This article is evidence to the whole long tail concept all over again, but now in the gaming industry.
It doesn't help the big players that a lot of the big games are franchises (what's the difference between Call of Duty: Modern Warefare 2 and Call of Duty: Black Ops?) that are nickle-and-diming players through Xbox Live and PSN. It costs more to buy a few hours of new content in Fallout: New Vegas than it would to buy many indie games.
Braid: A very successful indie game distributed entirely through relatively traditional channels (e.g. XBLA, Steam), in contrast to the other two. It was also rather unexpectedly innovative in a realm dominated by relatively generic platformers.
Touhou: Almost certainly the most successful indie game series in history in terms of the total profit across the franchise it created -- but probably not as profitable as Minecraft for its creator because he intentionally limited his monetization of it.
Minecraft: Probably the most successful indie game in history in terms of net income... and the topic of the article, obviously.
That's not even true. Adjusted for inflation, the most expensive movie was made in 1963:
http://en.wikipedia.org/wiki/List_of_most_expensive_films#Mo...
That rather supports the assertion that making blockbusters is becoming more expensive.
Like the movie industry, it's becoming easier and cheaper than ever to make professional-looking films, but making "A-grade" blockbusters is becoming more and more expensive.
This sort of transition inevitably hurts the biggest-budget producers, as they're forced to spend more and more money to compete. Similarly, it helps the lowest-budget producers, as they can make millions of dollars off what would have been 10 years ago been considered a hobby project.
People tend to look at the most absurdly successful indie games like Braid, Touhou, or Minecraft and dismiss them as one-hit wonders or exceptions to the rule. But how many exceptions to the rule it takes before the rule becomes invalid -- especially as games begin to compete more and more on features besides graphics?
Many of these features are cheaper to develop and benefit from the agility and risk-taking abilities of a smaller development team as opposed to the designed-by-seventeen-committees-in-separate-cupboards approach of the big game studios.