@Mrt144; while I can see your point, I don't agree some of your maths will hold in a real game. For example;
If I'm going to be running that many specialists, then I'm going to be running representation, so Sushi will be giving twice the beaker output you've calculated here. 28 sushi resources would in this case be giving 130.2 beakers relative to ethanol's 68.2. With the advent of Cristo Redentor, there are strong arguments for running representation anyway in all but the strictest CEs.
Here my objection is that on the vast majority of map types, seafood is more abundant than the ethanol resources (and rice counts for both). Seafood also clumps more, making it easier to trade for it. 38 sushi resources isn't that hard on the larger map sizes, but I've never come close to 30 of the ethanol ones.
Finally, Sushi has considerable advantages in any city type thanks to the flexibility of food (Gp farms, production cities, food poor sites, etc). It's easily the best to found for other purposes. Founding the ethanol one is costing you an extra GP (and probably slightly more maintenance since you can't have both in the Wall Street city). Aluminium there is slightly more argument for, since it doesn't clash with Sushi, but coal is rarely that abundant. It's not bad for overseas spread though, and you could certainly add it to your science cities if you lack mining inc.
There's also the issue that Sushi shows up so much earlier than Ethanol. You could even argue that the larger cities will give more commerce thanks to trade routes, though that's a two edged sword due to health/happiness issues.