concerning your patch: Can you name your Python functions according to the xyzXyz scheme instead of xyz_xyz? The latter is the pattern for variable names, that's the scheme I've been trying to follow in that Python script, and I'd prefer if we try to stick to some consistency here - if you don't mind, it's not really
that big of an issue for me. And maybe break up your code blocks by throwing in an empty line now and then, IMO that's a bit easier on the eyes.
Gosh, I'm starting to sound like Geoff...