#2 - Dec. 12, 2008, 5:41 p.m.
There's not much logic to it, at least in the sense that we have some kind of Standard Operating Procedure that we follow.
Some changes are very simple and a designer can just go to their desk and make a change in a few seconds. We try to hit lots of these rather than let them all get bottled up behind the big, complex issues we are wrestling with.
Other changes are almost as easy. A designer feels like nobody will really object to a change and that it's pretty low risk. At most they might bounce it off one or two more people as a santiy check.
Other changes require a lot more discussion. We might not all agree. We might need to run some tests or calculate some numbers to see what the effects of changes might be.
Sometimes we don't know how to fix a problem. A great example of this would be a case where a spec is doing too little damage in PvE and too much in PvP. You can't just juice talents or coefficients without making them too good in PvP. These kind of changes require a lot of brainstorming and creative solutions.
Controversial changes require a lot of research. We read the forums. We ask our friends or thought-leaders in the community what their opinions are.
Some changes are risky from a bug standpoint. These require a lot of testing to make sure they work. Having a spell just fail is bad. Having a spell cause an exploit where players can get loot they haven't earned or win an Arena fight they should not have won is worst. Crashing the servers is about the worst thing you can do (but most of us have done it at least once.)
Some changes are hard to make and require programmer support. Maybe we need new tech to implement a feature (this was the case for a lot of death knight abilities).
Some changes require buy-off from a large number of people. Changing a talent is a relatively minor fix. Making big changes (say changing how arena works, or major class mechanics) are the kind of thing we want to run by just about every senior person on the team to get thier feedback.
I get the sense sometimes that players expect (or at least want) us to attack problems based on which is the most pressing problem. That is definitely an important consideration. But it is all about triage. Sometimes even a pressing problem isn't worth handling right away if that means 30 minor problems can't get fixed. Developing games is a pretty chaotic process. There are a lot of interruptions. There are a lot of considerations that have nothing to do with the actual product going on the shelves. In a game as large as WoW it can be easy to get demoralized because you just don't have enough hours in the week to get to everything you want to do. Features get bumped to subsequent patches or expansions all the time. (But for all the chaos, it is still a lot of fun).
Also, when you say that an entire tree was unfinished, you have to understand from our point of view that we feel that way about every tree. Blizzard is it's own worst critic. If you asked us which tree was the closest to being perfect, we would be very hard-pressed to answer that. We aways think we can improve on the game, down to individual talents. We are never going to stand back and say "That's it. We nailed the Survival tree. It's done."
