Perma-death is the idea that just won’t die. Lum came up with good alternatives. Ubiq intimidated me into quashing my third, fourth, and seventh stage arguments and ideas for permadeath. My wife (on prior occasions) has given me the simple “It’s a stupid idea and I won’t play games that implement it” sniff-test answer.

That doesn’t mean that I can’t bloviate about ideas I had around the idea of permadeath.

Reputation
First, I reconsidered the whole whuffie thing again, this time from the angle of reputation, rather than currency. If someone is going to lead you to death, you at least want to know that people think highly of them first. To that end, every active player gets a single vote for every other player, thumbs up or thumbs down. An active player is (arbitrarily) someone who’s logged in for more than an hour in the past 30 days. Players are independent of avatars or characters, and ideally, accounts.

Nothing’s going to stop people from gaming this (“I’ll thumbs-up you if you’ll thumbs-up me”) although I would hope, even expect, this to be a self-policing mechanism. Changes in reputation are not immediately available to observers (there will be some kind of random lag before the reputation is applied). Third-party reputation votes are to be completely anonymous from the player’s perspective: you can only see how you voted for that player, and what the player’s overall reputation is (i.e., a number). Reputation votes can be changed at any time.

Examining a player character will reveal the player’s reputation. Title changes (honorable, dishonorable, esteemed, dastardly, etc.) will provide general categories of reputation visible at any time. Ideally, title will be based on reputation divided by some factor based on active server population: low population servers may be easier places in which to gain good reputation than very large, anonymous servers. As reputation is bound to the player, and not to the character, players will be discouraged from using mule or expendable characters to defraud or injure other players.

Yes, this involves a lot of extra data that the server needs to keep track of, but I believe social systems are just as, or even more important, than game systems and world content.

Minimal Magic
Magic should be minimized. You want to play an assassin? Great! Go put on dark clothes, and practice hiding behind terrain, because there’s no such thing as invisibility. Fireballs, lightning bolts, and heal spells being thrown about with wild abandon lead to one of three conclusions: melee is minimized in its impact, magic damage is minimized to make melee a reasonable choice, or magic is artificially limited by level/power to like-levelled opponents.

Magic should be just that: magical. Hard to find, hard to master, and demoralizing when faced, magic in MMOs should be more than artillery. If players want to become a master of magic, they should first become a hero.

Heroics
What makes one character, or player, different and distinguished from another? In most games, it’s the ability to cat-ass their way to victory: put in the time, reap the rewards. I think reputation is part of the answer–making friends, influencing people, and building relationships will keep players in the game. Another aspect of the answer is to measures the risks the player takes with his character (permadeath, obviously, being the most risky endeavour).

Heroes have done something that most people won’t, or can’t, do:

A person normally becomes a hero by performing an extraordinary and praiseworthy deed. Traditional deeds are slaying of monsters and saving people from certain death. A hero normally fulfills the definitions of what is considered good and noble in the originating culture. However, in literature, particularly in tragedy, the hero may also have serious flaws which lead to a downfall, e.g. Hamlet. – Wikipedia: Hero

In game terms, a hero is a character that has achieved something normal players will never do. They have risked more, gained greater than average rewards for it, and have developed a reputation or mystique. To support this as a mechanic in-game is the goal of heroics. The opportunity to become a hero must always exist, but the risk must be so intimidating that few people are willing to attempt it.

Absent permadeath, the risks become fewer, or even nearly negligible. In EverQuest, the worst thing that could happen would be to lose your equipment: hours were spent in epic planar corpse recoveries. In response, and to minimize risk, the EQ designers gave players utilities (corpse recovery spells) and changed zones so that corpses would appear in “graveyards” where the players could recover their equipment. Still, any kind of corpse recovery serves to keep players out of the game as a kind of penalty, something my wife is highly opposed to. (I believe her ideal would be self-ressing at the location where she died, after the mobs wander off to a respectful, non-aggro distance.)

Mobs stealing your equipment after defeat was explored in Asheron’s Call, and has been pretty much discounted (or ignored) in current MMO design–players started carrying grave goods to bait the mobs into taking the expensive jewelry instead of the valuable equipment.

Limiting access to heroic areas (by time, total player count, total accesses, or another mechanism) isn’t “fun”–that is, it’s an abritrary limit imposed to prevent a player from exploring or achieving. Heroic areas or encounters therefore shouldn’t present limited access, but should rather be avoided by reputation or likelihood of failure.

The real question, then, is how to make failure feel like failure, without imposing either unduly harsh penalties, or making the experience un-fun, while also not trivializing the effect of failure (which is also not fun). When failure and it’s consequences can be defined, then heroic encounters can be described in more detail.

Imagine a world with permadeath
In a world with permadeath:

• People need to group, if only to minimize the chances of death
• Reputation tells whether the person is worth grouping with
• Lifekeeping, first aid, and (if implemented) resurrection become highly valuable in and of themselves, and not just to minimize down-time
• Players want a faster level treadmill if characters are more likely to die
• Players want some way of preserving some gains from character to character, even if it’s only an inheritence or better starter gear
• Heroic activities are those where the character succeeds against the impossible odds
• Players are more likely to tell themselves stories about the risks they’ve taken
• Most players are less likely to take great risks, and will instead stick to safe but steady activities (making it easier to identify the foolhardy and heroes)
• Death has meaning, so that a player that sacrifices his character to save others is renowned–and players who run from battle to leave their companions to die are known for their self-serving nature
• Encounters should be of both a non-leathal and lethal nature, with appropriate rewards for each. In non-lethal combat, characters who fail are knocked unconscious and robbed or otherwised penalized for failure. In lethal combat, characters who fail are killed, unless first aid or lifekeeping is applied by another player character.

The armchair designer in me believes permadeath could develop as an engaging way to play the game. I definitely don’t believe there is room for both PKing and permadeath in the same game, however, if only because people grief.