Short update this time. Puzzle barriers implemented this month: seven. (Some, again, with multiple solutions.) Also another substantial chunk of the automatic move-around-the-map code. That has been going in slowly because it's so integrated with the puzzles -- going from one area to another usually requires a puzzle solution or two.
As I said in February, this is a weird development process, because I am implementing both the puzzles and the mechanism to bypass the puzzles. So it feels like there's no more game here then there was in January. I can start the thing up, type "ZAP-OMNI" (to mark all the puzzles as understood), then type "GO TO ANTECHAMBER" -- that's the second-hardest room to reach in the game. Zwoop. 41 lines of automated activity, and I'm in the Antechamber.
To be clear, "ZAP-OMNI" is a debug command; it won't be available in the final version of the game. (And I'm skipping over a couple other debug commands I used, to compensate for links in the Antechamber chain that I have not yet implemented.) But "GO TO ANTECHAMBER" will work, just like that, once you have solved all the puzzles in the way. The point of the game is to make those tools available to the player.
So what makes this interesting gameplay? Like I said: puzzles with multiple solutions... and solutions that apply to multiple puzzles. It's no good getting into a room if you've used up the items you needed to have when you got there.
Beyond a certain point in the story, you need to start managing your automated solutions, making sure they all mesh together. That's going to be the real game. And I have now implemented enough of the map to start seeing these effects. Select certain combinations of solutions, and type "GO TO ANTECHAMBER", and the game says (very approximately): "Sorry, you got to the last door and you're stuck, because you've used up the rod of metal X getting that far. Also, when you swam through the flooded tunnel, the Y got soggy and the Z dissolved. Tough beans."
So we go.
Other news: My Unicode Parser for Inform 7, which I mentioned last month, is now available. I also helped get the Inform 6 web site back on line. It's been neglected since I7 came out; then, a few months ago, a server fell over and nobody's had the spare cycles to fix it. The site is still not particularly up to date, but it should now be stable, at least.
My Secret Project for April is not done, but it has reached the major implementation milestone that I was aiming for. I now flip over to the Secret Project for May, aka Secret Project STW-5, which begins... as soon as I upload this post.
Update on one Secret Project from January/February:
Graham Nelson just announced that the next release of I7 is "fairly close": http://www.intfiction.org/forum/viewtopic.php?f=7&t=7767
One of the new features is the absorption of my unofficial I7 manual index. I spent a good couple of months updating it for the new release, and it will be shipped as part of the IDE documentation henceforth. I am excessively pleased and proud.
Wouldn't it be a good idea to have duplicate the puzzles with multiple solutions and then selectively disable the solutions the player already knows?
I'm not sure if that would feel weird but it seems you could get a few more fun puzzles out of it that way. :)
I'm not sure what you mean. Are you suggesting to have several identical puzzles, but require that the player use a different solution for each one?
One could write a game that way, but it's not what I'm doing. The constraints are supposed to flow naturally from limited resources.