Groundhog day
Have you ever felt that no matter you do, nothing changes? How many times have the town criers proclaimed victory against some for or another, only to do so again tomorrow? Congratulations, you have figured out that you have been living the groundhog day.
In World of Warcraft, the world is static. Anything you kill will be back sooner or later, even players. Why? Because a dynamic world would need an army of content developers and gamemasters to maintain. Let's deal with PvE first, because it's easier.
The current method allows the content to be re-used. Let's say that we start with a new server, and everyone starts at level 1. Eventually people are going to level up enough to clear the Deadmines and kill Edwin Van Cleef. Now what? Since he's dead, nobody else can kill him. The content designers will have to design something else for players to do. Now imagine that you have a few hundred other servers full of players as well. Even if you had an army of developers and gamemasters inventing new quests, you still wouldn't have enough. You'd always have more customers than staff.
The sad truth is that content is consumed faster than it's created. This is even clearer when dealing with endgame content. Let's look at Naxxramas, for example. It had been under production for months, maybe even years. Only bugs are holding the leading guilds back from completing it. Static PvE content is all about reusability. Players that have just recently started playing World of Warcraft have a huge amount of content for them to explore. Even players who have played WoW from the beginning still can enjoy a nice instance run or a guild raid. And if they start a new character, they have as much content as a newbie at their disposal. If they have played on just one faction, they have almost half of the game's content still unexplored.
PvP content is a bit trickier. A common complaint about PvP in World of Warcraft is that it doesn't change a thing. No matter how overwhelming you are in Warsong Gulch, Arathi Basin or Alterac Valley, the next match is basically the same as previous one. The effect is even more apparent in World PvP, especially town raiding. Any guards you kill will be back in a few minutes. Even faction leaders will respawn in few hours. Even the skeletons disappear in a day.
But what would dynamic World PvP look like? We now have a glimpse of it in the form of Eastern Plaguelands towers. These can be captured, and the faction controlling them receive zonewide buffs. So far, so good. The buff's effect is minor enough not to unbalance the factions. No important NPCs are near the towers, so PvE content is not disrupted. The frequent killing of NPCs and the resulting impact on opposite faction PvE was the reason why dishonorable kills were implemented in the first place.
Those who were playing World of Warcraft at the time the honor system was introduced probably remember the Tarren Mill battles. While Blizzard did design the Hillsbrad Foothills as a hotspot, I doubt they anticipated the effect the introduction of the honor system had. Because Alliance had the majority on many servers, the Forsaken town of Tarren Mill was frequently sieged and all inhabitants killed. Combined with the major latency caused by several hundred combatants, the whole zone was practically unhabitable for Horde. Even if you could reach Tarren Mill before dying, all questgivers, vendors, innkeepers and even flight masters were dead, so there was nothing you could do there.
Due to the latency issues and resulting server instabilities, Blizzard had to intervene. They implemented two new features to discourage these crippling town raids. First, there was the concept of dishonorable kills. If a player killed a low-level important NPC, he would be penalized immediately with loss of honor. Somewhat paradoxically these civilians were also given a new skill: calling for backup. Civilians could yell for help and spawn some guards. At first these guard spawns were just a mere distraction, but then Blizzard changed it that guards would spawn for every player that wandered too close to a civilian. That meant that a 20-man raid group would spawn 20 guards per every civilian they passed. Even with a full level 60 raid, that is a significant hindrance. Pass a few civilians and you suddenly have hundreds of guards upon you. Thus, if you wanted to survive, either you stayed well away from the civilians or killed them very quickly.
The introduction of instanced Battlegrounds solved many problems. They offered lag-free combat among willing participants with no risk of dishonorable kills. The queuing system tried to ensure even teams. However, this PvP heaven came with a cost. Because it was simply faster to gain honor (and thus tangible rewards) with Battlegrounds, World PvP simply withered away on many servers.
In all it's flaws, World PvP had something that Battlegrounds have been unable to achieve: immersion. Defending the free orcs of Hammerfall against the human enslavers is completely different than chasing a piece of cloth in Warsong Gulch. Fighting on the blood-covered streets of Stormwind is different from spending a few hours fighting on an icy lake.
What would dynamic World PvP look like, then? In the early days, Blizzard promised plenty of World PvP, even siege engines. There are ballistas and catapults scattered around the world, but they are simply props. They serve no functional purpose whatsoever. Even if we stay at the level of Warcraft 3 and allow siege engines to destroy only trees, buildings and units, implementing them will be quite difficult. The concept of destroyable buildings is already in the game. The World of Warcraft engine is not capable of true structural damage (like in Red Faction or Silent Storm), so buildings would have hitpoints. Even controlling siege engines and designating firing zones would not be a problem. Mind Control and various devices already allow players to control other units. The details have been already solved by several real-time strategy games (including Blizzard's own).
However, the whole concept of destroyable buildings is troublesome. But what does this have to do with a static world? The sheer scale of World of Warcraft continues to be a problem. Even in it's miniature from, Azeroth is large. If destroyable buildings were implemented, the world servers would not only have to keep track of every mob and player in the world, but also the buildings. It also has reprecussions on client side. A common trick in many 3D engines is to include only those parts of the world in the calculations that the player can see. Because these visibility calculations can be costly, they are usually done beforehand. World of Warcraft also uses such visibility calculations. In addition to terrain, static objects like buildings, bridges and tunnels block visibility. But if those objects can be destroyed, the calculations cannot be done beforehand and thus the game cannot take advantage of them. Imagine how slow Stormwind would be if you could see the whole city from the Trade District.
But enough about mechanics. Let's get to the true killer of a persistent world: Lack of faction balance in numbers, attendance, levels, gear, skill and organization. Eventually, one faction will gain the upper hand. In a persistent world, they will be able to press that advantage, further skewing the balance. And eventually, they will capture or destroy all enemy settlements and defeat the racial leaders. A miniature version of this happens every day in Alterac Valley. So once the racial leaders have been killed and the capitals are in flames, now what?
While the war up to the final victory/defeat might be fun, it is also anticlimatic. And unless you want to run into the same problems in PvP as in PvE (running out of content), the world would have to be reset after every conclusion of the war. Or the ultimate victory/defeat would have to be prevented by a GM. Both distrupt immersion. Which is more immersion-breaking? Moving from one nondescript day to an another, or having a continent in ruins miraclously restored?
More serious problems arise after a few rounds of the war. Due to human nature, we prefer to be on the winning side. So if one faction starts to dominate, they will attract more people. The faction balance will only get worse. The winners will be bored of easy victories, and the losers are frustrated with fighting against impossible odds. They will switch factions, servers or quit altogether. Eventually, there will be nobody to fight. New players will shun the losing faction, because leveling up when the enemy controls 90% the world will be quite difficult. There are already servers where the faction balance is 90-10.
There are of course games which try to have a persistent PvP world. But they have once advantage over World of Warcraft: Multiple factions, or even player-created factions. A three-way struggle for power is more likely to end in a stalemate. Warcraft, on the other hand, has always been about two warring factions, only deviating slightly in Warcraft 3. The only currently possible candidates for the third faction would be either the Goblin Cartels, or Illidan's servants. With the inclusion of Blood Elves in the Horde and the Draenei in the Alliance, Illidan's possible faction lost a lot of the uniqueness, leaving only the Naga as an unique race. And the goblins currently have a very cushy position as a neutral faction, with an income flow from both factions. Open warfare would be quite costly. So it is very doubtful that there will be a third faction anytime soon, and thus no persistent game world for PvP.
