I am an experienced programmer (not in Python), who recently came to Panda3D. I think it might be interesting to see what hurdles I and others had to take, and how these could have been avoided.
So please, add the things that would have helped you starting more smoothly - I think that the collective experience will help improve Panda3D.
Some of the hurdles I had to take:
The first version I had installed had this habit of crashing on errors, and I am very happy that the new one leaves a window open with an error message and a command prompt. I propose that for the basic tutorials this window is not minimised: It took me a few hours to discover it! (Auto-hiding task bar. Stupid, I know, but still…)
Then, I found myself experimenting a lot regarding the co-ordinate system. I discovered the Point3 data structure, and that it was accepted as an argument by setPos, lookAt and the likes. Adding two Point3 strutures would not yield a Point3, however, but something that could be cast with Point3(p1 + p2). And then there was the whole relative co-ordinate thing, the “otherNode”. Some pages on that in the manual would be very helpful!
One quick proposal: cut the “Common state changes” page in two (position and colour), and add the FAQ question about lookAT to the page about position. Then, in due course, this position page can grow into a section on co-ordinates
I ended up writing the math for intersecting two planes, finding the normal of a plane, finding the plane going through three given points, etc. I am pretty sure all this has been coded already, but I cannot find it in either manual or API. So that would be very helpful as well in that proposed section on co-ordinates and vector math.
Hope this helps improving Panda3D even more!