1)No, a SSD-reader is a very high tech thing. But it is durable enough to last you for 45-70 years, during which you can copy all of the blueprints for it and everything required to build one to a longer lasting medium. Assuming you streamline your culture towards creating one of them, the gap between when the first one finally fails and when the next one is successfully finished will probably only be 2-3 hundred years. Assuming you seal up your drives when the reader breaks they should be still good when you manufacture your first reader, and in a worst case scenario if you have one of your children focus their life on copying blueprints and information while the first reader still works you should still be able to obtain much more information from the reader and a single drive then you could store in dozens of textbooks. A quick calculation shows that (assuming 1 TB drives) you can get about a 35 to 1 ratio of drives to textbooks (using my smaller Electrical engineering textbook and the
OCZ Octane as patterns), and the 1 TB drive can hold the equivalent of 10,000 textbooks worth of information, giving you a total information to space ratio of 35,000 to 1 with the drives. The idea is to trade about 50-60% of the lifetime of your textbooks (about 100 years) in exchange for the ability to store (35,000 x 'number of textbooks') times the amount of information.
2)In that case simply start with enough samples for all 3 (that would be around 6 vials, though you might want to go up to 9 to be extra safe) including the liquid nitrogen for the first week and inseminate over that period of time. You will still increase your genetic diversity quite a bit, and assuming you plant wheat ASAP they should all still be able to help quite a bit at harvest time (common spring wheat has an average growth time of 6 months). Since you only need to feed 5 people (8 counting the new children) getting through the first winter shouldn't be too difficult even with reduced labor at harvest time. After that you may want to stagger the pregnancies a little to ensure steadier manpower.
5)Take ingots, not ores. Ores require further processing that you won't have access to for a while, where ingots can be used immediately. Also rather then worrying about finding ore, why not just take a
map with known ore locations?
6)Paint can be easily created from natural materials, how do you think it was done in the past? For example here is a simple 1 1/2 quart recipe for flour paint.
1 cup flour
5 1/2 cups cold water
1 cup screened clay filler
1/2 cup additional powder filler, such as mica or sand
Gold on the other hand is heavy, unwieldy, fairly useless (many ancient cultures didn't value gold at all since it is useless for 99.9% of tasks). Within a few generations nobody will care if you make it out of gold since they won't value gold, where as if you make it out of a metal that can only be made with current technology then it becomes a one of a kind for hundreds of years, instantly raising it's value.
As for government it probably won't be worthwhile for you to plan one. Most likely you won't even break the 50 person mark with your "tribe" before your own death, so having a "government" is a little ridiculous. It will probably be at least 6-10 generations before you even have more then one village present, and probably at least a dozen more before you are spread out enough to be considered a tiny "country". As it is anything beyond your simple "village council" idea won't really be needed, though you would want to establish records of democracies that way when it came time for your 20-30th generation to finally set one up they would have a guideline. Of course if your teacher is expecting one then by all means go ahead and make one, but in the event of this actually occurring it would probably be a waste of time.