Glad you're interesting in JME3.
Im is also working on a project whose goal is to create educational programs (young and adult) in the gaming environment.
If you are going to let your children learn programming through game development, this is a good idea. But JME3 and Unity are much harder to get started (I suppose your children are still young) ... There are also several projects suitable for children to visually program programming.
Greenfoot ww.java.com/en/java_in_action/alice.jsp
Alice ww.greenfoot.org/door
Kojo ww.kogics.net/sf:kojo
Those things (languages ​​with an IDE) have a short learning curve and are easily accessible, require minimal knowledge and are suitable for children and starters. This is the educational side.
For the engine side. [It's my personal opinion]
I preferred JME. I also left Unity (made about 4-5 commercial games in unity of myself) to go to JME. Before Unity, I also worked in Ogre, UDK, Torque and many other engines (10 more). I also worked with a commercial engine in C ++ daytime work, whose code is dirty but flawless, and it cost millions of dollars.
The reason is that these mechanisms have tied you to their limitations and pre-authorized solutions. Of course, this is also half the reason you choose and move first. But when you click this restriction, for example, a license fee or proprietary technology. You will hate them as much as I do.
So that’s why I came to JME in search of “complete” game development and entertainment technology.
If you are experienced Java and C #, combined with a JME and Unity developer, I’ll give you a few things that can be * strong text * compared to two:
Please note that I still usually use Unity and JME3 at the same time, for my work and for my hobby. I used Unity 4 with fancy animations of mecanim, substantial stuff ... per day and using JME3 to research and improve it at night. IMO, JME3 - a game engine that is worth exploring, and it will swing and shine in the future !!!
Hope this help!