The SimCity launch last week was awful. Long wait times to join a crashy client into an overloaded server cluster that doesn’t even have the courtesy to replicate your profile to other clusters.

The Color.com launch in 2011 was awful. An unusably obtuse client connected to a creepily empty social network running on two expensive domain names.

Dust 514 seems to be doing okay. A lengthy and obvious beta period allowed them to grow and iterate the game to its current state, and will (hopefully) turn a profit and continue to be a fun game (Mac release please, thanks!) You’ll notice there wasn’t one single huge launch, just a gentle progression. No huge queues to join games

Twitter has been famous for teething issues (euphemistically known as “failwhales”), which is the opposite of Color’s overcommitment. While there’s been days when it hasn’t worked (frequently during large social media/marketing conferences like SXSW), they’re around today, will probably be around tomorrow, and they’ve done a great job rebuilding the service to handle the load without obvious hitches.

Iterating software is cheap, iterating business concepts isn’t expensive, and good will is expensive. While many people1 have sworn off SimCity and other EA products forever, there’s been no noise or New York Times articles about problems in Dust 514’s early days.

You can scale later, but somebody that hates you for a botched launch won’t come back later.

Notes

  1. Well, gamers, not the most committed demographic.