I’m back from my travels, I’ve mostly recovered from my bout with the plague and my head doesn’t feel like it’s stuffed with wet towels any more, and I’ve written my first review: for Dr Ludwig and the Devil! It’s here.
I also drew a map as I played:
Tubmlr being a bit weird about making some post (www.tumblr.comp/user/post) visible if you don’t have an account, this link should not prompt anyone to create an account. 
Thanks for the heads up! I’ve updated the link.
I’m curious, though, because I can’t get Tumblr to demand that I log in no matter what I do. I’ll gladly use the username-as-subdomain link format in the future, but if you don’t mind, can you describe what happened before when you clicked on the link?
Thanks in advance.
I believe it lets you read a certain amount but if you scroll too far it’ll ask you to log in? I just tried the other link and had to scroll well past the end of your review before it popped up the login blocker.
Ah, gotcha. Thanks. It’s probably blocking everything after the spoiler cutoff.
Thank you!
Fourth review: I spent a week going down the rabbit-hole of The Witch.
There is a partial map, excluding most of the end-game areas:
yes, this is my first game and i think you pretty much nailed it on your blog. it is very much a first game; sort of a proof-of-concept for learning inform6. in hindsight, ifcomp may not have been the best forum for a first retro parser game. but, i’m a newbie at this and now i know. plus, the process itself has been rewarding.
there will be a 2.0 release taking into account the judging feedback. but i agree, the endgame puzzle is fairly contrived and kind of just is what it is at this point.
your thoughts on the game were spot on. just curious, do you participate in beta testing? i have a second game that i’m much happier with that’s probably a month or so away from being in testing shape.
either way, enjoyed your review.
I’m really glad it was helpful! And I think it’s worth saying again that there’s a lot I liked about The Witch, and that it was particularly ambitious in a lot of ways for a first game. I think a post-comp revised release could fix some of the rough edges.
I’ve done beta-testing in the past (but I’ve also occasionally dropped the ball and just disappeared without sending in a beta-test report in the past, including on a game that’s in the Comp this year – I’m trying to be better about that). I’m totally willing to help if life isn’t too crazy for me when you need feedback, though. Send me a PM when you’re looking for testers!
The third ending provides insight into what the wretched thing is, which I definitely found necessary to understand the story. But yeah, it is somewhat unfortunate that seeing it relies on chance…
I guess that makes sense, and I can see it as a design decision, perhaps intended to incentivize further exploration to get more information. For me, though, it seems like a bad choice, because the PC’s motivation is kind of a fundamental thing I need to have to experience the narrative in the first place.
I’m just in awe over here that you wrote a python program to play the game . . . O_o (re: The Witch)
Well. I cobbled together some code I already had for another purpose, and then adapted it, and wrote some glue code to run the code I’d adapted. In all fairness, I’d been meaning to do that refactoring anyway. It’s not really true that I did all of that just for The Witch. 
I really enjoyed looking through your GitHub code for The Witch and the other games. I hadn’t played Rope of Chalk but having the “visit every room exactly once” puzzle bumps it up on my list.
I’ve used thecrude grep "^\>" walkthrough.txt | clip or GREP "^\>" walkthrough.txt > walkthrough-cmds.txt (I’m using Windows) a a go-to. Gargoyle allows cut-and-pasting to the terminal. But it’s neat to see an approach that does more.
The winnable-state tracking of The Witch is an interesting bit of programming to me. It’s obviously nontrivial, and while I prefer not to get in unwinnable states, it’s neat to know. I know there’s been a discussion of how to type out a walkthrough from where you currently are & this seems like a great companion question.
Thanks! You probably realize this already, but just for the sake of being perfectly clear to anyone else reading: the winnability calculation is being performed by The Witch itself, not by my code, which is just trusting and repeating the game’s own reporting on the subject. The really interesting question for me is the bit I originally wrote the frotz Python harness for, which was my desire to try to brute-force All Things Devours by exploring every possible sequence of events to see how big the possibility space is. Thirty-two and a half months of processor time in, having made almost a quarter of a billion moves, it still hasn’t managed to repair the prototype.
Anyway, I’m glad you enjoyed the code! It’s been fun to have a little repo where I dump small, semi-related projects over the last few years.
