Oh I see. All of these (perhaps except the last one) are what I hate about the current system, what feels weird and "unrealistic" to me and why I bothered to come up with a different solution in the first place 
Buying out caravans by steaks and stews might seem like a benefit, I call it an "exploit".
Right, I understand that. The problem is that the only remaining benefit for cooking is happiness, and the way the game works right now, that benefit is fairly small, especially if compared against something like increased risk of miasma. So without these benefits, inappropriate as they might be, cooking ends up being pretty much irrelevant to game play-- there's no reason to pursue it. Under the changes you're advocating, the game might be more realistic, but with the way I play the game, I would just stop cooking. And I don't think I'm alone. So then the question is, do you want changes that make cooking irrelevant?
As for the second one, food would be stored before cooking instead of after cooking, but it would still be stored! The change is almost merely cosmetic and it shouldn't affect the size of your foodstocks in any significant way.
Currently, a stockpile of, say, 500 plump helmets costs me 50 squares and 50 barrels. After I cook those plump helmets down to roasts, I only need 25 squares and no barrels. That's because I can store those roasts. That's why I cook. That's how cooking currenty affects stockpiles. If I were limited by some mechanic to a single roast at a time, say, by cluttering workshops, or because of prepared meal degradation, that same stockpile would cost me 47 spaces and 46 barrels. It probably wouldn't be enough of a difference to justify cooking for me. I don't consider that a purely cosmetic change.
If, as a second step, ingredients like vegetables or raw meat were made to rot even in barrels (they don't now), then we would get workshops to smoke/dry/salt/pickle them to preserve them indefinitelly. This might look as more micromanagement, but isn't, because these workshop would replace the current kitchen workshop and otherwise be almost the same. Instead of making roasts that last indefinitely, you'd make dried ham that lasts indefinitely. Not much of a change. The workshops could also be automated for less micromanagement.
I don't think it'd really affect much if you renamed roasts as pickles, or biscuits as jerky. I don't think it'd be bad to introduce food degradation. I just wonder if there'd be any purpose to roasts in that situation.
But I do agree the end result might be that it would be a bit more difficult to keep large foodstocks - even if only because you had to build more workshops and divert some workforce to meat smoking or something. Again, this is what I call "gameplay" because the current system of having 2000 roasted steaks in a cellar sounds too much like an exploit. In medieval times, people were always on the verge of famine. I'm not saying to go this far (and this suggestion wouldn't even come close), but a bit more attention to food couldn't hurt.
Well, what'd you'd have is people storing 10000 plump helmets instead of 2000 roasts. I agree that feeding your dwarfs is a little too easy right now, but changing cooking doesn't fix that.
Number three is very dependent on the actual implementantion, and whether we choose "individual cooking" or "communal cooking" or whatever. Dwarves right now use something like communal cooking. If they used my idea of communal cooking even in the new system, nothing in terms of bonuses would change. You seem to dislike the idea of "individual cooking" but please note this is only one of several proposed ways of implementation.
Individual cooking is the worst in terms of the way you changes would affect the impact of qualtity, but your changes would also affect communal meals. If a meal has to be eaten now, you can't make a dedicted cook and tell him to cook everything in the fortress, then forget about him until he eventually shows up idle. Doing so would invite famine. That slows down skill progression, which means less of an effect from quality.
I'm not actually suggesting dwarves should eat more food units in a single meal. I was suggesting to limit the supply of food in fortress mode by dividing all food sources by 10. The end result: a single cow gives 1 to 2 meat and a single dwarf still eats 1 unit of food per meal.
That would work. There's details to be taken care of though (stuff like, do you make a dog give a brain 1/10th of the time, or what?) I think that would also have a massive effect on the early part of the game. A lot of strategies that involve putting off farming, trading for food, or bringing reserves would stop working. I don't think glaciers, coasts, or deserts would be viable starting biomes anymore. I think 2 of your starting 7 would have to be farmers. So it would have a really radical effect on the rest of the game.
To answer Kohaku: I disagree. I believe that currently, cooking has a minor benefit to dwarf happiness. If food changes involved increased risk of rotting, that benefit wouldn't be worth the cost of miasma, which can be a major cost to happiness. (You get to eat only four times a year, but miasma is continuous.) As I said, if these changes were implemented, I would not pursue cooking, and I don't believe that I am the only one.
Why do you cook food? For the happiness? Because you can?
