I don't think she does a very good job of explaining why good game design is difficult.
It's not that game design itself is difficult, it is that GOOD (ie, fun) game design is difficult. She's basically addressing the wrong problem set. What she is describing is simply software design and engineering issues, which boils down to 3 real categories:
1. Functional / feature design: the rules which govern whether they exist and how they can function. AKA "business rules" in normal software development.
2. User Interface design: how the user (player) interacts with it.
3. Engineering/Implementation issues: how do you make 1 & 2 real and work, while reducing undesired side-effects.
1 & 2 generally form a specification for the feature's design, and 3 is the specification for how to implement it.
This is not unlike many common design and implementation processes for standard software design and engineering of complex systems. The real difference is that, while a software system designed and implemented correctly may fulfill all the intended design objectives, there is an additional objective which games add to the mix that is not generally present in normal business applications: fun. Unfortunately, it is not an objective criteria, and requires "play-testing" to discern whether a particular design is fun or not. It is very difficult to design-in "fun" from the very start of a project.
That said, with the advent of Serious Games, adding the "is it fun?" criteria to real-world business applications is happening more often.
Lastly, as a game developer, the single greatest challenge I have encountered is simply to keep going through the "hard times". Like any difficult software development project, there are times when things get dark and depressing for whatever reason, and there is difficulty keeping motivated to continue, but you have to bear down and power through the hard parts. The reason most game development projects fail that I have seen is that people don't really understand how hard it can be at times, and give up when the going gets tough. To me, this is a more difficult hurdle than in typical business application development, because many people get into the development of games with an incorrect level of expectation about said difficulty.