While I'm not in favor of content that's too easy, that doesn't mean I think every aspect of every game has to be hardcore, either.
I come at MMORPGs from (among other places) a psychological direction. I assume that the player base will be composed of people with what in some areas are radically different desires, and in some areas with the same kinds of innate desires. Although I think there are certain high-level similarities of interests, different people find different things to be "fun."
So when it comes to pure game design, what I favor is a system whose result provides the most fun for the most people most of the time. Done in a slapdash kind of way, that would probably produce a big bag of mush, but I think a focused design could lead to a product that feels like a richly diverse but internally consistent world.
OK, so how to define a game's challenge level to help achieve that goal when there are thousands of people with different notions of how much challenge they want?
The usual approach is to implement the old character level design model, then make the challenge level of each bit of content static and defined by location. This allows players to decide the difficulty level of the content they want, but as characters rise in level this makes some places off-limits and some places no longer worth visiting. Most MMORPGs take this approach.
Another way to define the challenge level of content is to dynamically scale it to the player's or group's capability (however defined). This is usually criticized for eliminating the pleasure of defeating enemies that were once too hard to beat. On the other hand, it has the advantages of getting rid of the need for the odd and arbitrary notion of "zones," and of insuring that content that's neither too easy nor too difficult is always easy to find.
I think each of these approaches alone has too many weaknesses. I'm trying to figure out some alternate approach that typically exposes the character to content that's difficult but beatable, while still allowing players to find some things to do that are especially easy or especially risky.
I don't have an answer to this yet, but yes, it does seem to me that too much freedom in this area is bad news for a game. Offer too much content that's too easy, and you get bottomfeeding for cheap XP -- i.e., grinding. Too much content that's too hard, and you get people doing things that lead to them dying over and over and over again -- basically behaving as though foolishly risky behavior was acceptable (a bad lesson to impart, even in a game).
Which (finally!) brings me back to alts, to limits on the number of characters per server per account. The more characters, and the easier it is to treat them as aspects of one ubercharacter, the more that a game's challenge is skewed to the "too easy" side of the ledger. That doesn't mean I'm trying to make these games "too hard"; it means I'm trying to keep them games.
If there's too little challenge, too little risk, it's no longer a game -- it's a software toy. I happen to really enjoy the software toys that Will Wright has created (I can't tell you how much I'm looking forward to Spore), but that's not what we're talking about designing here -- we're talking about games, where there must be some reasonable level of challenge.
So I'd like to see some game (perhaps a Star Trek MMORPG?) designed to insure that most players, most of the time, are exposed to content that's appropriate for their capabilities, but that lets them try really hard stuff every now and then, and lets them take a break every so often with returning to something that's become easy to beat. I'm not sure how to do that, but it does seem to me like a good idea from both entertainment and business perspectives if it can be done, because it maximizes fun for the most people most of the time.
And just to complete this little mini-manifesto, I'm also completely in favor (as I've noted in another thread) of offering plenty of content that's non-confrontational, or even that's purely social, that isn't about win/lose gameplay at all.
For a gameworld, game is important... but so is world.
Getting both of those in balance and keeping them that way appears to be very, very hard to do.
