Alabaster Open Beta

Okay — it’s much later than I would have liked, but Alabaster is now, I think, ready for beta.

This is an unusual sort of beta for IF. Usually I distribute a game file and collect information about how it plays. This time, because of the unusual origins of the thing, I’d like the beta to be more open, with contributors able to see how their materials were worked in and how the mechanics of the game function.

To that end, a) I am putting up the source, the conversation diagram, and the plot tree for people to work from. I have testing commands in place and have (I think) proven to my satisfaction that the plot tree works… which is exactly the sort of hubristic statement that guarantees the discovery of enormous bugs by tomorrow morning.

Also, b) I have left in a lot of the construction materials. You can, for instance, see in the status bar a bunch of stuff about Snow White’s mood and the current state of the conversation in terms of how many quips are used up / available / total in the game. We’ll naturally be getting rid of that in the final version.

How I envision this working:

— If you are testing, please keep a transcript and send it along with any bug reports. Without transcripts, it is nearly impossible to debug complex state-related bugs in conversational games. This is critical.

— The conversation builder is still turned on, and can still be used to generate new quips for inclusion. If you want to use it to fill in missing little snippets of dialogue, you may, but this is a good time to do it sparingly and with regard for what other people have written: the idea at this point is to fill in the remaining gaps in the game, not to add big glorious new plot threads. I reserve the right not to accept new conversation that would redirect things in massive, hard-to-code new ways.

— If you encounter something weird and want to go look at the code, you’re welcome to do that; I’m happy to receive actual patches as well as reports of things broken.

— I also welcome feedback about the way the story winds up being folded together, though I guess I welcome it more from people who had a hand in the first half of this project (or at least, I’m more likely to act on it). There were a lot of different ideas from the contributors that got combined into the final plotline, and I hope I achieved something coherent that doesn’t too terribly betray anyone’s vision. But… who knows. Tell me if I totally broke your idea.

— If you prefer to test in a black-box, how-would-the-player-really-experience-this way, that would also be very useful. In that case, ignore the source code and charts, and when you start up the game, type PLAYER MODE ON; this will turn off the conversation builder and create a somewhat more natural playing experience. Feedback about how the game is paced, how the endings are clued, etc., would all be useful especially from black-box testers.

Finally, re. performance: I’ve optimized somewhat, but at the expense of a longish load time. I’m working on further improvements, but for the time being, starting the game now does a bunch of work so that the lapsed time between turns will be shorter. Sorry about that. If you have access to a functioning version of Git, I suggest trying that.