|
If there were one thing I could say to Blizzard about my Starcraft 2 Beta experience it would be: “stop trying to control everything, and just let go. “
Like most people, I don’t know anything. Some people come to grips with their state of ignorance and others don’t. My stance is that it is better to know that you know nothing rather than believe you know something when in fact you don’t. At face value, this sort of declaration comes off as purely philosophical. However, two areas that interest me, biological evolution and economics, illustrate that there is a practical significance for knowing that you don’t know. Much can be accomplished without comprehension, and much that we are inclined to attribute to intelligence and understanding occurs without either. I don’t claim to be an expert in either biology or economics, but expertise is not a prerequisite for coming to appreciate a critical intuition: incredible outcomes can occur without the slightest comprehension of how they came about. I’d like to talk a little bit about both these subjects and some insights they may offer to the subject of Starcraft 2.
Biological Evolution: Complexity through Natural Selection
Evolution posits that we can develop very complex and apparently intelligent things without any sort of centralized design or authority. Daniel Dennett calls Darwin’s theory a “strange inversion of reasoning.” This is an incredibly apt expression. A human growing up in the modern American metropolis is surrounded by a man-made environment. Everything he sees is engineered by man for man. Man is exposed to one complexity after another, and all of them have an intelligent designer as their common cause, namely, man.
Intuition wants to extend this same explanation to other complex entities. The famous question is: If you found a watch lying on the ground, would you not assume the existence of a watchmaker? Evolution gives us an alternative explanation. Complexity can arise from blind trial and error. From a multiplicity of simple interactions, something surprisingly complex may result.
Market Interaction: The Invisible Hand
Economics offers a similar inversion of reasoning.
Most people intuitively understand bargaining. You have something I want. I have something you want. In many cases, we can trade with one another and both end up better off. But, things are not so clear when we expand our view to encompass an entire market. There are many agents interacting in the market and there is a great deal of disparity between agents. There are rich consumers and poor consumers. There are big businesses and small businesses – some businesses produce internationally, some are domestic. How can such a diverse set of agents be expected to interact?
Economics’ strange inversion of reasoning is often referred to as the “invisible hand,” a term used to describe the self-regulating nature of markets. Economics posits that efficiency may be achieved by simple interactions by selfish individuals. A central planner is not necessary for efficiency. People pursuing their own interests, buying and selling in the competitive marketplace, is sufficient for an efficient outcome. Although efficiency is not necessarily the same as desirability, this is still a very powerful idea.
Let me reiterate that the above result does not prescribe desirability. We have only that competitive equilibrium ensures Pareto-efficiency. Pareto-efficiency refers to a state where no one can be better off without making someone else worse off. Wealth can be redistributed, the market process proceeds, and that result is efficient. Through redistribution and market forces it is possible to achieve an efficient outcome to fit any set of values, except values that require impossible outcomes. We have what Milton Friedman termed cooperation without coercion.
Emergence
What I hope to have illustrated is that incredible results can follow without comprehension, without centralized planning. Complex organisms can arise from blind trial and error. Random mutations, trimmed by the forces of survival and reproduction can account for all known life. Agents pursuing their own self interest achieve efficient results without any oversight from a central authority. Selfish interaction sees results that are at least as efficient as the best possible outcome that central oversight can offer (and as oversight is not likely to be omniscient, omnipotent, or free of cost, selfish interaction is probably much more efficient).
Emergence and Starcraft II
Impressive, complex products happen without anyone understanding how they came about. This applies to Starcraft II. It may be weird to try and design units without a specific purpose in mind, to trust that people can find a use for your creation rather than to create a unit to accomplish a particular task, but experience shows us that this is the subtle spring of greatness. Design your melee units with claws, horns, and lightsabers. Design your ranged units with Phase Disruptors, machine guns, and Tachyon Cannons. Don’t design Marauders to beat Roaches. Let the players choose the units’ roles. It’s the curious nature of strategy games that the players create the game even more so than the game’s creator. Whose product will be better in the end? The product of a few hundred thousand hours put in by Blizzard employees or the product of billions of hours invested by millions of players around the world?
My question is: Blizzard, why are you trying to control everything? Why are you trying to mold your units to such specific purposes? Why are you removing or omitting so many multiplayer features? Why are you putting so many limitations on what users can create? Why are you limiting what users can share with one another? You seemed so much wiser when you were younger. Do you believe you can foresee everything - do you believe you can foresee anything?
Did you invent the Terran wall?
![[image loading]](http://wiki.teamliquid.net/starcraft/images/8/85/TerranGap3.png)
Did you imagine the PvZ fast expansion, with essential building placements?
In 1998, were you dwelling on the strengths and weaknesses of Vulture, Tank in TvP?
What about Corsair, Reaver PvZ? Did you see that one coming? Did anyone?
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/KTCorsairs1.png)
How was your Mutalisk micro back in ‘97? Did you prefer to stack them with Overlords or Larva?
Lessons of the Past for the Future
What you created in Starcraft: Brood War was a playground with lots of cool toys. When one toy proved more desirable than the other toys, you increased its cost, or you delayed its production. Or you improved the other toys. The toys had distinctive features. Some toys were sturdy, others were fragile. Some were very dangerous.
But you didn’t tell the kids what to do with their toys. You didn’t make toy A to beat toy B, and then toy C to beat A. You made cool toys to be cool toys. You gave them cool functions so that they could do cool things. And you let the kids decide how to use the toys and their functions.
We’re not asking for the same toys or the same playground. You promised us a prettier playground, and you promised us new toys. That’s what we want, and that’s mostly all we want. You’ve given us a prettier playground. We like this. The toys are shinier than before. We like that, too. Some of the new toys are really cool, but a few of them not so much. We don’t want toys that can’t play with each other. Toys should play best together. Toys should have lots of friends.
Don’t tell us we can’t use the jungle gym without parental supervision. Don’t tell us to stay out of the sandbox. The playground should be open 24 hours, and people of all races should be allowed to play together. Put the cheaters in timeout but don’t punish an essentially good kid because his name is a little unusual. Don’t get so caught up deciding between blue paint or red; just make sure whatever color you choose isn’t lead-based. If you’re torn between a plastic swing set and a wooden one, just make sure we get a swing set.
It’s not too late. There are two months before you release the first installment of Starcraft 2. You have two expansions. You have patches. You have a lot of money. You have a staff. You have a reputation worth upholding and improving. Be willing to make changes right now. We are.
Some related links
+ Show Spoiler +
 
|
FREEAGLELAND26781 Posts
This is very good, and the tie-ins to biological evolution and economics really drives the point home. Oh, and the analogy at the end about the playground is pure gold.
5/5 
|
I study economics but I fail to see the point in beta with economics. Are you talking about the managing the resources in game or the blizzard market POV?
|
I agree with flamewheel 
5 stars. I think that this will open people's eyes if they weren't aware that blizzard is being overbearing. (if that's possible not to notice.....)
|
i saw corsair/reaver pvz coming 
naw, in all seriousness good post, just like most of your posts I have read
|
5/5.
This is the most succinct description of the differences between BW and SC2 I've ever read. Your analogies are spot on.
I don't think Blizzard will change the game so radically now though. And we might see the game become something absolutely great, as Brood War became, but I doubt it.
On May 25 2010 04:28 PaD wrote:
I study economics but I fail to see the point in beta with economics. Are you talking about the managing the resources in game or the blizzard market POV?
I think what he's saying is that the function of a unit shouldn't be handed to us by the game designers, and if the players are given a good enough sandbox, then these functions and unit roles will arise themselves as the players get used to the game.
|
On May 25 2010 04:28 PaD wrote:
I study economics but I fail to see the point in beta with economics. Are you talking about the managing the resources in game or the blizzard market POV?
No, the point of talking about economics was not to talk about economies but to illustrate emergent systems with a hopefully thought-provoking and less common example.
I think that many people appreciate the dichotomy that's usually created when we talk about theories of creation and evolution. Hopefully it's clear that evolution is a remarkable example of emergence.
I don't think that most people appreciate the marketplace as being a very remarkable emergent system in itself. I doubt many people are aware that a competitive market has an outcome that is at least as efficient if not better than the best possible supervised system of trade (at least in many reasonable cases). In other words, an emergent system is at least as good or better in terms of efficiency as the best centrally planned alternative.
I'm trying to make a case for Blizzard maintaining a laissez-faire approach to certain aspects of Starcraft II that history shows to be the province of emergence rather than of planning.
Blizzard makes the units. Blizzard creates the map-making software. Blizzard creates a single player game. Blizzard makes Battle.net 2.0. We buy Blizzard's game. Blizzard profits.
Blizzard doesn't specify how the units should be used. Blizzard doesn't artificially limit mapmakers. Blizzard doesn't deny us LAN support. Blizzard doesn't deny us replays with friends. Blizzard doesn't require us to buy multiple copies of the game and to create 3rd party software in order to play with our friends who live in other places. Blizzard doesn't require that all tournaments be run through Blizzard.
There are a lot of things Blizzard is doing that it shouldn't do. And there are some things that Blizzard should be doing that it isn't. Emergent systems are a great alternative in many cases, and they work really well. Both in video games and everywhere else.
|
Thanks for everyone's support ^.^. When I'm stuck performing a pointless desk job my favorite entertainment is reading and writing. it's reinforcing to receive positive feedback
|
United States33339 Posts
just saying, you could have just invoked common sense instead of any economic theory and made the same point.
|
On May 25 2010 05:02 Waxangel wrote:
just saying, you could have just invoked common sense instead of any economic theory and made the same point.
Yeah I realized that while writing but common sense isn't common
Anyway I'm an econ grad student / assistant, and I'm at work. It's academic policy requirement that I reference econ whenever possible while I'm on the clock
|
On May 25 2010 04:02 Failsafe wrote:
It may be weird to try and design units without a specific purpose in mind, to trust that people can find a use for your creation rather than to create a unit to accomplish a particular task ...
Actually i remember Browder mentioning their design policy (interview from ~a year ago?) to be just that, to make "cool" units with as much diversity as possible and then test if they feel right and are fun, and balance them later.
I think it's a good policy but they failed to make game engine and basic stats that could suppress the "best option is much better than second best" effect - like exponential growth of unit(s) effectivnes through massing it or making a timing rush / push that enemy cannot deflect without almost equal tech or army size.
For me the problem lies in the lack of things like: longer distances on maps, smaller chokes (comparing to army ball), high ground advantage, unit clumping and splash, worker/units glitches witch make room for error and recovery, or less than 100% efficiency to work.
anyways, i agree with OP and 5/5 nice read
|
On May 25 2010 05:13 Failsafe wrote:Show nested quote +On May 25 2010 05:02 Waxangel wrote:
just saying, you could have just invoked common sense instead of any economic theory and made the same point. Yeah I realized that while writing but common sense isn't common Anyway I'm an econ grad student / assistant, and I'm at work. It's academic policy requirement that I reference econ whenever possible while I'm on the clock
+ Show Spoiler +
5/5 I agree completely with you and the analogy with the playground was superb. oh and yes i had to post the image
|
Really nice blog entry, thanks for this.
|
|
|
5003 Posts
Economics’ strange inversion of reasoning is often referred to as the “invisible hand,” a term used to describe the self-regulating nature of markets. Economics posits that efficiency may be achieved by simple interactions by selfish individuals. A central planner is not necessary for efficiency. People pursuing their own interests, buying and selling in the competitive marketplace, is sufficient for an efficient outcome. Although efficiency is not necessarily the same as desirability, this is still a very powerful idea.
I've always said that the best way to balance the game was to host large, frequent, cash tournaments while making all information public. Providing incentives and making conditions ideal for a competitive market is key (like perfect information), or else you could still have asymmetrical information or some other conditions and have the market adjust slower, or not be at an equilibrium to begin with
Let me reiterate that the above result does not prescribe desirability. We have only that competitive equilibrium ensures Pareto-efficiency. Pareto-efficiency refers to a state where no one can be better off without making someone else worse off. Wealth can be redistributed, the market process proceeds, and that result is efficient. Through redistribution and market forces it is possible to achieve an efficient outcome to fit any set of values, except values that require impossible outcomes. We have what Milton Friedman termed cooperation without coercion.
A "technical" point but you're not guaranteed pareto efficiency on a competitive equilibrium, since a competitive equilibrium is more like a nash equilibrium in this case if anything since this is game theory (ie: strategies used by other people affects me, so I consider that). The optimal strategy is likely going to shown by some nash equilibria and the nash equilibria doesn't have to be pareto efficient. In fact, you can end up getting some nasty equilibria if there's racial imbalance within the game.
But that also depends on "how you define it". Suppose there's an imbalance that makes TvP and TvZ broken, and TvT is a terrible game, but PvP, ZvZ, PvZ is amazing. Everyone stops playing Protoss and Zerg slowly and plays Terran, which means more and more people will start playing Terran, and depending on actual composition you could have ended up in a much worse situation for all parties.
A competitive market will make the imbalances obvious, but it doesn't mean that the competitive market will guarantee there will be nothing broken. If something is broken, the competitive market will show it since everyone will use it, and that's how you start making adjustments after.
I study economics but I fail to see the point in beta with economics. Are you talking about the managing the resources in game or the blizzard market POV?
Economics applies to "more things than money".
Think of the game itself as an economy. The player maximizes utility by winning the game, and given that they know what strategy that the other players are using, they will find the optimal strategy in order to accomplish that. The optimal strategy is the one that maximizes my expected utility, and my optimal strategy is made up of a "bundle" of units and timing pushes.
One of the key things you need to learn to do with econ is to learn to apply it to anything and everything. It'll make things more fun
|
Dustin Browder did say that their approach to unit design was to make cool units without any specific purpose and see how it went from there. So in some way were doing what you were suggesting.
Anyway, I don't think the analogy between evolution and game design is good because "evolving" a good game takes too many resources. You would need many copies of a game being play tested having random mutations distributed among them. After measuring the fun of each game you'd have to devise a mechanism determining which versions of the game go on to the next round, and so on. Creating a game like that is just crazy since anything could be a parameter, and the funness function would be hard to define.
Your last section "Lessons of the past for the future" is closer to intelligent design than evolution since there is only one game that Blizzard changes, and not many different games competing with each other for balance and fun.
Another thing to talk about if we're to talk game theory is Blizzard is only one agent looking out for itself. Even if Blizzard ruins it for everyone else by not including features, it obviously believes that not including those features benefits itself. Including LAN might encourage other BNET like entities to pop up quicker which may generate more competition, but is less beneficial to Blizzard. So saying features that Blizzard should or shouldn't include makes no sense. The reason why Blizzard is trying to control things is to benefit itself, not the entire ecosystem. Maybe this will backfire but like you said Blizzard can't see into the future with certainty.
|
Wow, nice.
This should get the spotlight on the front page imo.
|
|
|
|
|
|
EPT![[image loading]](http://www.synapsepc.com/Hosting/Images/Funny/%23Demotivational%20Posters/Common%20Sense.jpg)
