Various surveys over the years showed that MMORPG players on average spend around 20 hours per week with their favorite game, that is 1,000 hours per year. As a consequence of the RealID debacle I turned on parental controls for my World of Warcraft account, and the weekly play time reports Blizzard sends me confirm that I too play around 20 hours per week. That makes over 6,000 hours since the game was released, so it isn't really surprising that I and many others (see comments in yesterday's thread) are somewhat burned out.
But still the question remains whether it would be possible to make a game in a way that players *don't* burn out when playing it 1,000 hours per year for several years. Basically there are two major axes along which a game can entertain for thousands of hours: Content and replayability. Of these, content is currently the focus of most game developers: Blizzard made a huge effort to tell more engaging stories in Cataclysm using a mix of quests, phasing, scripted events, and cutscenes. And Bioware promises epic storytelling as the "fourth pillar" of their upcoming Star Wars: The Old Republic MMORPG, with "over 50 novels worth" of voiceover text.
I have repeatedly voiced my doubts about that approach. It is not that I don't believe that games can't tell stories well. But I do believe that they usually do it well only for shorter blocks of time. If you play through lets say a Call of Duty game in 10 to 20 hours, you will have experienced a gripping story. Keeping that level of story-telling up for 1,000 hours and more is a challenge that hasn't been met yet. Even voiceover is no help when ultimately the game goes back to telling you to "kill 10 womp rats".
Replayability is about "making interesting choices", the hallmark of a good game according to Sid Meier. Unfortunately MMORPGs don't appear to have made much progress in that area in the last decade. In some areas the making of interesting choices even regressed, due to the growing popularity of MMORPGs leading to theorycrafters finding the mathematically optimal solution, thereby eliminating choice. Game developers also contributed to that, by trying to make their games more convenient, and less punishing, thereby eliminating the *need* to make a right decision. The only thing we gained since Everquest is better technology that enables us to play much faster, thus making a "do not stand in the fire" type of challenge possible. Unfortunately "not standing in the fire" is *not* an interesting choice, and ultimately boils down to reflexes and excellence of execution. It makes it possible to fill 1,000 hours of gameplay with 10 hours of wipes each for 100 different raid bosses, but that isn't everybody's idea of fun.
Replayability suffers a lot from different MMORPGs having so similar basic gameplay. An Everquest player frozen for a decade and thawed up at PAX East to play a SWTOR demo would have found the controls and combat very familiar, just prettier and faster now. If anything the choices are less interesting now, because in a modern game randomly mashing buttons has a higher likelyhood of success and a much lower penalty for failure. That carries the risk that players burn out in a new game much quicker than they burned out in the previous game. Once the player consumed all the new content of the new game, the gameplay isn't holding him, because he was already bored of that from the start.
I do believe that if MMORPGs want to ever make the next quantum leap in popularity, developers will have to come up with new forms of gameplay that offer more interesting choices, and thus better replayability. Chess has been doing quite well for hundreds of years, in spite of it having lousy story-telling. Instead of creating interactive television, which by nature is limited in the number of hours of entertainment it can provide, developers will need to remember some of the core values of game design. It is hard to fill 1,000 hours per year of gameplay if that gameplay doesn't offer interesting choices.