I like the idea of experiments to get data, and deliver the data to Kerbin to get science done. Then there is the type of science and how it applies. I am also all for being limited to sub-orbital until a few experiments are done on the upper atmosphere, the you can progress to orbit, them the mun, Duna, etc... I just don't want to have to go to the Mun before I can build an orbital platform. or the other way around. I await the actual science equipment and experiments that go with it. Harv described the experiments as one-shot deals where data is collected than delivered back. I hope that the experiments are not only time based, i.e. deploy experiment, wait a day, collect results, but also include actions. such as collect data while driving a rover through the mountains successfully to make better materials. or have a "black box" part to get data from failed missions to improve computing. I am also assuming the Kerbals' Stupidity levels will factor in to the experiments as well. In the tree, I hope it is more free-form than the rigid examples from strategy games. I want to pick my own successful or unsuccessful path. I want to invent the aerosol spray can before the wheel! Nevermind lighter materials, get a better ISP and thrust! Never invent the RTG and rely completely on kolar power, triple-K batteries, and an Exercycle connected to a generator. (hamster powered probe anyone?) KSP has a silly and ridiculous side to it. please keep it. The only pre-requisite to an achievement should be the ability to construct the equipment to achieve it.