Well, my suspicions were correct. Another playtest of My Little Vineyard with a "looser" set of rules did many things, all of which led to a better experience. Many of the rules changes weren't changes at all, but removals. Which is a good thing. Solving problems by subtraction instead of addition, or worse, addition of exceptions (as discussed a little bit later) is generally a bad place to be.
A lot of the complexity of rules centered around the removal of spoilage. During various parts of the game, most of your warehouse of supplies would be returned to the stock if you didn't use them "before winter came." Thematically, this made some amount of sense. But from a game play aspect, it really required way too much pre-planning for a player to think of, especially in a game using dice where you might not get the die roll you want. Not only did spoilage punish the player who rolled unlucky, or poorly selected a roll based on bad stategy, it resulted in a case where it "reset your world" thought various points in the game.
And that reset of the world turned out to be a problem. Early on in the design of the game, the desired reasoning was that the risk to make the better scoring wines was the potential that your supplies for it would be spoiled out of existence. As it turns out, just the fact that the higher scoring wines simply "cost more" is good enough. One of the more interesting statement on the last failed playtest which turned out to be wiser than it seemed at the time was that a game of this type should be a steamroller game...one where you start by building little things, and end with building big things. Spoilage never let you reach beyond the little stuff.
Another aspect that was changed to fit into this simpler world are the way the research books are implemented. And their change resulted in killing two birds with one stone. The research books now apply across the board to any barrel of wine you make, and you can purchase a research book at any time for yur turn for the cost of one die. Before, you only got a book when you produced a barrel...and it was only good for the type of wine you produced to earn it. Which resulted in some confusion as to when the book applied it's bonus (when you first picked it up) and tended to make player concentrate on a single wine type.
Now, with the one die pick for the book, it gives player something they can always do when there are no other actions available (the game was missing a fallback action when the was no other decent option available) as a beneficial side effect of a generalized bound-for-all book.
Anyway, I'm back to feeling good about the game again.
One other discussion about the game that we had was about defining different rounds to as doing something specific. For example, when you produce a barrel of wine, you can store it in your wine cellar, whee it will "ferment" and gain points at the end of every round, or simply ship it to score it's current value. The problem with barrels in your cellar is that they don't score for you unless you ship them out of your cellar which requires an additional action.
One of the things I liked about this system is that over the course of the four or five rounds you play, the first rounds are spent storing barrels, while the last rounds are spent shipping barrels. However, it is up to the player to decide whethat tipping point occurs. If players aren't careful, players will wind up at the end of the game with barrels stuck in their cellar scoring zero points.
Anyway, this is a choice OF THE PLAYER. There was some discussion of programming the rounds (round 1 and 2 are storing only rounds, 3 and 4 are shipping only rounds, etc). I believe that this is needless rules creep, and could prevent players from experiment with unusual strategies as they get better with the game, and falls into the category of "rules exceptions" which I don't like. Otherwise known as "the rules are this, except when..."
Chess has a few "rules exceptions" that bug the he'll out of me. Castling your king, and en passant, which are both rules that break piously discussed rules.
As it is, the game currently has one rule exception...on the first round of a season, every player must roll at least 5 (changing to 4) dice to add to the center pool. After that, it's one or more. I'm not a big fan of this exception, but it helps with first player advantage, and gets a lot of dice out in the middle of the table to start the round in a methodical manner. So, I can live with it.
I hope to have the next version of it up soon, but another, stranger project is sucking away at my time. Which I may or may not talk about depending on how that works out.