The talk was obviously geared toward game developers, but I found it incredibly interesting to hear a Blizzard executive's opinion on which systems and game features were successes or failures. Additionally, past failures helped the company make improvements in a later game.
Up first, Rob discussed Diablo II's economy, which he labeled as a failure. Gold became useless to the player because they would accumulate more than they could use. Players evaded the death penalty of losing half of his or her gold with a workaround and thus, the end-game economy revolved around bartering high-end items.
Learning from this, Rob labeled World of Warcraft's economy as a success. In it, gold has value and was carefully designed by the developers. The cost to repair items is used as a "tax" on players. Higher quality items incur larger taxes to repair. Additionally, there are always extras (like "exotic mounts") to spend gold on. Finally, the Auction House system continually removes gold from the economy while facilitating player trading.
When speaking about creating depth before accessibility, Rob discussed the idea of hero units in Blizzard's RTS games, labeling the implementation in StarCraft as a failure. In SC, heroes were supposed to be awesome units that lead their massive armies into battle. In reality, hero units were weak and ultimately useless in battles of over 50 units. If the hero died, the mission would be lost. Players would hide their heroes in a base and use other units to win the mission.
Taking the lessons learned here, Warcraft III made several changes. Army sizes were reduced to make the hero more meaningful and the hero was always the most powerful soldier on the field. Finally, if a hero died, it could be resurrected at an altar for some gold. In StarCraft II, Blizzard isn't adding hero units, but will play around with expensive elite units that are lost once they die.
In talking about "concentrated coolness" and the fact that more isn't always better, Pardo discussed World of Warcraft's vehicle system as a failure. The technology to ride vehicles in WoW was built well before its debut in the expansion packs, but the designers went a bit crazy with the tech. Soon, designers had crafted over 100 quests and dungeons that used vehicles, which didn't use any of the abilities of a player's class. It was scrapped and eventually reintroduced in a much more limited fashion in the expansions.
A potential success for "concentrated coolness" comes from StarCraft II and its unit selection. Instead of just bringing every unit back from the original and adding more, the team was forced with a tough decision. If the designers wanted to create a new unit they had to remove an old unit from the game. Using this method, the unit count was kept low enough so that the overall complexity level did not grow too high.
In storytelling, Diablo II quests and the World of Warcraft quest "The Green Hills of Stranglethorn" were failures. In Diablo II, quest NPC's would speak for around 2 minutes giving backstory and details about a quest that just ended up sending the player off to kill a demon. Everybody skipped them. In the WoW quest, players were meant to slowly collect quest items that would drop randomly off of any enemy in a particular zone, but it was too slow. Players just bought the missing items from the auction house.
For success, Pardo identified the "Culling of Stratholme" from Warcraft III and the Death Knight starting area from World of Warcraft's Wrath of the Lich King expansion. Both presented the player with story content during gameplay instead of frontloading the content with tedious briefings. The Death Knight area also served as a playable tutorial for the new class that was introduced in the second WoW expansion.
When game developers have the luxury of time to look back on past mistakes and learn from them, gamers end up with incredibly polished games. Very few studios besides Blizzard have this kind of time, but we will hopefully get more frequent releases out of them with StarCraft II set for a mid-year release and the next WoW expansion, Cataclysm, following later in the year. Presumably, Diablo III will follow in 2011, but this is Blizzard we're talking about--it'll be out "when it's done" and not a moment before.