Right now in KSP the only way to get serious science is by sending a kerbal to everyplace in the system, with no need for long term missions, like Voyager, Hubble, LandSat, or any long term satellite. I would like to see some additions and changes to fix this issue. First could be the addition of a duration mechanic where the science you get out of a project is equal to the amount of new surface area that a satellite covers. So an experiment similar to Maven will take years to explore all it needs to. But this can work the other way where an mission could require hang time over a subject to get detailed reading on weather or whatever, making a geostationary mandatory. Also why do we lose science from measurement based data? Like when we take the temp of a planet transmission loss dose not adequately explain why we only get <60% of a single time reading? I get the goo canisters and the materials bay needing kerbals to process the data, which i think shouldn't be tests that should be able to run on a probe, but a thermometer gives exact values which just needs to be broadcast. Sure with the tears of communications equipment should give us better range and data speeds. Maybe the further a satellite gets from kerban the longer it takes and more power is expended to broadcast data, until the limits of the spacecraft and or transmitter are met.