Being already in the middle of the next economic crisis, one would think that that the lessons of the previous dot.com crash have been long learned, and the thinking that lead to that bubble long discredited. Unfortunately that isn't the case. Some people working on various online projects *still* think that the most important number to count is "eyeballs". They still measure the success of anything by simply counting the number of users / viewers / visitors / players, with no regard of whether these are paying customers or not.
For example Raph Koster has found the game that has more players in North America than World of Warcraft: YoVille. While he admits that "there is little doubt that WoW makes a lot more money", he still considers YoVille an "answer" of "whether there is room to go around WoW". Sorry, no, it isn't. Social spaces aren't even games, and most of these millions of players use YoVille mainly because it is free, and it is on Facebook. Very, very, few people will be willing to spend the $200 a year on YoVille that a typical MMORPG costs. To beat WoW, a game does not have to beat it's subscription numbers, but it's revenue and profits.
On the other end of the scale some people seem to have realized that big isn't always beautiful, even if the customers are paying. Buying a copy of the recently released Darkfall is extremely hard. Darkfall only has one server, and Aventurine is only selling as many copies as are necessary to keep the population of that one server stable. Copies are sold once a day, at a random time, and are usually sold out quickly, reducing some fans to actually camping the website. Amidst jokes that even buying Darkfall is hardcore, commenters often realize that this strategy isn't all that stupid. It avoids the boom and bust fate that both Age of Conan and Warhammer Online suffered. And of course by NOT letting in everybody, it makes those who are in feel more special. Financially it produces a more steady income both from a stable number of monthly subscriptions, and a steady stream of new players buying the game. Assuming that the development cost of Darkfall was low enough, the lower cost of limited infrastructure hopefully leads to Aventurine having a decent profit margin. That must be better than buying lots of serves for release, and then having to shut them down later. A profitable company can always find investors for their next project. A company that only has great user numbers, but no profits, will have to rely on investors who like high risks. And due to the current crisis, there aren't many of those around any more.