On the subject of the FTC ordeal and "did Bioware lie?" concept...
Question: With the ending in Mass Effect 2, there were so many different variables and possibilities for the outcome and what could happen. As players reached the end, they started comparing notes and trying to figure out how it worked. A few months after it came out, we ran a chart in the magazine that showed the layout of how to get the different endings and how things happened. Is that same type of complexity built into the ending of Mass Effect 3?
Answer: Yeah, and I?d say much more so, because we have the ability to build the endings out in a way that we don?t have to worry about eventually tying them back together somewhere. This story arc is coming to an end with this game. That means the endings can be a lot more different. At this point we?re taking into account so many decisions that you?ve made as a player and reflecting a lot of that stuff. It?s not even in any way like the traditional game endings, where you can say how many endings there are or whether you got ending A, B, or C.
It?s more like there are some really obvious things that are different and then lots and lots of smaller things, lots of things about who lives and who dies, civilizations that rose and fell, all the way down to individual characters. That becomes the state of where you left your galaxy. The endings have a lot more sophistication and variety in them. It would be interesting to see if somebody could put together a chart for that. Even with Mass Effect 2?s...
Emphasis added mine.
As for the difference between Mass Effect and the Mona Lisa (or paintings in general):
Mona Lisa is unique. It's ONE. And if you don't like it, you don't buy it.
If you go to a restaurant and they give you bad food, you don't pay for it, right? Cooking is art too...
If you're an architect, and you design a building that just doesn't work properly, aren't you expected to do it over, and do it RIGHT this time?
Larry Niven has said more than once that one of the big reasons he made a sequel to Ringworld (his most popular novel) was that fans ran the math, and were walking up and down his hotel hallways dring a con shouting "The Ringworld is unstable!" His sequel RetCons many of the assumptions in the original.
Movies using test audiences all the time. As Good As It Gets originally had Nicholson's character just be a douchebag. Audiences didn't relate, so they added the OCD thing to make him more likely...
Akira Toryama planned to have Gohan be the main character in the last season, but brought Goku back because people liked him better.
When Peter David took over the Hulk, he RetConed past story out as a dream by Nightmare.
Broken Steel ALREADY RetConed the ending to a game because people shouted enough.
When a game's gunplay is considered broken or unbalanced, a patch (FREE patch, mind you) to fix it. In a game where the story is at least as important as the gameplay (arguably more) why SHOULDN'T the game fix a broken ending?
Fixing problems fans don't like is something that happens in cooking, architecture, books, TV shows, movies, comics and even GAMES themselves. What's the difference here?
As for bad precedents, giving MORE money to people who put out a flawed product is much worse than "developers give fans what they want."