Architectural Service Design

Now that’s a heading that should make people’s heads hurt.

I’m in the middle of a rather large service design project now and it’s making me think long and hard about similarities in games.  There are 4 main phases; design, migration, steady state and close out.  I am chest-deep in the first one and I talk about this a lot on the blog.  The other three, let’s get a bit more meat on.

Migration is the period between nothing and operational state.  This is paperwork stage, signing agreements and whatnot.  It’s when you buy your ticket to the ride.  Steady State is the day to day activities.  Close Out is when the service is about to be shut down.  There’s a lot of this one lately.  In simple terms, from a design perspective you need to figure out how to minimize impact to users during migration and ensure that steady state meets expectations, otherwise close out happens.  In practice this is more complex since expectations are all over the map for steady state.

If I take a console game as an example (BioShock Infinite or Ni No Kuni), the process of migration is simple enough.  Buy the disk, put it in, patch (maybe) and play.  There are no extra bells and whistles, you’re in.  MMOs you can’t really buy the games anymore, you’re downloading them.  There’s the signup, payment methods (PayPal should be an option everywhere), patching and then you get into the game.  That game part is also a problem since character creation, for many games, is done poorly.  Customization options are often lackluster and irrelevant after a few levels.  Class/race selection usually have a dramatic impact on gameplay but without the context for players to understand.  Someone starting an MMO cold is going to be confused and likely alone.  I went back to SWTOR recently and it took about 4 hours of reading forums and websites to have an idea what was going on.  Barrier of entry is a problem.

Steady state is also a fun one.  Again, the console example has you play a contained experience which is cohesive.  I mean that the game from start to end is logical, systematic and if you play the game you should be able to follow track for all content.  Batman doesn’t suddenly turn into a FPS game half way through. MMOs again have trouble here.  For some reason, many try to make 3 games in one.  First, is the leveling experience.  Heavy on story, exposition, relative balance.  Very lackluster on world integration.  You consume, move on and never really look back or understand your relation to the rest of the world.  Second is the “end game” aspect, where you’ve reached the end of the levelling experience and now have a list of a dozen things you can do.  Hunt knick knacks, get bigger numbers on your equipment, beat big bad guys.  This is, sadly, skinner box material.  Third is PvP.  This is usually a bolt on mechanic, with parallel gameplay and rewards.

 These 3 components are rarely integrated.  Leveling is often-time the only part people want to play since the disconnect at max level is just a wall of grind.  There’s no real progress except for numbers.  I mentioned in the last post that WoW leveling is a face roll of challenge, and then you reach the max level stuff and realize you actually need to use some of those skills you got 50 levels ago. SWTOR is somewhat interesting in that you need to use ALL skills to do leveling content.  PvP, other than a handful of games, has no bearing on PvE.  Since UO took the knife to the problem, no game has really put effort to figure out this problem.  Heck, FireFall has pretty much thrown in the towel even though it was pitched as PvP only.

Games today have a significant challenge to come out of the gate.  First, there are few people entering MMOs cold and they have expectations.  If your game’s Migration phase is different than existing models, it need to be ultra smooth and intuitive or you’re going to lose people.  If you want people to stay around after the leveling portion of the game is done, make sure it is tightly integrated with other systems.  GW2 is a good example where leveling content is also seen as end-game content.  If you want PvP in the game, make the social aspects obvious and integrated.  Have it affect the PvE world and vice versa.   Change zone “availability” based on PvP results and make those zones relevant.

I love the challenge of architectural service design.  I think it’s one of the most complex and overlooked parts of development.  If done well, and expectations are clearly understood, then meeting those same expectations is in the realm of possible.