Interesting to see how Mojang/Valve's "work on what you want" policy might be Considered Harmful. Not wanting to finish a game, tune a webserver that processes your sales or freeze and document an API because it's hard.
Even with the API where Mojang hired the developers of the de facto mod API Bukkit, the development seems to have just continued on Bukkit while the promised API languishes for 8 months.
Yeah, I had a talk with someone from valve awhile back and he mentioned that only a certain kind of developer works well in the "work on what you want" strategy. If you don't have goal driven people who know what they're doing, nothing will ever ship.
On the other hand I see Mojang, the company who in its infancy hired the popular and attention-obsessive "minecraftchick" to do PR/Advertising rather than someone who actually knows what they're doing.
On top of this, Valve has the kind of financial backing where they can screw around all day and still not go bankrupt for awhile thanks to their previous work in Steam. Mojang on the other hand is only making income on merch and the 5 people who haven't already bought minecraft.
As much as I hate to say it, the more I read about Mojang, Minecraft, and the decisions they're making, the more I feel like the success of the game was an act of serendipity rather than one of skill.
> the more I feel like the success of the game was an act of serendipity rather than one of skill.
I see it the same way, and honestly when I hear Notch speak about all of his success he seems about as confused as everyone else. I'm happy for the guy, because he worked hard on something he was passionate about and it was a huge success. But at the same time he kind of hit the lottery with Minecraft.
>Interesting to see how Mojang/Valve's "work on what you want" policy might be Considered Harmful. Not wanting to finish a game, tune a webserver that processes your sales or freeze and document an API because it's hard.
In terms of maximizing profit, sure. Seems like they have plenty of money though and are optimizing creativity/happiness whatever.
I do wish they'd figure out how to handle an API but it looks like a really challenging problem.
It might also be said that Valve showcases the dangers of this philosophy on their own. Isn't there a game about lives being halved or something that people are waiting for a new version for? :)
It's not necessarily a danger to not release a sequel immediately if they've got other things to work on. People will still buy Half-Life 3 in a few years time - similarly Blizzard have had ten-year-plus gaps between Starcraft - Starcraft 2 and Diablo 2 - Diablo 3, which hasn't seemed to do them any harm.
That's distinct from not working on something simply because it's "boring" or "too hard".
Notch could be suffering from programmeritus - a condition where a programmer works a difficult, but fun problem that doesn't help reach the end objective (in his case, making a fun game).
We've all been there - spending months doing something that in the end doesn't matter.
Notch has the ability to make fun games quick, I'd gladly pay for a full version of the last game jam he participated in. If he made a series of demos, then chose the best one he wouldn't be in the dilema he's in now.
"I kind of see this a little bit like my hobby: just being able to just work on games and not have too much external pressure that (those games) have to make money or whatever."
Maybe that's the problem. No pressure can also take toll on finishing something.
It's not a problem by itself. Nothing wrong with fooling around. However, if your objective is to produce something then it becomes a problem. When there is a set deadline or a resource limit you're forced to make decisions, while if there are no limits it's easier to fall into "creative blockages".
Maybe more structure and less treating Mojang like its your personal frat house is the answer.
Minecraft has allowed them to bankroll certain proclivities that the whole "we're cool nerds" culture Persson likes to perpetuate; but if they want to be taken seriously as a gaming company, they might want to knuckle down a bit and use some of the tried and true methods for bringing a project to market.
I work for a Swedish father/son company. Interestingly, the son has the same problem; take something interesting 90% of the way then wander off to a new shiny before the project gets done.
> But you couldn't use the spaceship CPU to subvert the basic tenets of the game. Put another way, players would be working within the confines of the existing system when coding in 0x10c, not dipping into the game's raw code.
Modifying the game from within actually sounds like a good idea.
Even with the API where Mojang hired the developers of the de facto mod API Bukkit, the development seems to have just continued on Bukkit while the promised API languishes for 8 months.