Jeff "Tigole" Kaplan — World of Warcraft Game Director
Slashdot: Could you walk us through how you balance a particular class or ability — especially going into Wrath as you're adding all these new and extra skills, and trying to find a new balance?
Jeff: Definitely. It's really an ongoing, iterative process. Every piece of content we add in terms of PvE — a new raid boss, new abilities that creatures are doing — and every other class ability are going to come into play in whatever original class ability we were trying to tune in the first place. So, as we add more things, we constantly have to go back and look at the things that were previously fine, but now might suddenly be overpowered or underpowered. We also have a lot of philosophy that comes into play. It's very easy to do what we call — it's kind of a Blizzard 'cardinal rule of never-do-this' — balancing to mediocrity, which means that you always notch everything down because you're scared of certain things feeling overpowered and are literally living by the numbers. I think numbers are a great guideline, and you should always understand the math behind what you're doing, but at the core, you need to follow the gut and ask "Hey, does this feel really great?" The best place classes can get, in our mind, is where everybody thinks everybody's overpowered. That's kind of the Starcraft balance at work — I think it's best illustrated in Starcraft — "Oh my god, all three races are wildly overpowered!" Yet, somehow, the matches seem to come out even most of the time. Something that we also try to communicate to players — it's difficult for them to understand, and it's not really their responsibility to even worry about it — if we never touched the classes — let's say we all agreed that the classes were perfectly balanced, and never touched them, and let them go for three months, they will eventually become unbalanced because of different strategies that evolve. The players are really driving how the game goes, and it's our job to play referee at a certain point. But not referee in the sense of "No, no, no, you're breaking the rules," but just asking, "Hey, are you still having fun?" And we have to make sure that you're having fun, but not at the expense of someone else.
Slashdot: Along those lines, how are you trying to balance for the arena at level 80, with the addition of a new class and all the new skills coming into play?
Jeff: The arena is particularly challenging because of the three brackets; a class that might be absolutely godly in 2v2 doesn't even get invited to 5v5. What we've claimed, and what we've stood by, is that we don't balance the classes in PvP in a 1v1 scenario. We have zero expectation that every class will be able to beat every other class. The key, and the real core to our arena tuning is making sure that tuning the classes in the arena doesn't step on the toes of the classes in other parts of the game. It's very important when we look at a class that we say, "Here are the fun things that this class has to do leveling up," and some level-up abilities have very little use in the arena at all. "Here are some uses in a 5-person encounter, and here are uses in a 25-man." So, the arena is just one part of that. It's very important that the arena doesn't become the sole focus of class balance for the game.
Slashdot: Professions in The Burning Crusade — tailoring, for one — the curve wasn't very smooth for progressing into raids. Are you trying to change that this time so it segues into raids better?
Jeff: We totally agreed with that; we felt like a lot of the trade skills had a very odd curve. Either they were way too easy, and you just sort of had everything, and you were instantly done for the expansion, or it felt like you could never get there. There's a new system in place for Jewelcrafting that I think is really interesting, and is a direction we want to swing more towards, which is a Jewelcrafting daily quest that rewards a currency for Jewelcrafters only. Then, there is a trainer that literally has dozens and dozens of Jewelcrafting recipes. Some of them are what we like to call "selfish" recipes — they only benefit the Jewelcrafter — and for others you're being the "outgoing, good guild member" if you get them. So we're really putting the power in the player's choice as far as "Hey, I want to get my guild in good shape for raiding Naxxramas," or, "No one on the auction house has yet purchased this recipe and put these gems up," or, "I really want this powerful item for myself, personally." So, we feel it will let player choose the direction of their tradeskill development while still having access to all the content over time. By the end of the expansion cycle, they'll probably have it all, it just depends how soon you get it, and how you prioritize. It's us putting that control in the players' hand, rather than us trying to anticipate what's going to matter to you, because there are so many different play styles.
Slashdot: As you go into raids, are you still going to be able to get upgrades through your professions?
Jeff: Yeah. You're going to continue to get upgrades. There's also the concept of the Primal Nether in The Burning Crusade. It used to be Bind on Pickup — we switched it to being trade-able. The equivalent to that is already in Wrath of the Lich King, and it's already trade-able. It can be purchased for the equivalent of Badges of Justice. So we feel like we have things that gate the tiers of raiding, and when we add a new raid tier, we can add another one of those items, similar to the progression from Primal Nether to Nether Vortex, and continue to progress those throughout the raid tiers.
Slashdot: Wintergrasp is a bold new direction in terms of creating world PvP that's something in which a lot of people can and want to participate. What's it like to design something like that and commit so many resources to it before seeing the fans' reaction to it?
Jeff: Well, I think it's our job as the WoW development team to try to anticipate fan reaction; anticipate not only what will be cool about things, and hold that vision, preach it through to the team, and make sure we realize the vision, but also to second guess ourselves the entire way. We have to ask all the questions of "What happens if...?" We need to solve a lot of those problems ahead of time and test it as much as possible. We had a test last Monday on the beta realm where we had 350 people fighting at once, which was a tremendous feat for us, because it was, on the server-side, completely lag-free. Player's clients, with 350 people in the area, are not always going to have systems that can handle it. For some people on lower-end systems, the sheer video lag or lack of processing power is going to slow them down. But, on the server side, as far as the network went, things were great, and we were supporting it. Another key component of doing something like Wintergrasp is playing it ourselves. There's no way to watch over an encounter like that and know, "Is that guy having fun?" Is it fun to be the one guy getting hit by four siege engines at one time? So, we need to play it ourselves and form our own opinions of what is fun, what is balance, what is overpowered. Is it too short? Is the reset time too long? For those sorts of things, you're never going to have enough design intuition to have the answer short of playing your own game. That sort of applies to us throughout World of Warcraft. We're very fortunate that our development team.. we have the problem of saying, "Ok guys, let's try to keep the WoW to lunch time or after hours." We're still in the mode where everybody's playing the game.
Slashdot: Speaking of siege engines, are you happy with where they're at, and where do you want to take them from here?
Jeff: I think the vehicle system has turned out to be one of the best things we added to the game. Originally we added it to do the Wintergrasp vehicles and the Strand of the Ancients vehicles, and to use those against destructible buildings. We wanted the Plaguethrower, and the Demolisher, and all these other things we added. What we ended up doing was developing a very robust vehicle system, which our quest designers, who weren't even working on the PvP content were kind of looking over our shoulder saying, "Well.. let me see what I can do with that." And they've ended up writing some of the coolest quest mechanics you've ever seen using that same system. You're on the back of a guy's horse throwing flaming bombs at Worgen as you go; you're flying a frost wyrm; the Malygos encounter — not to spoil any surprises — it ends with a big finale where the entire raid is on what are the equivalent of vehicles. So, that's turned out to be one of the coolest things. Also, just watching in Wintergrasp.. the sounds of the Plaguethrower are just amazing. Our sound team really nailed it. It just feels super-visceral to have that barrel of plague-muck and wing it across the battlefield, and see the destruction that it leaves. It's just fun, visceral, very video-gamey. It's fun to click the buttons; that was the general idea.
Slashdot: Are there any plans for vehicles to come into the arenas.
Jeff: Not into the arenas. We do have bigger battleground plans on the horizon. We have nothing specific to talk about, but siege definitely comes into play in those new battleground plans. The arena, we feel, is a more pure environment. The more we do gimmicks in the arena, the more the arena players complain. Our long-term goal for PvP is to actually shift the focus off of the arena, and get it back onto the Horde/Alliance conflict in the battlegrounds. We think Wintergrasp is really cool, and public PvP is cool, but when we don't have control over the balance — which we really don't, in a thing like Wintergrasp — we can't guarantee a fun play experience for people. But we feel that in the battlegrounds, it really strikes the right number of people, accessibility, less focus on individuals and more on teams, and definitely more focus on Horde vs. Alliance, and big, epic feel, which is really what Warcraft is all about.
Slashdot: Are there any plans to make WoW friendlier to Linux?
Jeff: Friendlier to Linux.. Currently we don't have any plans to release on Linux. WoW is actually extremely Linux-friendly, internally. There are many Linux WoW servers and WoW clients. But, publicly, we haven't released WoW on Linux, and don't currently have any plans to announce that.
Slashdot: Is there any sort of vehicle that might allow the people who have gone through the trials and tribulations of getting it stable on Linux to share their experience?
Jeff: Possibly. It's definitely not out of the realm of probability. But, at this time, we don't have any plans to announce it on any other clients than we currently have.
Slashdot: Are there any plans implement some kind of a spectator mode?
Jeff: Yes. We would love to implement spectator mode. We've had some really great ideas about it. Not only to do it for the arena, but we've had some really good suggestions about doing spectator modes for the 10 and 25-person raiding as well. There's been a desire for people to watch the top end guilds. It's definitely on our list. I can't say that it's coming out any time soon; it's quite a bit of development. Priority-wise, there's not a lot of new gameplay there. But, it is something that we'd like to do. We'd also like to get to replays, too. In a lot of ways, I feel like, for the arena, replay would serve people better than actual spectator mode. In spectator mode, you have to know to be watching the match when it happens, whereas a replay mode would allow, "Oh my god, this turned into the match of the century, you've got to see it." Those are definitely on our list. We think they're super-cool ideas, and it's just a matter of finding the right time in our production schedule to get to it.
Leonard Boyarsky — Lead World Designer, Diablo 3
Leonard was joined by another Blizzard representative.
Slashdot: How do you define the limits of the world and the individual levels?
Leonard: Well, the limits for designing the world and the limits for designing the levels are completely different things. For the world, we've done a lot of design even in areas that you won't be seeing in this game. We really wanted it to feel like a living world. One of the things they really started to scratch the surface on in Diablo 2 was expanding it and talking about some of that stuff. But you really didn't get all of the history. You got the feel for what these places were at the time you were there, but you didn't get a lot of depth. There's never too much you can do. There's too much we can try to shove on the player, but there's never too much for us to do. If we have that intense, deep knowledge, we can just drop tidbits here and there; intriguing things that will hopefully get people to research it further. As far as the levels, it really comes down to play time and what kind of feel we want for the dungeon. The perfect example for that would be the demo. There will probably be no levels as short as the one in the demo, except maybe the very first dungeon you go into. But, we crafted that level specifically for Blizzcon, because we knew people only had 15 minutes. So, that's a perfect example of the way it works. You look at it and you ask, "Ok, what kind of chunk of gameplay do we think this is going to be?" And then you make it how big you think it should be for that gameplay, play through it, fix it because you're generally wrong, and then just keep iterating until you get it. It's an iterative and complete process. It's not like you can look at one level separate from all the other levels. You have to look where it fits in the game.
Slashdot: Is it going to be the same system where there are Acts, and then the Acts are subdivided into different sections?
Leonard: Yes. We're probably going to do a little bit of tweaking. It's not going to be exactly the same as it was in Diablo 2, but we are sticking with the Act structure. We were trying to move away from it at first, but it just kept coming up, so it was obvious that the game needed to be structured that way.
Slashdot: Have you given any thought to having a less linear type of gameplay, where you have a world - similar to the things they're doing to Starcraft 2, where it lets players make more of a choice — or are you going to stick with the linear style for now?
Blizzard: The advantage we have with Diablo is that we can actually tell a story. You have a huge impact on the world with what you're doing.
Leonard: In the previous Diablos, you would get the quests in a linearly doled-out manner. Our main story arc is going to be linearly doled-out, but there are a lot of side quests. There are a lot of random quests that you may or may not get per game. Those are very non-linear in that you can do them earlier or later, depending on when they come up. So there's that form of non-linearity, and there's probably going to be ... other stuff we'll talk about at a later date that'll enhance that as well.
Blizzard: It's like a mix of the two. You have the backbone story structure and then you can decide whether to take some of the quests, and also the game decides whether you can see some of these things, because they're random each time you play through.
Slashdot: By that, are you referring to the scripted events that they mentioned yesterday?
Leonard: Yeah. A lot of those are random. Some of them have to do with the main story arc. Did you play the demo? When you run in and see those guys talking about the crown, that's supposed to be giving you a hint about what you're doing. You're basically summoning his spirit into the physical world so you can fight him. So, that's a way for us to give you guys a little more information. That one would always happen, because you always fight the Skeleton King. Other ones, the escort quest, for example, that's a completely random one. There's another one in the dungeon about bringing this mysterious box and putting it on an altar. There will be random quests in the world which work that way, probably some in the dungeons. So, even with letting you do quests in a different order, there's going to be a ton of varying content.
Slashdot: How does the new checkpoint system work?
Leonard: When you the end of a dungeon level, before you go to the next one, you hit a checkpoint. That's so, number one, you don't get teleported back to town if you die and say "Okay I have to run allllll the way back to the dungeon." The reason that we didn't put it at the beginning of the next level is because half the time you're saying "Oh my god, I'm going to die," and you're running back to the entrance to get away. So, you have this mob of monsters following you to the entrance, and you'd just spawn and die, spawn and die. It's a very simple system; it's just a way of keeping you from having to run from town to wherever you were every time you die.
Blizzard: It's nice for the users because you don't have to work about saving and crawling; it just works. You don't ever worry about saving your character.
Slashdot: Are there any plans to release some sort of map editor?
Leonard: No. The way we put together our maps is very art intensive and artist intensive. We talked about it a lot at the very beginning. First of all, you have the random dungeons, which are very technical. Then you have our out-of-doors, which gets put together in a very specific way. So, there's really not the raw materials for people to make their own stuff. It would take a lot of work for us to build our editor so that it was usable by those on the outside. It just didn't seem to have enough bang for the buck.
Blizzard: One of our key goals for Diablo 3, going back to the world layout and design, was to bring the world of Sanctuary into more focus and have more specifics. This is why Leonard and Chris Metzen are looking at having more back story for the classes you can pick. They're not just faceless archetypes, they have this whole back story. The regions you visit are more defined, and the whole world is becoming more and more real. We're just putting a lot more subtle and outwardly noticeable things in the world design to tell a story. You can come to a city that has this great past, and you can see it in the actual level design.
Slashdot: Along those lines, will we be seeing a similar amount of cinematics as were in Diablo 2?
Leonard: I don't know if there will be more. In Diablo 2, it was almost like a separate story from what was going on in the game. The story we're telling through our cinematics is directly related to what you're going through in the game. So, it is very much going to have an emotional impact on what you're seeing and doing. And we're doing scripted events in-game — you've seen some of the small ones, and there could be spots where we do bigger ones. But, once again, we're really keeping in mind the fact that it is an action game. We're balancing the action and the RPG. We don't want people who want a hardcore action game to feel like they're being bogged down with story, but we want it there for people who want to dig deeper.
Blizzard: It's mostly opt-in. You can decide to stay and listen to that little scripted event, or you can just go on and start cracking more skulls.
Slashdot: As far as the single-player campaign goes, will there be changes in balance as the difficulty scales up, so there are fewer instances of players running into a proverbial brick wall against certain monsters with extreme immunities and resists?
Leonard: You mean between the Normal, Nightmare, and Hell difficulties? Yeah. You know.. all our sympathy for the player starts being removed brick by brick as you get into those levels. (laughs) In the main game, we say, "We can't do that because players need to be able to get by this," and we don't want them to have to go back and build up their character even further just to handle this one guy. When you get into Nightmare and Hell, we figure that's what you're asking for.
Blizzard: Not that we just make it harder for harder's sake.
Leonard: Yeah, we don't screw you on purpose.
Blizzard: We require you to dig deeper into your kit of abilities to really pull things off. We require you to be smarter, use more tactics, and really dig into what you're able to do to handle some of the challenges.
Slashdot: From what it looks like, there's more of a toolset, or more of an incentive to build up a larger toolset.
Blizzard: Yeah. Say there's a boss that could be in a side-quest or the main quest that has a resistance to frost, and frost is what you've been focusing on. Well, you can go back and learn some new skills — fire skills, or something else — and return to tackle that guy. Those kind of specifics will be more prominent in the higher difficulty levels.
Slashdot: In the demo and the trailer, we saw parts of the environment that seemed to be destructible. Is that something you evolved from busting barrels in Diablo 2?
Leonard: Yes. We started off saying, "You know, it'd be really cool to be able to destroy a wall." At first it didn't do any damage, but we thought, "If you're destroying a wall and it falls down, it should damage whatever's there." So then we put the damage into it. And then everyone said, "That's the most fun in the whole game," so we decided, "Hey, we need to do more of that." So now, there are chandeliers. I didn't see anyone using chandeliers in the demo. If you click on the thing on the wall that holds them up, they'll drop. That's one of the fun things we do in co-op, because it stuns players. We'll wait until our friends get underneath and click it to stun them. (laughs) We are exploring what we can do with destructible environments a lot, because it's a crowd-pleaser.
Blizzard: Our engineering group is amazing. The engine they've created — there's these technical aspects with the random dungeons as well as all the things they've advanced going from Diablo 2 to Diablo 3; the 3-D characters, all the effects, real-time physics, destructible environments. They're just pouring everything they can into making that engine sing and have all these great features.
Slashdot: In one of the panels, they mentioned their goal of making the character animations very visceral and having a big impact. What is your direction for the backgrounds — the levels themselves? What's your inspiration for building them, what are you trying to do with them, and what are you trying to make them look like?
Leonard: We want them to look fantastic. We want them to be very impressive. We look at a lot of real-world stuff, starting with a lot of real-world cultures, and then try to extrapolate from there. We try for the balance between believable — that it can actually be constructed — and fantastic. You don't want it mundane. You don't want it to look like you could go to Europe and see a destroyed 13th century castle and have it look the same. But, at the other end, we're being conscientious to not make it look like a full-on fantasy world. I think one of the reasons Diablo has worked so well in the past is because it feels very grounded in reality. That's our biggest balancing act, right there; trying to get both of those together.
Blizzard: It's kind of different; this is a world of dark fantasy. There are no elves or dwarves or that sort of thing. It has this off-kilter view that's foreboding, and we're trying to have that seep into and drip from all the environments we're creating.
Slashdot: Can you tell us about the multiplayer aspect of the game, in terms of Battle.net and the possibility of LAN play?
Blizzard: We're not supporting LAN play. We're basically focusing on making the best multiplayer experience we can, and that's all through Battle.net. There are tons of features we're going to be supporting both for cooperative play and competitive play. One of the things we can talk about with the new Battle.net is security. Fixing some of the problems we had with the earlier Diablos — item duping, cheating, and griefing — we're going to be addressing all of those things with the new Battle.net, as well as some pretty awesome competitive play ideas we're working with right now. So that's going to be the biggest advance, especially for previous Diablo players, to see all these we're planning. It going to be really awesome.
Chris Sigaty — Lead Producer, Starcraft 2
Slashdot: What is your approach when you look at balancing, not only player versus campaign, but player versus player?
Chris: Really, it all just happens through lots of play. We definitely are looking at numbers, and there are tons of numbers plugged into the game, and a lot of different dials that we can manipulate to try and hit balance. But, a lot of it comes through play, and often times we find things in beta or even after ship. We have patched Starcraft many, many times; we have patched Warcraft III many, many times. We definitely ship out the gates with a good balance, but it evolves. There are things that these really great players come along and figure out that we just aren't able to figure out. These are things that we don't necessarily anticipate, even in beta, so for us I would say the biggest part is lots of play. For example, Starcraft 2 right now is in alpha internally. Our whole company — we have over 3,000 people now, and many of them are playing when they have time between their jobs to give us feedback. The other big balance tool we have is a couple of pro-gamers, who basically spend all day just playing each other, and they try to develop strategies against each other. One of the unique challenges that we face in doing this, because the original was such a huge phenomenon from the e-sport perspective, is balancing those numbers really well, yet making things for the more average player, or new player, that feel really powerful. What happens is, in the hands of an expert, everything is powerful to some degree, and so then you end up balancing the numbers down to mediocrity. One thing we just did to try to combat this is that we just made the Zerg 30% faster on Creep. Many people were like, "You can't do that, that's gonna break it," but I wanted to try it and see what happens. We are trying to work in some things like this, so we would add an ability, balance it out and then explore some bigger things, like what does this mean in the hands of a pro. Those are some of the tools that we use to approach balance.
Slashdot: What types of caps are you looking at?
Chris: We are sticking with the hard population cap. We have had them in all of our games; it's 200 food again, and we will be sticking with that unless something changes significantly. Right now, we're definitely balancing the game in that direction. There isn't anything we have seen that has made us desire to do otherwise. Where we had a bigger reaction was with Warcraft III; we put in a much smaller cap — 90 — and a lot of people felt that was pretty heavy handed. It was a very different game, though, and a very different time. Graphics engines, too, were very different. So, part of it was system requirements, but now that it has opened up, the main reason is that going beyond 200 doesn't seem to mean anything for us. 200 seems like a good, hard cap, and even though hardware has come a long ways, we still see performance issues, especially in some of the 2v2 matches. If all four players have the 200 food cap, and they are running those units against each other, other games may solve that problem by choosing a performance solution. From a visual fidelity standpoint, we wanted to keep the same unit that you have at the beginning of the game that looks a particular way to look the same way when you have 200 units. So, it's partially that, and again, it's partially just management — once you get to 200, what are you really doing at that point? You want to force the guys to actually have action at some point, and 200 is really out there. We think far enough out there that it works.
Slashdot: How constrained are you going to keep players with respect to the UI? Some players have specific questions about the zoom level, and is there any possibility of giving a broader view of the battlefield?
Chris: It's interesting that you bring up UI. UI has been a point of contention since the beginning, for Starcraft 2. There were the biggest arguments about things you would never believe; unlimited selection is a great example of that. People thought unlimited selection would break it, arguing that you have to control groups of twelve units at a time, and that is the strategy and the way you have to play. Each one of those has been a big conversation. As far as zoom level specifically, we made a decision to keep the action pretty close on purpose. We are actually allowing players to see a little more with wider monitors, which is a big decision, but we kept it constrained, really, because that's how we feel is the best way to play the game. We are holding the players to some standards. The user interface that we have isn't going to be some sort of scripted user interface that you can update at will, like World of Warcraft does. It's going to be set, and this is how we envision the game being played. That said, you can go ahead and create all sorts of different visions for what gameplay should be by using the map editor and changing things up.
Slashdot: What are your goals for system requirements? Is there enough attention being paid to both the low and the high end?
Chris: We have definitely done a lot of cutting-edge things on our end, but we are about to (post-Blizzcon) run back through and make sure we are hitting the low end. We have generally done a good job at Blizzard about trying to make sure we support that lower end. I would even say that we have spent a little too much time making sure we have all the bells and whistles and crazy things, like our new cinematics, so now we are going back to re-balance. I'm not sure we are discussing any specifics yet, but we are going to be investigating those questions as soon as we get back from Blizzcon.
Slashdot: What capabilities do you foresee for the map creator, and is there any chance that the map creator could be released early, in a manner similar to Spore?
Chris: There is a possibility; we have actually been talking about releasing our editor in the beta. There are some concerns that if we put it out there for multi-player, we feel like a lot of players will just respond by saying, "Here is how you should balance it," and send back out re-balanced maps. We really don't want to go down that slippery slope and have players just joining in to those types of maps. We want to balance the game we're trying to make. At the same time, we do want to enable people that are doing more extreme games — things like the DOTAs out there that are re-envisioning games. So we are talking about it, and we may or may not include the map editor in the beta, or potentially roll it out halfway through. As far as scope of the editor and how big, I can tell you it's much more powerful than what we had in Warcraft III. We are trying to ensure the things that were done in Warcraft III are still possible. On top of that, we have this actor system now that allows players to do a lot more as far as putting together a series of events to make up abilities. For example, "Make this missile fly using this particular piece of art, play this explosion, make this sound go off." This way, people can now encompass abilities that have their own customized versions of that ability. So, beyond that, all the stuff we are doing with in-game cinematics are possible using the editor. DOTA is awesome. We love phenomenons like that. We want to make sure things like that and beyond are possible.
Slashdot: With Ghost suspended, do you ever see Starcraft evolving beyond a strict RTS setting?
Chris: We don't have any specific plans like that, but I definitely feel that the universe itself is very well loved within the company and other places. I personally think that it would be a fantastic universe to do any of several different genres. An MMO, an FPS, or anything in between. It's always something that is in discussion — "Where is this going to take place?" Always high on the list somewhere is the Starcraft universe.
Slashdot: With the recent controversy surrounding DRM, what is your take on DRM with respect to both Starcraft 2 and Battle.net?
Chris: We don't have specific plans ironed out. We are definitely aware of things that have happened with Spore, and some of the other games that have come out with big uprisings. We want to make sure that we are protected, but at the same time we want to make sure that it feels like the "right way" and not the "wrong way." Our biggest advantage is Battle.net, so I think the solution is that Battle.net is the premiere place to play, and that's where you want to be playing. So, that alone is the best sort of solution. You want to be up there and in contact with your friends, see what's going on, so there is your copy protection, essentially. As far as specifics, we haven't really worked that out. There are definitely some things to discuss still. One of the main things we are talking about is that there has to be a way for people to play offline — on the plane or wherever — but those are discussions that we still have to have.
Slashdot: How would you describe your AI in Starcraft 2, and how does it stack up against earlier AIs?
Chris: Our AI is a very script-driven, home-rolled system. We're going to expose all of that this time, and let people see that. Every time we change significant balance, we have to go back and look at what is going on. We have a pretty strong AI in there right now on hard, and it is just the beginnings of what we want to have in there at ship. There are different tactics and different paths that they can go down, so that you'll be surprised, but we're going to be rewriting all of that again before beta. It's not a learning AI, so we have to do a lot of hand tuning. We have certain algorithms in place that will attempt to analyze data and see what the player is doing. One thing that we are doing this time around that is totally new — our AI cheated in the past. Starcraft and Warcraft III saw the whole map, so the AI could see what the player was building or doing at any given time. The AI in Starcraft 2 now has to scout, and it's much harder to do, but there is a pretty effective AI in there for now.
Paul Sams — COO
Slashdot: What types of challenges do you face on a daily basis?
Paul: I would say that we have such a high level of confidence in our developers that I don't spend a ton of time worry about them, because I feel like they always are going to deliver great content, and they are always going to make sure that they are meeting the quality proposition. We have such an experienced team in that area that we kind of feel that they "got it." What's really new for us — and I guess it's not so new, because we have been doing it for four years now — is the customer service piece of the puzzle. It's really challenging. When you are dealing with 10.9 million subscribers globally and you are trying to provide customer service to them in a way that is consistent with their expectations, it's challenging. You never know what is going to be happening the next day or what challenges there may be with patches, griefing, bugs, or something. You never really know, so the customer service piece is probably the most challenging. Certainly in the earlier days, managing the network was very difficult. We have worked out many of the different challenges that we encountered. We periodically will have a challenge here and there, but for the most part the really nightmarish, difficult times are behind us, thankfully.
Slashdot: What types of measures do you use to ensure quality control?
Paul: One of the things that we have been doing for a while is surveys immediately after you get a ticket closed. It's a short survey that helps us to track the quality of our reps' performance. We also track how quickly they are able to resolve the issues and how long the customer had to wait. We are continually trying to evolve and improve that. I think that our customer service is pretty good. I would say that it is amongst the better customer service solutions in our industry, but it's not someplace where I am comfortable or "happy." We are actually putting forth a ton of focus right now — it's actually one of the big initiatives in our company and our leadership, to try to take our customer service experience from what I think is one of the best in our industry to being amongst the best of class regardless of industry. This is something that people are going to see over the next year. We are putting a huge amount of investment (time, energy, and money) towards really evolving and improving our customer service. It's something that our customers are telling us that they think needs to be better, and we are hearing them. I would envision some significant improvements in the coming months.
Slashdot: Given your position of market dominance, have you ever considered using this position to effect wide, sweeping change? Specifically, and most important to our readers, is with respect to Linux. I know you aren't about to jump in with both feet and support Linux outright, but I know there are people within Blizzard that have clients and tools running on Linux. Is there any sort of vehicle that could be put in place to allow that experience to be communicated to the outside world?
Paul: I know that there are a lot of people at our organization that are big fans of Linux, and I would say that it is something that we have looked at and have an openness to. But, we have built this system, and it is a system that we have gotten to be stable right now, so I wouldn't anticipate that we would do anything to make meaningful changes for right now. For future products it is certainly possible, but for us, there is certainly a comfort level in the way that we are doing it. All of our staff is very comfortable with the way that we are operating and the systems that we are using. So, when you go and shift your focus to something else, you don't necessarily know how to do it as well. I think that whenever you do something like that you have to do so with a lot of caution, so that it doesn't negatively impact your consumers. That being said, I would say that it is possible, but it's not something we are working on with any amount of focus right now.
Slashdot: Given that Blizzard has such a wonderfully rich backlog of games, has there any been any thought to open sourcing some of the older products?
Paul: I don't think we have ever talked about open sourcing our games. We own a few older games, like Lost Vikings, Rock and Roll Racing, and Black Thorn. Those were originally Interplay published products, and they owned those franchises. A number of years ago, we went to them, because those games hold a special place in our heart. So, we wanted to get those back, if for no other reason just because we didn't want anyone to do something different with those franchises and cause them to not have such a fond place in people's hearts. So we acquired those a few years back, which was good timing, since Interplay had a desire to have some revenue coming in, and we released GBA titles of each of those, and that was a fun thing for us to be able to make those available to a newer audience. My kids were both into Gameboy stuff, so they really dug that. That is something we have done with older titles, but open source isn't something that we have ever really talked about, and I don't know what the thought process would be. Certainly, if it is something of interest by your readers, they should let us know that officially so we could talk about it.
Slashdot: What are some of the details that you can give us about the evolution of Battle.net and some of the thought processes behind the decisions being made?
Paul: The things that we have shared thus far as it relates to the new version of Battle.net is, number one, it will ship together with Starcraft 2. There is going to be a real focus on e-sports, and there will also be some strong social networking elements to it as well. As far as it relates to what the business model is — we really haven't gotten there yet. I know that there is a lot of speculation, but I think that part of the speculation is simply because Blizzard isn't answering the question. The simple fact that Blizzard is not answering the question of whether Battle.net is for-pay or not suggests to people that it could have a pay component, and people should not assume that. The reason that we aren't talking about it is because we want to focus on the game, and we want to focus on the game network, and if you look at Blizzard's history, that is how we always do it. We almost never talk about how we are going to distribute it, what the price is going to be — we never talk about that stuff until very, very late in the game. There are multiple reasons for this. One, when you start talking about stuff like those things earlier, you start making decisions and potentially compromise on what the gameplay experience is. We never want to do that. The other thing is that even though we are a very game-first focused type of company, we are still mindful that we have competitors. Those competitors want to know what Blizzard is going to do, and how they are going to charge, and to be candid, I'm not interested in letting anyone else know how we are going to do it until it is too late for them to react. I think a lot of other companies, the way they run their business is they sit down at a table much like we are sitting at right now, they've got marketing and finance people sitting around, and they may ... maybe ... have a developer sitting at the table. They get out reports, and they look at what type of products are selling well, and then they notify the game development team, "Oh, by the way, you are going to be making a game about sailing." Well, no one is necessarily enthusiastic or excited about that. We do it differently. We actually go to the game teams as they are coming off of a game and ask them what is it they would like to play next, what they would like to work on next, and let them decide. Then we tell them to make the game and we decide what type of business model to wrap around it when it's time, but let's not have the business model dictate how good the game is. I think this is something that a lot publishers and developers make the mistake of doing, because they spend so much time thinking about how to get your money and they don't spend the time thinking about the game. We try to build a process such that we focus on the game, and our feeling is that if we make a great game that you and I want to play, that we'll vote with our pocketbooks.
Slashdot: Has Blizzard given any thought to consoles, both for existing and future games?
Paul: This is going to sound strangely similar to my last answer, and here is why. When we are making these games, we make the game for the platform that makes the most sense for the gameplay we're looking at. It's not that we don't like consoles. It's not that we aren't going to be on consoles, because we started on consoles. The games that we developed back with the Lost Vikings, Blackthorne, and Rock and Roll Racing — those, to us, felt better and more appropriate for consoles. Some would argue that it's because PC wasn't as big back then, but that was where those games naturally fit. Games that we have been making since that time have felt better and made more sense to be on PC. But, if our development teams came to us and said, "Listen, we want to make and play the following game next," and if that made sense to be on a console, then we wouldn't hesitate to do that. We're not afraid of consoles and we're not against consoles, it's just an issue of making the right decisions for the game and for the customers.
Slashdot: There has obviously been a lot of DRM fallout recently. What, specifically, is your mindset with respect to DRM?
Paul: Obviously, we want to protect our games. We put a lot of time and energy into building these things, and we feel like we prioritize the right things and make the right sacrifices to create a great game for the gamers. As a result of that, when we have done all of those things, and we think we have done them right, then we do have a desire to see the fruits of those labors. Just like other companies, we are looking at how we can protect those products and how we can ensure that the people who are playing our games have paid for them. How, exactly, are we going to approach that? Certainly Battle.net is a piece of that, and there may be other components there together with that when we launch these future games. World of Warcraft kind of has it built in — you can't play it without an authentic copy, but the other products aren't as easy to protect, so we are trying to come up with creative and clever ways to be able to do that, but it's challenging.