Like all blogs, particularly about computer game development, this post contains a lot of conjecture. So, with that in mind...
What is interesting about Warhammer Online (WAR) is how many of its core things it missed out doing well when it launched. A lot of the stuff that had been promised - the "WAAAAGH is everywhere!", the intensive RvR experience, the taking of learnings from Dark Age of Camelot (DAOC) and applying them to a new game - fell very short. War was hard to find in RvR zones because players were too spread out and in-game incentives promoted other types of play. RvR was so undermined by server crashes that it 1) because a griefing tactic, and 2) Mythic had to implement population limits around fortresses. So war is everywhere, but don't all come at once. And there are other numerous problems that WAR has experienced that I consider above and beyond acceptable for launch from an established MMO development studio.
So what happened? I've got no inside sources, no exciting "the lead dev stole medication from employees' desks" -style stories about why WAR failed to perform. But looking at WAR I see plenty of evidence of second systems effect. Second systems effect is the well recognised human behaviour of over-designing a second project to include all the things that weren't included in the first project. This leads to huge development times, large and complex products that are more error prone thanks to that complexity, and generally a product that promises to exceed the original in all ways but actually under-delivers. Sound familiar?
Part of this second system effect has to come from the gap between DAOC and WAR... and that gap's name was Imperator. Imperator was meant to be mainly PvE with very limited PvP and it certainly seems from the state of WAR that this was carried over as a design reality (if not intent). It's unlikely that there ever be a full accounting of what design decisions and systems for Imperator were carried over into WAR, but its unlikely that everything was thrown out when Imperator was cancelled in favour of WAR.
Imperator: You're going to be waiting there a while, Legionnaire.
The next evidence of second system effect is much easier to see - the race / class / realm design. Everything was bigger in WAR. The scale of the conflict increased with three pairings (Greenskin vs. Dwarf, Order vs. Destruction, Dark Elf vs. High Elf) rather than just three regions (Albion, Hybernia and Midgard). The original design had every race with a massive capital city (six in total) even if only two were available at launch.The size of the overall landmass was bigger as a result. The number of unique abilities for each of the already large number of classes increased. WAR gave players a lot more options, but this had the accidental effect of spreading the player base further and making WAR seem empty while also being harder to find specific classes to play with.
Mythic also added in a number of extra advancement measures - DAOC had an XP bar and Realm Points; WAR got an XP bar, Realm Ranks, a PvE Influence bar for each chapter, an RvR influce bar for each Tier, a stack of separate measures in the Tome of Knowledge et al. That's a lot of bars to track, or to grind out for the rewards, with it being arguable exactly how much extra fun they bring to the game.
And so on. So much seems to have been squashed together in PvE, scenarios and PvP that the good ideas - Public Quests, for instance - are overridden by features and systems that impede gameplay. In other areas, basic systems - like text chat - were completely neglected and possess only the bare minimum functionality.
It is good to be ambitious - ambition takes us places we might not have gone before. But ambition needs to be tempered with a slight bit of common sense. Mythic tried to make a title that did everything following the years of experience they'd had, when a clearer focus on one system. RvR, which they hyped as the "major focus" of WAR, was buried under a raft of other systems that all aimed to be best of breed (to out-WoW WoW even) but all ended up tripping over each other and delivering an unsatisfying play experience.
It's interesting that few MMO studios or developers release a second title that is better than their first. EverQuest 2 tried to completely redesign EverQuest and took a very long time to gain traction. Raph Koster designed Star Wars: Galaxies to offer more than Ultima Online but ended up with numerous flawed systems in a title that lacked the heart of the Star Wars IP. NetDevil's fan favourite Jumpgate was followed by massive flop Auto Assault. Second systems effect accounts for some of these failings as developers attempt to outdo what they got mostly right the first time, only to fail to reinvent the wheel.
Time will tell if WAR can overcome its shortcomings and transform into something greater than it currently is. However, it certainly isn't the title it was promised to be from launch and a lot of those shortcomings arise from overdesigning WAR to have everything bar the kitchen sink.