you · can't · see · inside · something · unless · it's · broken

Gemslinger has gameplay!

If you want to try it out, here's the deal:

• Clear your browser cache.
• Click on the link above.
• Pick out some clothes if you haven't already.
• Click on the Action tab.
• Move around with the arrow keys.
• Red and/or blue gems will appear near you at regular intervals.
• What gems appear is based on your clothes. Some clothes do not produce gems at all (the first item in each list).
• Move next to the gems to collect them.
• Collected gems appear at the top of the screen, and move to the right.
• When collected gems move a certain amount, they disappear and are wasted.
• Use the spacebar to fire.
  
  • What happens when you fire is based on what gems are in your collection.
    
    • Red gems do damage.
    • Blue gems make the rats go slower.
  • Once you fire, you lose all your collected gems.

Your range is kind of pathetic right now... there will be gems to fix that. ;)

(If you want to restart, you have to reload the page.)

While there are many, many bugs and you're not in any actual danger, the basic mechanics I want are here: gems appear near you; what gems appear is based on what clothes you're wearing; the more of one color you get, the more effective that color is.

So 1 red does 1 point of damage; 2 does 3; 3 does 6, 4 does 10, etc. You want to get lots of the same color in a row, the way gem matching games should be. :)

I think it's basically working. The point is to get you to move around a lot, instead of the stand-and-shoot effect you get in many games. As for having to plan your wardrobe, I can't realy test that without different monsters and non-combat encounters for the player to try it.

Had a mind-bogglingly wonderful date with reasie last night.

We started off with a wonderful dinner at Gusto. I had some yummy steak and Marie had the lobster ravioli, but the best part was the bruschetta we had to start off with... instead of the usual totato and basil, it had, duxelles, seared scallops and these really nifty crunchy fried onions of some sort.

After failing to finish our deserts, we walked around Little Italy, hoping that Christine's (a consignment shop) might be open. Sadly, it wasn't. But we stumbled into the gallery of this random photographer who wanted to take lots of pictures of Marie and her hair, look at my web site and talk about Photoshop effects and Flash programming.

It's always cool when you actually get to interact with a store owner, rather than just looking around and quietly leaving.

So, giddy from that experience, we walked back to our car, stopping to admire the statues and the freshly polished dome on top of the Holy Rosary Church. I don't think I'd ever seen one of those copper domes copper before. It looked magical in the light of the setting sun, with pink and grey clouds behind it.

Right then, it started sprinkling.

And as we rushed to our car, I looked back at the church--

--and the rain was unlike anything I'd ever seen before: tiny flecks of shimmering, golden light that formed a wall with a crisp, well-defined edge, all the little flecks connected, moving in unison, one slowly falling thing as tall as the sky--

--and for a fraction of a second, it was like I'd never seen rain. It was beautiful and it was otherworldly and it was scary.

Really makes you want to write poetry or oil paint or something. But I had to settle for writing beautiful Actionscript instead. It can be lyrical, in its own way :)

Over at gemslinger, you can move around and bump into stuff.

Sometimes the simplest things confuse me. I had a really hard time doing the collision detection here.

No actual game play yet.

There is also a little widget at the top called Spare Change, where you can (eventually) pay to play the game. So yeah... I'm moving forward on the whole this-isn't-free thing.

Maybe that will turn off piles and piles of users, and it's bad from a business point of view. But from an ethical point of view, I'm convinced that a flat pay-per-session model is way less shady than paying for in-game items or sending you over to possibly shady marketing sites, which some apps do.

You can name your currency and define your own exchange rate, which I hope to set to the boring ratio of 1 dollar for 1 whatever, so you know precisely how much you're really spending. No volume discounts, no sales, no specials. Everybody pays the same amount.

I want to keep things simple. And honest. And hopefully, it won't be too much of a barrier to entry. From what I can see, you can get your Spare Change account started with just a $2 balance, which is actually higher than I'd like, but not particularly heinous.

virtuality
was quite good, though I think they were pushing the interpersonal conflict aspect of the whole reality tv thing a bit.

One of my favorite things was something they didn't do.

So we've got this big interstellar spaceship, and in the middle of the show there's this big pretty launch sequence. And there's no sound in outer space, and they're using a propulsion method I've read about in science magazines, and the ring-shaped habitable area of the ship stops spinning and sort of turns inside out...

...and I know that they're doing this whole ring-twisting thing because before this point, their artificial gravity was made by spinning the ring, and now it's going to be made by the acceleration generated by the science-magazine-drive, and gravity is going to go in a different direction... but you know, I don't really care. I like that they didn't waste any of our time explaining this stuff. What little explanation we do get is more about making the guy explaining it look like a total ass than educating the audience. It's about character.

Also, I like that we never get an external, telephoto beauty shot of the ship gracefully cruising through outer space. They just focus on the characters reacting to what's going on inside their ship.

And speaking of tv,

merlin
is definitely a keeper. I love how wholesome it is, I love the corniness of it all. It's like milk and cookies and a little Camelot playset that opens up to reveal little poseable knights and peasants with smiles permanently drawn of their smooth, noseless heads.

Except Uther. He perpetually has a cute frown.

incidentally, I heart

first watch

, a little breakfast-only restaurant really close to our house,  that I expect to go to entirely too often. We've been twice now, and while it's perpetually packed, the wait is totally worth it. The food is really realy really good and tastes real, which is quite refreshing given my all-junk-food diet lately. 

So after yummy breakfast food yesterday, reasie  and I went to

the cleveland museum of art

, the first time we'd been there since it re-opened.

It's looking nice, though there's just not that much on display right now since they're still doing work on it.

I really liked Industrial Section by Jolan Gross-Bettelheim... it totally feels like Cleveland, despite how abstract it is.

I always wanted to do a tile-based game where the cities looke cubist like that, but I'm not sure I can pull it off.

I have started a new font for Gemslinger.

Or, to be more precise, I'm revising an old one.

My main goal, as usual, is to avoid spending hundreds of dollars in licensing fees.

So here are my design goals:

• It must match Penumbra Flare, which I'm using for headlines.
• It must have a monoline, science fantasy look that meshes well with my game art.
• It must be very readable, since I'll be using it for these choose-your-own-adventure style portions of the game.

To get the whole science fantasy look, I'm using a handful of uncial glyphs with a wide, geometric style. While I usually despise fonts populated with random uncials, I think it's working ok here. I'm working very hard to make the uncials not look archaic. So for example, the lowercase R is unconnected, which makes it feel kind of modern and makes the counterspace match the normal lowercase letters.

As for readability... that's not something I know that much about. I think this needs to be spaced wider, and some letters, like the g, feel too compressed.

board games

this wednesday? Let me know! I'd like to get things started a little earlier; it's not like I'm cooking fresh meals anymore, so anytime after like 6:15 is fine with me.

merlin
was pretty corny, but I think I'll give it a shot. While the first few scenes were pretty excruciating, it quickly settled into that awkward, stilted but eager rhythm you get with new shows. I kind of like that awkwardness. It's cute. You want to pet the new show and tell it that when it grows up, it will be a good show and a pretty show.

And I think John Hurt makes a great dragon.

I defeated a shadowlord
on my lunchbreak.

So yeah, I've picked up Ultima V again, and luckily, I still had my old save games. So this is farther than I've ever gotten.

Despite the antiquated graphics, despite being turn-based, it's been a pretty tense affair. It's exciting in a way that modern games just aren't to me.

A lot of that excitement comes from not knowing what the heck is going on. You don't have a status screen showing you how many mantras you've learned or how many shadowlords you've dispatched. You don't have a to-do list. You don't have a built-in map or a note taking system.

You really have to remember all this crap. And while I find myself wishing, constantly, that there were in-game aids to help you along, their absence probably makes the game better.

I don't think that these sorts play aids are necessarily bad. I just think that, in modern games, they're presented in such a way that they suck all the drama out of what's happening. Most to-do lists in games are just that -- lists. They're about as exciting as a PowerPoint slide, and about as forgettable. Once you've completed something on your list, it's gone. Forever. There's no sense of position, of movement within the game as a whole.

If you're going to have a game screen that tells you what quests you're on, you need to put those quests in context. Some sort of memorable, visual context. The original Legend of Zelda had a little graphic of the Triforce, and as you collected little triangular pieces of it, the graphic would get filled in, one triangle at a time. It wasn't much, but it was memorable. It was tangible.

So I'd like some sort of visually rich to-do system in Gemslinger. I don't know what the quests are going to be yet, however, so I don't realy have any specifics right now. :/

So as I was chowing down on some McLunch yesterday, I thought it might be a nifty idea to have Gemslinger combine two popular activities on Facebook: taking silly personal quizzes about yourself, and ,of course, playing silly RPGs.

Ultima V started off with a sort of personality test, where the game figured out how you ranked various virtues like sacrifice, honesty or humility.

Each question had you weigh one virtue against another. For example:

Entrusted to deliver an uncounted purse of gold, thou dost meet a poor beggar. Dost thou:

A. Give the beggar a coin, knowing it won't be missed.
B. Deliver the gold, knowing the trust in thee was well-placed.

Here, A is the compassionate choice, while B is the honest one.

Your character was then made with different qualities, based on your rankings.

I'd always wanted to see this sort of thing integrated into the actual gameplay somehow, and I think I can do it here.

What I need to do is get a set of virtues, or personality traits, that are easily measurable by the game. It would be difficult for the game to detect that you're doing something compassionate, or something humble.

What I can detect, however, is what quests you chose to go on. Therefore, I propose a virtue system based on having quests come from different people.

* * *

friendship (requests from friends)

Being a social game, your friends will be able to ask favors of you, and add things you your quest list.

integrity (requests from yourself)

Being a giant marketing tool, I want you to keep playing the game as long as possible. So I want you to keep setting explicitly defined goals for yourself that the game keeps track of. Setting them will be mandatory, but completeing them is not.

loyalty (requests from NPCs in a position of authority over you)

The main quest arc would probably consist of taking orders from some king or queen or whatnot.

charity (requests from random NPCs you don't even know)

On occasion, you would get a random request from someone you'll never encounter again, that has no direct benefit to you whatsoever.

* * *

So hopefully, I can arrange it so that you have to choose between these things.

For example, you might set a goal for yourself that you want to get these nice Pants of Freezing. But your lord, the next day orders you to find some Freezing items for him. You have to choose between your own personal Integity and your Loyalty to your Lord.

You'd end up with a ranking of these virtues based on how you play the game, so NPCs would react to you differently based on your rankings.

It's a long way off, but I think this could work.

I don't think it will really say anything about you as a person, but then again, I don't think regular Facebook personality tests say that much either. It's just some silly trivia you can share with your friends.

Back to pet-project silliness.

I have started on a map for Gemslinger. (This isn't an illustration so much as a schematic right now.)

If you've ever heard me blab about world building, you know I see it as a complete waste of time. If this were a story-based game, I'd only  fill in what details were required to flesh out the characters and move the plot along.

I actually started out trying to come up with some random continental & oceanic plates, deciding what directions they were moving, figuring out where subduction zones and hotspots were, making island arcs on some plate boundaries... but decided that that approach didn't really serve my goals well.

Since this isn't a story-driven thing, my main goal is just to have a place with memorable shapes, a variety of terrains, and some repeated elements where you can have long, multipart adventures. Like the island chains in the upper left; I need more stuff like that.

It should be easy to see and remember where you are, and where you can go. And I want a long, skinny place where you can start out and go on a fairy linear series of tutorial adventures before getting comfortable with the mechanics of the game and graduating to a big, fat place with more freedom of movement.

So you'll start on the bottom left and work your way up and right in the beginning.

And that's it. Those are the needs of the world.

on a more technical note,

I'm editing this as a 200x200 bitmap, and that will serve as my data for building the actual map out of tiles at runtime. It's really convenient because I don't have to make data structures for all this; I can just use Photoshop's drawing tools to make my map, use the builtin bitmap-loading stuff in Flex, and then map different colors to different tiles. Really simple.

 

For those of you who read gwacie  or reasie 's journals, this will mostly be a repeat.

Our oldest cat, Eenie, died this morning.

There might have been heat exhaustion or something involved, but she was just really, really old, like 21 or so.

I'm not sure what else to say. Starting yesterday afternoon, there was lots of running around and worrying and thinking she was getting better and having to decide that it was best to let her go... but in the end, she just died naturally anyway.

I'm generally fairly oblivious to pets, but I will miss her, and her smooshed, absurdly cute, I-hate-everything face.

first off,

board games

This Wednesday if people are interested. I just got Galaxy's Edge, a new thing by the people who made Conquest of the Fallen Lands (Which means it's nice and simple). Eager to try it out.

california

I am in california now, for reasie 's cousin's wedding.

Today I have eaten:

• Pancakes from McDonalds.
• A little round turkey sandwich on our first flight, plus Ruffles and a Kit Kat.
• Some yummy potato skin appetizers topped with bacon, onions, cheese and andouille.
• A salad, and a bite of reasie's little round cheeseburger from our second flight.
• A couple of potstickers.
• Some sort of Chinese duck-filled burrito.
• A couple bites of the best sweet & sour pork ever.
• A bowl of wonton soup.
• A bite of some really great honey walnut shrimp.
• Lots and lots of Coke.

I quite like little, round airplane food.

Surprisingly, I do not feel particularly full. I'm confused.

I am also very tired.

Some rough characters.

Hopefully, the one on the left looks boyish and the other looks girlish, though the only difference is the hair.

genders

I'd like to go with totally unisex outfits here. Because I hate the cheesecakey gender dimorphism in most fantasy games. And so I won't have to draw everything twice.

Adventure Quest actually has totally unisex outfits. It's the only game I've seen that does that, and I think it works fine.

For NPCs, I could make the genders more obvious and have different body types. Adventure Quest also does that, but I kinda don't like it. I think I'd want to stay with the same bodies -- in different poses -- to emphasize the non-importance of the body shapes. If there's one big, round npc, people will wonder why you're forced to have skinny avatars. If everyone looks the same, you're more likely to assume that it's just a random stylistic choice regarding human beings in general, the way my rats have no legs.

no legs

The long robe is just there to fill in the square.

So I figure everybody PC will have a cape or robe or something, um, because you're all wizards. Yeah. And that's just what they wear.

You'll probably never see anybody's feet, which is fine. I want a very limited number of clothing 'slots' anyway. So here I've got a tunic, robe, hat and gloves/arms, and maybe that staff if I want a fifth slot. But four or even three would probably be ok. I could ditch the hat or arms slot.

negative space

I like all that negative space on the upper left, but maybe I could fil that in with a health bar, and do that on the monsters too.

I'm not sure. I like negative space.

Saw The Terminator for the first time last night.

It felt awfully dated, especially the sound mixing. But I liked it. If anything, the lack of eye-popping cgi effects made the action more visceral, more structured and easier to follow than many modern action movies. And I liked the total lack of dialogue in many of the scenes.

I even liked the stop-motion robot at the end. I always found stop-motion monsters to be really creepy, creepier than they would be if they were better realized.

It really is just a monster movie, isn't it? All that sci-fi backstory really seemed to be about making the monster movie cliches feel a little less corny.

Forgetting all the sequels and spinoffs, I found myself wondering why they couldn't just have Sarah be the future leader of the resistance and drop all the stuff about her unborn child coming along to save the day. I mean, it makes her a really passive character here, just being there to be the off-screen hero's doting parent at some point in the future. And I think the whole shtick in time-travel movies about going back in time creating the need or ability to go back in time is pretty overdone.

It all felt kind of weird from a narrative standpoint, having this John Connor person never actually show up.

Or would Sarah leading the resistance have been too progressive for 1984?

Though I like the little bit at the end about Sarah waffling on whether or not to tell John who his father is.  There was kind of a personal spin there on the paradoxical nature of that whole time-travel thing, and I thought that was a really interesting, tender moment.

Anyway.

I still actually like Salvation too.

Have a square rat & bat.

Facebook is apparently looking into letting developers add micropayment support to their apps. Sounds intriguing, and I would like Gemslinger to pay for itself using such a system, if possible.

I said before that I'd never make people pay for items in my Facebook game(s), and I'll still stand by that. But I don't object to having people pay for the game in general. I'm not real comfortable with the ethics of not charging for a product. It seems dishonest.

I find myself thinking about amusement parks and restaurants here.

As I understand it, most amusement parks these days just charge you a flat fee to enter the park, and you can ride all the rides you want. But a few amusement parks (or, more frequently, the traveling things at street fairs) make you buy tickets for the rides, and the more you ride, the more you pay.

Conversely, in most restaurants, you pay for each item separately, but at a buffet or a mongolian barbecue or somesuch, you just pay a flat fee.

I'm not real sure why there are different standards here, and I'm not going to argue that the flat-fee model is fair and the a-la-carte model is not.

I'm not even going to argue that the flat fee model encourages people to have more fun or relax more because people are free to forget about cost once the fee has been paid. While I love going into an all-you-can eat buffet and getting tiny samples of a bunch of items that, added together, equal a still tiny amount of food, another person might stress out about getting their "money's worth" out of the experience.

What I do have a problem with is going into this sort of thing with friends, and having different experiences because each of you paid different amounts, even though you all went together.

If I go out with friends to a fancy restaurant, get scared by the prices and feel like I have to nibble on a cheap appetizer while my companions eat steak, I'm certainly not going to blame the restaurant. It's more a question of poor planning; we should have gone somewhere where everyone would be comfortable.

But, to eliminate the need for that sort of planning at all, I'd feel more comfortable with the flat fee system in my own products.

So the question then becomes, how often would I charge, and how much?

You could do it like an arcade, where you pay a few cents every time you log in or travel to a new area, a truly micropaymenty system. But I think I'd prefer something a little less granular, like a quarter for a weekly pass. That way, it's a large enough amount of time that you can play when you want to play, and not worry about having to run off and do other things and lose your quarter.

I don't know what typical prices for these sorts of things would be, though. I'll have to look into that.

So here's what I mean when I say I want the bounding boxes in Gemslinger to be absurdly clear:

I should be able to make the tail wag and the head bob pretty easily, to make it feel kind of lifelike in-game... but it should be pretty clear this is just an abstraction of something, and not really representational.

So I tried to model a 3d rat over the weekend, for Gemslinger.

I didn't get much done, and it did not go exceptionally well. I'm not good at visualizing things in 3d.

I think I shall go for a very abstract look. More abstract than anything I've seen, in fact. Partly because I don't think I can do the 3d character animation right now, but also, because I don't think I actually want 3d rendered characters for this project.

Been playing Star Wars: Empire at War, an RTS. I'm finding the visuals a little frustrating. Sure, they're attractive enough. It looks like there's a miniature Star War being waged on my computer screen. It all looks very authentic.

But it doesn't feel like Star Wars. It just feels frustrating a lot of the time. I can't tell my blaster-weilding grunts from my blow-up-giant-robot specialists; they're both collections of people a dozen pixels high each.

And I'm having trouble targeting enemy units, since I'm not exactly sure where their bounding boxes are.

Maybe if the graphics were more abstract, I wouldn't be having so much trouble with those fast-moving, hard-to-click-on tanks squishing all my people. And maybe it would be a little more fun.

My goal with Gemslinger is just to make to most fun rpg I can. I'm not trying to tell a great story or flesh out a great world. I don't want you agonizing over where to spend your character points every time you level up. I just want something fun.

And part of being fun is being clear.

I'm going to encounter some resistance, of course. People aren't used to abstract visuals in computer games.

But nobody looks at an in-progress board game and expects the combination of full-color maps, cardboard counters and paper or plastic characters to look like a scene from a movie. They expect it to present the mechanics of the game clearly.

I'm hoping I can sell that idea here. A sort of realtime board game.

So I'm already thinking about my Jennifer Ann 2010 submission, and my rewrite of my 2009 submission, which I think of as two separate projects. I want the first to be a head-on, modern-day tackling of a real-world issue, and the second is an interactive story.

I don't like tackling social issues head on. I don't think it's necessarily effective, especially for a teenage audience. But I want to give that a try.

And, of course, there's Mika's Tavern.

I would like Mika's Tavern done late Augustish. Or, at the very least, late October, because I want to submit it to the Independent Games Festival and I think submissions are due in November.

The bar for the festival is really really really high. Placing in this year's Jennifer Ann thing was cool and all, but I'm going to have to do a whole lot more work, and manage things a whole lot better to even get noticed at something like the IGF. But I figure I might as well give it a shot, and give myself another deadline, another challenge and potential prize to work towards.

I'll probably be poking at the Facebook thing too; I don't think that will really interfere much with this, because of the whole doing-it-on-my-lunchbreaks thing.

I am perfectly content at this point to let Basil Street Bridge fester. Sure, I like the story and the aesthetics of the game world, but there's just not much meat from a gameplay standpoint. Many of the story elements may get absorbed into Mika's Tavern; they fit pretty well. If I'm borrowing so many Basil Street characters and plot points that it obviates the need to actually do Basil Street, that's fine. I can just work out some mechanics and get a fresh story for that world that fits there.

Or not. The 1920s sci-fi world in my head won't mind.

Well, that was kind of annoying.

I finally got Symphonic Choirs installed, and have a couple problems with the word-building interface.

It doesn't work with my music editing software, but I can still run the choirs app as a standalone thing. Presumably, I'll be able to record vocals and import them as raw sound into MixCraft. That's annoying, but at least I can still use it... sorta.

The second problem is that I just don't have the memory needed to run it. When trying to do actual words, it wanted more than half a gig of ram, and was crashing because I didn't have that much to spare. I might be able to use some less robust versions of the libraries that take up less memory, where you can't use the mod wheel and stuff like that. Trim down the features a bit.

So if I want to use this in MixCraft, I can't use the word-typing thing; I just get vowel sounds instead. Which isn't so bad; they're very nice sounding vowels.

* * *

Also, I got a bundle of random Star Wars video games since I figured I needed some structured downtime in my schedule. But, Knights of the Old Republic doesn't run on Vista at all, and while Empire at War seems to run ok, there's no opening crawl at the beginning of missions... just the Star Wars wordmark running away from you, dramatic music... and an empty starfield.

Ugh.

Well, I got a runner-up prize for the Jennifer Ann entry.

Since I only heard about the contest a month and a half after it was announced, I guess that's not too bad... but really, I'm pretty mortified that this is up there where the public can see it.

While I keep thinking of it as nearly unplayable due to the whole spacebar issue, I've actually been afraid to look at it since submitting it. I don't know how playable it really is. (Which, I suppose, might make my postmortem a bit premature....)

Anyway, If you have a look, I'd love to know what you think.... if you love it great; if you find it hugely problematic, please say so...

Well, they were supposed to announce winners for the Jennifer Ann contest yesterday, but I haven't heard anything and there are no updates on the web site about it

But here's my postmortem anyway.

what went wrong

• character animation. I think that, more than anything else, the character animation is what sunk this project, just because it took so long. I should have been spending my time on level design, playtesting, and debugging instead. In theory, doing character animation using Illustrator vector art and Photoshop layer styles sounded like a pretty cool idea. I could make just a few keyframes' worth of animation and use Illustrator's blend tools to do the inbetweening for me. But nothing quite worked the way it should have, and I probably would have been better off learning to use one of the 3d character animation packages I already own. Here are some details:
  • blending. Mostly, the blending worked. But sometimes, with rotating objects, the blending would be completely wrong. Say you wanted someone's upper arm to rotate clockwise 5 degrees. Instead, the blended frames would show the arm rotating counterclockwise 355 degrees. As far as I could tell, there was no way to change this behavior.
  • layering. I had to use dozens of layers in Illustrator, one for each body part. Importing new vector artwork into Photoshop while retaining the Photoshop styles applied to each layer was not possible without using scripts. So I wrote a Photoshop plug-in using PatchPanel to automate the process. When it worked, it worked well, but it would frequently stop in the middle of my script for no reason that I could tell, so I ended up merging all my layers manually anyway.
  • gigantic files. My photoshop file for the heroine is 8000 pixels wide, 400 high, and 30 layers deep. Photoshop didn't like that. I was frequently running out of memory.
• dynamic music. There was supposed to be this system where the music being played would dynamically change its arrangement based on the current player position, scare level and flashback state. Also, once a phrase was complete, it was supposed to pick an appropriate phrase to go with next. I was only able to implement position detection, and I didn't get to the phrasing at all; it basically moved from phrase to phrase randomly, which sounded very disjointed. It probably would have sounded better if I had just had looping music with the variable arrangement thing.
• level design. While I'm reasonably happy with the levels I got, I didn't allocate nearly enought time to this process, and only got half the levels in that I wanted (though a big part of that was also not wanting to make new characters for new levels). Mostly, the problem here was failing to settle on my mechanics until very late in the process.
  
• debugging. My debugger crashed as often as my program. Not much more I can say about that.
• word balloons. It was really hard to tell who was talking. I didn't have time to add little lines going from the balloons to people's heads.
  
• tutorial. Yeah. I forgot to mention how you talk to people. Big, bug problem there.

what went right

• scripting. While my XML scripts could sometimes be a pain to edit, going with XML probably saved me a lot of time, since I didn't need to write an editor. This also made it blindingly easy to save and load the program state when you needed to go back to a safe save point.
• music. I'm not happy with the way the music sounds in game, but I am immensely happy with the music on its own. There's only a couple minutes of material here, but I didn't think I'd be able to get even that done by the time this was over.
• graphics. I love the look of this. It's unique, it's reasonably polished, and goes with the theme well.
• flashbacks. The initial concept seemed to work, and I think that what gameplay I have meshes with the story well.
• layer comps. The heroine has three different costume changes, but there's only one photoshop file. I was able to use layer comps to set different styles for different pieces of clothing, or hide articles altogether, for each scene. I've never done that before, but that sped things up greatly.
• having a totally one-dimensional interface. I wasn't real sure that was going to work. You can't jump, you can't duck, you can't climb ladders. Your movement is completely, unquestionably one-dimensional, which is unlike anything else I can think of. And yet, I think there's a reasonable amount of gameplay you can get there.
• keeping the code manageable. While this was written very quickly, I never allowed myself to do things in a fast, unmaintainable way that I'd regret later. As far as I can tell, the code is pretty solid.

conclusions

• I can't emphasize this enough: I really need to learn character animation, in Strata or Daz or both. Any project I work on is going to need this, and I just can't do it quickly right now.
• I am very happy with what I learned during the project, about every aspect of game design.
• I got very, very sloppy near the end. There's no other way to put it. I didn't manage my time real well, and ended up being absurdly rushed my last day. I must not do this again.
• I would very much like to revisit this, which would probably involve a complete rewrite, since this was built to be excessively limited in scope. I'd still want to keep it very simple, though.