What I beleive people want to have are random QSO's that they initiate themselves and that are aren't controlled by anybody else.
Agreed. Like we do on HF to work a DXpedition - i.e., K5D.
The drawback, or shortcoming, to the "take a list of calls and confirm the list" approach is that - unless the station taking the list hears a confirmation to his/her report from each of those stations in the list - I'm not sure that is really a contact. It's more like an exchange of SWL reception reports. Having one station act as a net control etc. to orchestrate the activity then brings the same questions into play as on HF.
If you get more than half a dozen Amateur operators active in the footprint of a single channel FM satellite they'll inevitably be congestion issues.
This is true, just like on a terrestrial repeater. Either everyone cooperates or there is congestion that makes it useless. With cooperation, there could be many more stations on a pass that get a chance to make a contact with someone in a rare country or grid (I've worked as many as 26 on one AO-51 pass, last November from DM22/DM32).
We'll just have to live with that for those satellites that have already been designed, but for future designs a linear transponder seems the obvious choice. A 40 kHz linear transponder would allow for multiple SSB QSO's over areas with a high Amateur population but would still enable the use of FM in the many areas of the world where there are few Amateurs.
This is a good idea, as demonstrated by VO-52 where hams in South Asia are encouraged to try it in FM and the rest of us are to use SSB or CW. But don't overlook the interest in FM satellites for the areas with high ham populations, also. Otherwise, there goes a low-cost entry to amateur satellites in those areas. KiwiSAT is interesting, in that it will have an FM repeater and a linear transponder.
Patrick WD9EWK/VA7EWK http://www.wd9ewk.net/