One delay may be regarded as a misfortune. Two seems like carelessness.
I appreciate wanting your product to be 'the best it possibly can be', but I only accept that excuse from developers once per title, and DA:I was already postponed for a year. When a product has multiple delays, especially lengthy delays, that tells me that the studio has very poor time management skills.
Such an excuse wouldn't fly even once in other industries, it doesn't even work in school.
But I don't think DA:I is going to be delayed again.
The difference with school and a lot of other stuff is that it's not like delays don't come with a cost (it's not cheap to delay a title).
Plenty of other industries have multiples delays too. The issue comes with whether or not, when it comes time to delay, if it's deemed worth it to spend more money. If it's concluded that spending more money will result in a net increase in revenues to cover the further expenditure, it's useful to do it. Note, that even if it's been a financial mess up until the point and you've sunk million and millions of dollars into it, if you think that an extra $100k will net you $500k in revenue, you should delay.
The issue is that frequent delays say "we have low confidence that our assessments will be accurate." This creates a risk.
With regards to video games, the customer honestly shouldn't care, in my opinion, about delays. The only thing that matters to me as a gamer is whether or not the released product is something that I enjoy. While a game that takes a long time in development is not a guarantee that the game's quality will be good, I will take a slightly altered perspective: any game that is delayed is probably better served by being delayed than released at its current state. Even a game like Duke Nukem Forever, which took a looooooong time, likely doesn't provide a superior experience if it was released any time before its eventual release. The only way an earlier release date for DNF would make for a better game is if they were also able to change other aspects of production. That's the kicker, and that's the real issue for a game being subpar. Unfortunately, by the time you release the game is subpar, you've already spent that time in production on something that is subpar.... you can do one of three things at that point: delay and assume the risks that come with that; release the game as it is; cancel the game outright.
But if a developer feels the game isn't even good enough to release yet, I'm skeptical that me demanding that they actually release it would be an advantage. Canceling the game is no impact (I'm not currently playing the game anyways), and delaying the game simply means that I may play it in the future... it's not suddenly stopping me from playing the game.