"Most good products are designed around the person, not the technology." ” Donald A. Norman , author of The Invisible Computer , quoted in a Technology Review magazine article [1] [1] See "Handhelds of Tomorrow" by Claire Tristram, Technology Review , April 2002. For an industry with over 30 years of history and populated by some of the brightest people you'll ever meet, we sure seem to be unable to learn from the mistakes of others. There seems to be an almost emotional, arrogant need to reprise those mistakes, sometimes by the same teams on subsequent projects. A commonly heard refrain from developers is "<Insert feature here> hasn't worked yet because no one has done it right." So, even though non-consensual PvP combat (the player being attacked has no choice in the matter) has been shown in any number of games since 1988 to be extremely unpopular with over 80% of the player base of subscription-based online games , many new development teams plan it into their game on the theory that it just hasn't been done right in the past. Considering that concept, we thought it would be useful to include advice from other people who have been there too, in the hopes that the information would sink in institutionally. We asked several experienced executives and developers the following question: In your experience, what mistakes do development teams make during the online game development process? The answers we received offered some excellent practical advice for anyone starting an online games project. Richard A. Garriott, creator of the Ultima series and currently a principal at NCSoft in Austin, Texas: -
Teams regularly bite off more than they can chew. An MMP can easily become an impossible -to-complete reality simulator. -
Teams regularly hire on-paper game designers and/or solo player gamemakers, but MMPs need unique skills not found in other arenas. -
Teams often cut corners in code expandability or general quality to try to make schedule. This usually haunts them later. -
Companies often hire people too soon. A great deal of groundwork can be done before you need the majority of the staff. Gordon Walton, executive producer for The Sims Online for EA/Maxis: -
They don't build for the long term , that is, they don't make the code extensible and maintainable . The live team pays for this for years and the game's financial potential is reduced. -
They don't focus enough on quality of service, especially mean time between failure (MTBF) metrics. Failures in the servers deny service to thousands of people, and failures in the client irritate players, raising churn. -
They underestimate the complexity of the task. More moving parts drive complexity nonlinearly, and MMP games have a lot of moving parts . Rule of thumb: An MMP game is three times bigger than a standalone game, but 10 times harder to complete. -
They underestimate the testing and scaling challenges and end up with a fire drill when their service scales up. -
They overestimate the value of the initial game content and overestimate how long it will take the players to consume it. -
They forget that some design ideas don't work well within the MMP arena, particularly static puzzles and fixed-content discovery elements. These elements end up on player web sites within days of launch, reducing the value of that part of the game for most other players. -
They spend more time on game mechanics than social mechanics. This makes it harder for players to find and play with each other, which doesn't help retention and churn at all. -
They forget to bring customer service (CS) in as an equal partner in design and development. Thus, customer satisfaction is lower, and the game is much more expensive to run. -
They think they are just building a game (rather than a social venue with gaming elements). Thomas Howalt, project manager of Funcom's Anarchy Online ( AO ) : -
Confidence is good, but control is better. As stated earlier, an online game is a huge task, and it is very easy to lose track. Check everything thoroughly again and again. Do not accept promises; you have to keep track of everything. -
Problems don't solve themselves . Personnel conflicts have to be dealt with quick and clean. Do not accept discrepancies to pester the atmosphere. Conflict can be valuable if managed well; tension can be fruitful if managed well. It all depends on how mature your team is. Remember: Moss does not grow on rolling stones! Make 'em rock 'n roll! -
Everything moves. Everything in the game may have to change. Be careful what you hard-code. -
Repeat after me, please : "Tools! Tools! And more Tools." You will need tools for everything! Hire lots of tools and library coders. Use money to develop brilliant tools. Make decent and efficient graphical user interfaces, or GUIs. And, force the coders to work with their own tools before they ask anyone else to work with them. -
Mountain high, river deep. Be aware when the "stars" appear. The closer you get to launch, the more need Marketing will have for someone who knows the game and is good at talking marketing-type talk. To all you "stars": Fame is very addictive , but keep your feet on the ground and stay humble ! Tomorrow you will be forgotten; that's the way fame works ”it does not last. And remember: It is a team effort! Alex Macris, CEO of The Themis Group and designer of Spearhead : -
They focus too much on excitement (gameplay) and too little on involvement (the community). Within six months of playing 20 hours per week, any player will find your gameplay boring, period. There is nothing you can do that can make your gameplay more interesting than that of a brand new AAA title coming out six months or a year or two years after your title. But if you have him in the thrall of your community, he'll stay. -
Everyone talks about the addictiveness of EverQuest 's gameplay, likening it to a slot machine, but they lose sight of the meta-game that overlays EQ and that makes the achievements meaningful. And I think future games will have a difficult time in succeeding with an EQ -style approach because they won't enjoy the network effects that come from being the biggest game. -
They focus too much on content and too little on features. Content can be successfully added over time (as Asheron's Call proved). Features are far harder to add. The number of game mechanics is a lot more interesting than the number of dungeons. Andre Backen, former president of Funcom: Jeff Anderson, CEO, Turbine Entertainment: -
The worst thing I have seen repeatedly is a lack of focus. The teams feel compelled to engage in what I like to call "competitivism." Before I was making online games, I worked for a company that made combat flight sims. I noticed that the focus of those games changed from what was "fun" for the general audience to what they have to have on the back of the box "to be competitive." So instead of spending their precious development time innovating , they worried about the physics on the F-22, or making sure they had the latest laser-guided bombs . Maybe it made for an easy marketing angle ("We've got that!"), but it made lousy gameplay. You see a lot of that in the online space already; fervor to make sure that my game hits some kind of feature list ”even before there is an idea of what makes the game fun. Damion Schubert, founder of Ninjaneering and former lead designer of UO2 and M59 : -
A mistake that I've seen a lot of in the MMP space is a concept I call "ant farming." This is when developers start to envision cool experiments that would be fun to observe from the heavens ”but may not be terribly fun for the player base. Usually, these experiments involve ways that players can take advantage of other players' lack of knowledge, experience, or common sense. Developers usually have great defenses for these features, such as "interesting emergent behavior," "case studies in trust," or "fascinating guild tactics," but frequently what they really turn out to be are support nightmares and retention killers.  |
