It’s also possible for development of a game to be rough.
There are two kinds of endeavors: ones that are bound to succeed with time (sorting papers, assembling simple machines), and ones that can potentially fail, no matter how much time a person puts in (trying to solve a math problem, trying to invent a drug).
Considering that Valve’s games (l4d2 notwithstanding) are never cookie-cutter, and always try to fulfill strong ambitions, the development of HL2:Ep3 is definitely the latter kind of endeavor.
The development of Ep3 (or most games) is not something that you can connect to a ‘progress bar’, and you can’t really expect it to have gone very far just because a lot of time has passed.
Imagine trying to write a fitting ending to the Half Life saga. Writing fiction is not easy, and even though they have Mark Laidlaw, they may have run into a dead end with the story (remember, they don’t fully plan for sequels when writing the story of a HL installment).
Also, they’re suddenly faced with all these games whose graphics dwarf source engine’s graphics, and make the kinship between source and gldsource all-too-apparent. Updating an engine, or making a new engine, is a monumental task today. Valve has some very capable people working for it (many veterans of Computer Science and game development), but still, the task of keeping up technologically is daunting.
Finally, they’re trying to realize all the ideas that they cut from the previous HL games, which can be problematic.
That being said, looking forward to new info from valve every year and being disappointed is, at this point, hilarious. I’m just going to ignore Valve until EP3 shows signs of life (yeah, I’m buying Portal2 when Ep3 is out, no sooner).
