You know that feeling you get when you see a continuity error in a movie and it completely distracts you from keeping up with what's happening until you're not only hopelessly confused but you missed what the bad guy said, too? That's how my life feels about now.
You see, I'm writing a non-linear, choose-your-own style lecture. There are multiple choices, multiple paths, and I'm doing the entire thing in Powerpoint, with pictures. I'm used to writing non-linear dialogue in an editor, but when it's a matter of "Do I create 21 alternative question choke points and just try to remember to click on the right one?" or "Will they really care if they can take the same option twice?" my brain becomes a little mushy.
Luckily, I have a programmer on my side. He says the latter. And I'm very grateful for it.
Similarly, if an interaction could possibly be called on twelve separate occasions, or possibly never, do I make 12 variations and hope? Or 6? Or just one, and replay it? I've gone with one, because time is a factor (I'm presenting tomorrow, whoops!) but even if it weren't, is there really any merit to spending time on split-second interactions that may or may not be seen?
And, really, that's what game writing is all about. It's a little easier than what I'm currently attempting - it's only words, usually, with perhaps a gesture or two - but the principle is the same. How do you decide where to cut corners? What happens if the player notices?
Since this is a lecture, I can wing it a bit. I can steer my audience away from repetition, and hopefully make them feel smart in the process. But in a game, where you have no direct impact on your audience, providing that guidance is so much harder.
When writing branching dialogue, I try to link different paths to different outcomes, but never repeat the same line. Sometimes that's simply not possible, as with end-of-conversation comments, but in that case, I simply try not to make them too memorable. Writing a memorable parting line that's going to be repeated will only make the player more aware of the repetition. Better they experience one shining moment of brilliance than the reflection of a candleflame in a hall of mirrors.
But what of those paths that are all equally valuable? How do you make them shine? Personally, I try to imagine my own delight if this were all new to me - if I were playing this conversation for the first time. If I wouldn't be excited by the dialogue choices, why would the player be? Of course, that's when outside testing will be very useful, to get new eyes looking at your project and picking out the moments when you let the shimmer slip, but I find that trying to imagine the game as one I'm playing is a good start.
For a lecture, though, which is guided, trying to teach, and ultimately only has one ending (in this particular circumstance, anyway), it's far easier to know the paths and guide the audience away from the dangerous (i.e. repeated) ones. In the end, they'll never know. But with the ability to replay your dialogue, the player will.
Practice with Powerpoint first. It's as good a place to start as any.