It was the best of times, it was the worst of times. If I go back to about 1993-94, when I was learning C++, already knew some Pascal, C, and ASM, just started jr. college, I really wanted to write a tile-based game like Zelda, and was willing to re-invent lots of wheels to do it. That's when I figured it would make the most sense to use some kind of data file to describe a game rather than defining it in source code. (This was always pretty obvious to everyone.) There were ways to describe doors, enemies, persistent variables and counters, dialogs, etc. But it was all data, and no code. Next I realized that was stupid, and with the same amount of work, or less, I could just write a program that could be interpreted as commands rather than just data definitions. That's when scripts and objects came in. Having no idea how to write an actual compiler, I made up something with if-statements, loops, and variables, and lots of symbols. If I remember, it looked something like this:

object fred
[
100 -> x
10 -> y
]
script myscript
[
-> var1
-> var2
y + 1 -> var3
somefunction -> var1 + 10
start
[
fred:
loop
[
delay 1
dosomething
]
]
]

It was a very simple but effective language, but it would have been harder for other programmers to pick up, because it was so unusual. I can't even remember all the peculiarities, but the core ideas are there, as you can see. That was written for DOS using Watcom 32-bit mode and DOS 4GW, with some of the utilities written is 16-bit Pascal. I called it Build and Engine. Around that time, everything was going to Windows 95 and DirectX, so I decided to re-write more concisely using everything I had learned. This system I called AGHAST. Around 1996 perhaps, I started going to CSULB and took an actual compiler class. Whereas jr. college was a lot of hands on stuff that encouraged me to play around, this instruction was more theoretical. There was already a well-established science behind compilers that took me away from my own cute but incomprehensible language to something that looked a lot more like C. So I threw away pretty much everything and started from scratch on AGAST.

The name didn't take me long. I already had AGHAST which took me a good 5 minutes to come up with. It had to be a term which didn't appear all over Altavista, came up early in alphabet order, and was easy to type and remember. It probably took me twice as long to remove the H.

Then one day I read a usenet post about what was to become what was to become SCRAMM, and I thought I'd better put mine up on the web so that anyone else making such systems would pick up some ideas (like start and delay!). Then, even if I didn't finish, someone else would make a system I could use. But people read it, and most couldn't tell the difference between AGAST and whatever else was out there. There were maybe about 5 people who wrote me and seemed to really say "wow, I see what you're doing, and this is something!"

One of them, of course, was Russell, and he even offered to help. I never took these offers too seriously, because most people didn't realize the amount of work it really takes. I'm talking like 15 hours a day, sometimes, for months at a time minus school (except I somehow found more time to work on it during finals). But Russell seemed serious enough, and so I asked what he's good at, and he said "graphics" along with some other things. But "graphics" is what I mainly remembered, maybe because that's where AGAST was really slow. My routines had been place holders for what should have been much faster ones. They were good enough for something like Monkey Island, which was my main criteria for AGHAST and AGAST, but faster graphics meant much more could be done. Much to my amazement, Russell came through and made the graphics routines much faster in a short amount of time, has done hundreds of improvements in probably every area (language, compiler, sound/music, etc.), and then redid the entire subsystem in Direct3D acceleration (Morningstar). I haven't done much in the last couple years except cheer him on! I don't know how he does it, because when I was working, the last thing I could stand to do was come home and programming. I wanted stuff done, but just didn't have the energy to do it. This has led me to the theory that Russ is actually an android that was sent from the future.

And of course lots of people have helped immensely in lots of other ways, like with programming, and utilities, and graphics, and I hate to gloss over all that. But I think everyone around here knows about the more recent stuff.

—Todd Zankich, AGAST developer

Todd got most of it.

I was looking at building a generic adventure game toolkit for (at first) my own use. I went looking around at what other people were doing. Back then, a lot of people still believed SCRAMM was going to come out, and BOT would still be in active development for a while. (At least, I think it was. I could be getting things wrong within the range of a couple of months.)

Anyhow, I stumbled over the AGAST home page again, and took a serious look at the system for the first time. What I found was that it did a lot of the things I wanted, the way I wanted to do them. It also had elegant solutions for a number of things I hadn't devised elegant solutions for. I wanted to contribute something, though.

So, I posted on the board and e-mailed Todd, and he asked if I could write some graphics routines to spec. I said "sure," and whipped some up. At the time, I spent a lot of my free time working on optimized software rendering routines, so it didn't take too long. The improved graphics engine made its debut in AGAST 1.06.

After that, things just flowed. Darris appeared on the board working on Passage (the art for which was mostly done when he found AGAST). A while later Jonathan Lessard showed up with a really pretty demo game, and made vague noises about a commercial project. Roger (ad7) released the first complete AGAST game (I think), JBR. There was a snow demo (Roger), and a rain demo (me), and a strange little game called Secret Agent X. At some point in there, I saw a DOS version of AGS.

There was a new board. Passage got released, as did Ozzie and the Quantum Playwright. The Hunter: Part 1 appeared. SLUDGE came out. AGAST 1.08 added classes, faster graphics, and a crapload of other new features.

The new board got broken, so we got another new board. For a while I kept getting told AGAST was pointless, but Glumol would be better. AfterShocked! came out. So did Eye of the Kraken. I quietly wrote the new graphics system at the heart of Morningstar. (I actually did that very quickly, but the release it was in took a while because it involved more testing than any previous release.) I had the good fortune to see an AGAST demo in which an adventure industry professional was involved.

We got another new board. Morningstar came out. You probably remember stuff since then. :o)

I may or may not be an android from the future. Either way, maintaining AGAST takes a lot of energy and time I don't have. It's a wonder I manage it, and I couldn't without all of Todd's help and support.

—Russell Bailey, AGAST developer

—Russell Bailey, AGAST developer