Regarding the use of fire, I’ve always been intrigued by how early Homo was able to continue its trek northward (ex. Dmanisi) without it. It would seem that a traveling hominid would frequently find itself out in the open (at night!!) without access to secure shelter while, at the same time, it was also experiencing more dramatic seasonal changes.
I understand that the two-stone method of making fire isn’t particularly easy for an amateur. It would seem, however, that bashing rocks together to make tools on a dry savannah for a few thousand generations would have produced a clue as to how this worked. In fact, I would be surprised if they weren’t accidentally burning the neighbor-hood down on a regular basis. Maybe the initial production and control problem was learning how to put all these blazes out, not how to start them.
There is evidence for fire in Swartkrans Member 3, which may be as old as 1.5 million years. The really good evidence from Gesher Benot Ya'aqov is sufficient to demonstrate control and habitual use of fires by 800,000 years ago. So it is not a safe assumption that the early occupation of temperate latitudes preceded fire use. If a 1.8-million-year-old site had evidence of fire, I think few of us would be surprised.
The fire drill was repeatedly independently invented in different populations during the Late Pleistocene and Holocene -- it's one of the classic examples of diffusion and independent invention in cultural anthropology. So friction methods for fire making seem intuitive enough that humans come up with them again and again. To my mind these is easier and more consistent than the rock striking method, but who can say for sure?
It does leave the question of why the systematic use of fire for landscape control is so late.
