There's a review of the Sumerians
here.
Opening moves :
Settling in place would 'waste' 10-shields-in-4-turns, while settling on grassland would 'waste' a two-food tile; which is more important ? I guess it doesn't really matter, since both kind of tiles seem to be in abundance.
Still, if you prefer to settle a forest-tile, you might as well take one of the dyes, since the extra commerce doesn't get lost, and it saves your worker 7 turns (4 chop, 3 road) to connect them. [Actually, the dye-hard gold not getting lost is a supposition - can anyone confirm ?]
That said, I'd rather have my worker mine grassland than chop forest at the start. A sequence of 'move-chop(x4)-road(x3)-move-mine(x6)' yields 11 shields after 15 turns, while 'move-mine(x6)-road(x3)-move-chop' yields 19 shields if the first citizen works the mined field (which makes sense, following the general 'food-over-shields' civ-axioma). The first sequence yields 10 shields earlier, but I can't think of anything I'd want to spend them on - in the first 15 turns, that is ! The only thing that would make the first sequence more beneficial, would be if a BG tile were discovered under the forest - but that's just plain (terrible pun

) gambling.
Another small detail is that I want the town and the first road to be on the same side of the river, so that the road really does speed up movements. Based on the minimap, those movements will most likely be north, or west or east, but not south, so locations south of the river seem less appealing.
Conclusions ? Settling NE or E is out, since our worker wouldn't have a close-by grassland tile to mine west of the river. there are no inclinations to move SW, other than revealing a couple of tiles, so the only viable options seem to be :
1. Settle in place, and if no bonus food is discovered, send the worker NE to improve that tile, then continue with the dyes north.
2. move north and settle on the dyes, hoping for a 'trail of bread crumbs' to get to bonus food, if no trail : worker again NE to mine the grassland.
I think I'll go for the second option. I can't imagine Ainwood leaving us without bonus-food nearby at demi-god level.
Early production :
If no bonus food is found, Enkidu warriors are a priority : they need to go look for it. Building a granary or not should become clear early on, based on SirPleb's excellent advice
here.
Research :
on Demi-God ? There is no way I'm gonna spend effort on a first-tier tech; not the alphabet, not the wheel and not warrior code either. Iron working would be the only tech I'd consider, but I think I'll keep research at zero, trade for alphabet, and then start writing at minimum.
General strategy :
With defensive bonuses and cheap libraries, I'll go for maximum expansion, placing my cities agressively next to neighbours, pushing their borders back when necessary. This seems logical to do for the Sumerians, so that's another reason why I can't imagine not having food-bonuses closeby.
Don't know yet what victory-condition I'll go for. I won't be able to start playing before the 15th, so what type of play takes the least effort in General ? A quick diplomacy win ?