I would love to see something like this. Possibly even a few more choices - volcanic worlds, gas giant stations, desert worlds,  barren worlds, jungles, metal worlds...

+1 for me too

This is similiar to what is found in Distant Worlds. Its an ok idea but Norseman is correct on limiting how the AI would seek planets. If there was a flavor of one planet over the other of two that are available ok, but only if the AI does not gimp itself....

choosing a Jungle class 5 over a continental Class 15 is silly....

Having some races start with the requisite techs to colonize some extreme planets is enough.  Competition occurs when species occupy the same niche.  This suggestion gets away from that.

I dislike the idea of modifying the planet class by a habitability index. Planet class has always seemed more likely to be related to usable landmass than actual habitability despite the game's claims that class 26 worlds are "like heaven" and class 4 worlds are hell, as habitability doesn't have all that much to do with whether or not I can build something on its surface. However, something that a habitability index would be useful for is as a modifier for the maximum population, the maintenance costs, or the happiness of the world in question. I would further submit that these three candidates for modification by a habitability index are more appropriate than planet class as they are things that the player can attempt to counteract if he or she chooses to do so - I can build extra entertainment centers or farms or markets and play around with the wealth setting of the planet to influence the population cap, the planetary approval rating, and the net income of the world. There's little that I can do to counteract penalties to planet class.

I would further add that a penalty to planet class based on the founding species of your empire makes little sense as a determinant of the actual planet class. Maybe you believe that all 8 billion inhabitants of those planets you invaded were put to the sword and that there's no such thing as resident aliens or immigration, or maybe you believe that citizenship is never granted to non-founding species residents on any planet (since planetary population is not the total population but rather the number of your citizens residing there, according to how it was handled in the game lore for GCII), but I don't. This argument is also an argument against penalties to population, approval, and maintenance, but at least with population and approval you can argue that aliens (i.e. members of species other than the founding species of your empire) are less likely to be citizens of your empire and less likely to be entirely happy under your rule, and so planets which are less hospitable for your empire's founding species are likely to have a higher fraction of noncitizens and dissidents. It's a lot harder to use this argument to support the idea that large portions of a planet's surface will become completely unusable based merely on who owns the planet.

That's how it worked in GalCiv 1.

PQ 15 was average, and the lowest you should aim for initially. PQ 14-10 required life support systems, which costs 5bc to maintain. PQ 9 and lower cost you 5bc for the life support, plus another 10bc per level of government (Imperial is 1, Republic is 2, Democracy is 3, and Federation is 4). A PQ 4 planet would cost you 15-45bc per turn just for colonising it.

PQ also determined pretty much everything for a planet: max population, pop. growth, approval, max tax income, max production and research, and max bonus production and research. A low PQ planet will always suck, and be a huge burden on your empire. Even after terraforming it. Unlike in GalCiv 2 and 3, where terraforming can turn those planets into decent, if not even great, worlds.

In GalCiv3 there is no way for a multi-civ-system. Ok, you can divide planets, but instantly the most populated civ will take over the others. Before we could think about different enviromental planets, this part should get fixed. I'd suggest to denie takeovers till the diplomatic situation is worse then 'neutral'.  

I wouldn't want a game that the star system was automatically closed to other factions just because you were their first. You would have to at least have an even number of all planet types to make this work. I don't really see much a difference game wise than the current system where you can have multiple factions in a solar system. Influence might be a little important. I also would like to see a astrobiological molecular description on each faction. This would be a good option on the game to pick from. We also should randomise the starting class of the planets and the other planets in the solar system.

There is always a balance beten realism and the needs of the game.  moo3 for instance tried to modal different races liking different worlds. While probably more realistic it didn't really add anything and cut down competition for planets the major resource that drives conflict in space 4x games.

I just read my previous post and realised to tell you I like the idea. The only game mechanic I disagree with is the preference for small planets unless they could use small planets just as good. But if they could use small planets just as good then they are penalised for big planets. I like the idea of preferring diferent planets.