Company History As Recollected In 2021 - Part 3
Table of Contents
This history is talking about 12 years, way more than 12 games, and I’m not even sure how many products. I’m keeping it brief, but there’s a lot to cover, even so.
In part two, we talked about another two years — 2011 and 2012 — one major game release, and two product releases in general. It was a stressful but a profitable time, despite some of the errors I made out of anxiety (and to some extent inexperience).
Let’s talk about 2013 and 2014! The former was the year of our most releases ever, and the latter was the year of our most profit ever (so far).
A Valley Without Wind 2
That said, commercially this was always going to be hobbled from day one. It was free to everyone who had ever bought Valley 1, and anyone who bought either game in the future would get both. Whatever commercial prospects this had was severely blunted.
I also made some pretty strange (in retrospect) decisions when it came to handling this sequel. This wasn’t just an upgrade of the first game in an art sense or any other sense — it was a complete reimagining. The first game was very clearly a Metroidvania, and it also had a lot of influences from Actraiser, and had citybuilding, etc.
The sequel plays a lot more like Contra, is in no way a Metroidvania, ditches the citybuilding for a really cool strategic layer on the overworld (hence part of why Tom Chick loved it), and so on. In other words, the broad genre kind of stayed the same, but the subgenres all completely shifted around.
I had my reasons for doing this, and I actually really love both games… but when it comes to making a sequel, you have to consider not just what makes a good game, but what makes a good sequel. I think that if this game had been sold under a different name, and actually sold, then it would have been much more popular.
As it was, a lot of what people loved about Valley 1 was not there in Valley 2, most notably the sense of exploration. The combat was a lot more engaging and kinetic, and the game flow was a lot more conductive to controller support. The overworld map game was more engaging to some people compared to the citybuilding of the first game, and less engaging to others. They’re just different games, with different pros and cons.
Since Steam has stats for playtime, and since everyone gets both games whenever they get one of them, it’s pretty easy to compare which games engaged people more, though: Valley 1 gets more playtime by about a 5:1 ratio. The people have spoken, they preferred the original.
Extra interesting were the comments on the art. The art in Valley 2 is gorgeous, but immediately you then have people coming out and saying how they preferred my art in the first game. How it had more character, more of a bleak feel, etc, etc. In other words: the art in the first game really and truly was doing what it needed to, and the whole enterprise of the “free art upgrade” that turned into a sequel was a fool’s errand to begin with.
Valley 2 was the first title that we ever released into Early Access, and Early Access was still a pretty new concept back then. Since this was a free sequel, it made sense, and my feelings on that whole process are pretty neutral. It wasn’t the typical EA project path; the whole thing only lasted maybe four or five months.
Valley 2 cost something like $80k to make in all, although it’s hard to be sure exactly how to allocate that. I strongly doubt that it generated $80k in revenue that would not have been generated by Valley 1 on its own, but the pair of projects remained profitable, so in the end it’s not a crisis. It was very foolish on my part, but hindsight is 20/20.
Alden Ridge was the name of the game, originally, when I was working on it in 2008. I wanted my own version of Silent Hill in terms of a name (Silent Hill 2 is one of my favorite games of all time, right behind FFVI and Chrono Trigger), but it was generally agreed that this name sounded like a British soap opera. After a lot of tossing names around, Shattered Haven was the new name.
As I mentioned in part two, I brought on my old friend and modder extraordinaire Zack Cataldo to help with level design. I had already designed something like 50 levels for the game, and Marisa had done maybe 10 as well, but I needed more overworld content and I also wanted to push the level count to over 100.
This game was… a colossal failure. It’s “Overwhelmingly Negative” reviewed on Steam. I have some theories on why this is, but a lot of it comes down to things that aren’t inherent to the game itself. Let’s take a look at the factors I think are most relevant:
- From a trailer and marketing standpoint, I focused a lot on story rather than gameplay (because I didn’t want to spoil puzzles), and so people were very confused as to what this game was. This is an action puzzler game like Lode Runner, but Rock Paper Shotgun called it a turn based tactics game. You should have seen my face. They weren’t the only ones confused.
- Tonally I was trying for something very serious with the story. Something along the lines of Silent Hill Lite. But it was… pixel art. The visuals just did not have the gravitas to carry that kind of story. There have been a few people (I think three) who have told me that this is their favorite game ever, and it’s because of the story. I was really touched by that, and also surprised given how widely reviled it is. It was mainly a matter of the pieces not matching.
- The early puzzles are too easy. I was worried that people would get stuck, and so I made them overly simplistic. And therefore a lot of people would try the early stuff, get bored, and then bounce out of the game. Some of the later puzzles are truly devious and fun, but you get none of that in the very early game.
- The game just isn’t very pretty in general, and the overall design style of it makes it kind of impossible to make it all that pretty. It’s retro, but not quite retro enough to hit that nostalgia vibe.
- I think that Lode Runner: The Legend Returns was fairly niche to begin with, so it’s not like a lot of people would really be nostalgic for that sort of gameplay, if they even recognized it when it is flipped from side view to top-down.
This game was a massive commercial failure, and it was a critical failure, and most players hated it. I’m really proud of the fact that at least a few humans in the world list this as their favorite game ever despite all that, though.
You would think that my anxiety would have gone through the roof with this project bombing, but honestly we had so many irons in the fire in 2013 that I just kind of shrugged it off. I was starting to develop a bit of thicker skin. Not that I didn’t have some bad moments, I’m sure, but it didn’t alter my behavior like the reaction to Valley 1 had.
Side note: Pablo’s intro music for this game, which is in the trailer below, is one of my favorite music tracks of all time. He just absolutely nailed “Akira Yamaoka, but Pablo Vega style.”
Fun fact: this came out in March 2013, less than a month after Valley 2.
Reconsidering Development Strategies
Keith and I had a lot of discussion about the recent failures of several products in a row, and both agreed that we were not fond of giant projects that were so risky if they failed. We decided to split our efforts again and work on two separate projects. He would helm his first completely solo standalone project, Exodus of the Machine, and I would work on a solo project called Skyward Collapse. After those were done, we’d do more small projects, and keep going like that for a while.
This is very similar to my mindset going into 2022, incidentally. But more on that later.
Exodus Of The Machine
This game never saw the light of day, for a lot of reasons that I don’t remember super clearly. The really short version of it is that I failed in my job as a project manager and producer (which was the extent of my role on this project).
I threw Keith into the deep end of development on a new IP (set in the AI War universe, but still) with less preparation than he was expecting. I also had artists queued up and ready to work on this project, and asked him for a worklist for them way before he was expecting it. A lot of that comes down to communication failure on my part, and also forgetting that not everyone is redlining extra overtime like I was (and nor did I want them to).
A lot of art got done for this, and a beautiful logo as well, and a fair bit of code and design, but we pulled the plug when it was clear it wasn’t going to be all that fun, and was going to be a money sink. And/or some other details that I honestly just don’t remember. But anyhow, this sort of thing is the responsibility of the producer, and in this case I (as producer) put Keith (as the lead designer) in a pretty impossible position. This was only the second time I had acted as producer without also being the lead designer (the first time being Tidalis), so I was still relatively new to the whole idea, but so it goes.
Anyhow, this was a great example of us realizing pretty quickly that things were not going as planned and then just scrapping the project rather than throwing good money after bad. This was the first time we had outright abandoned a project during the prototyping phase, but it would not be the last. I learned to put less money and effort into art and logos and such until a prototype was actually fun.
Oh, right. The premise of this game! Essentially this was going to be a certain version of the AI’s path to sentience, I think. Something along those lines. It doesn’t match with current Arcenverse lore, but that doesn’t really matter now. This is something that I would describe as being kind of a cross of Oregon Trail and a visual novel. Very different from anything else Arcen has tried before, but that spirit of experimentation is something that’s at the core of how I like to do things.
I don’t recall when this was scrapped, but it was sometime before the summer of 2013.
This came out in May 2013, after a development period of only maybe three months. That was blazing fast. Daniette “Blue” Shinkle had just been hired as our new art director, and she set about finding some contract artists to help with really pushing through the art quickly.
On the design side, I did the overall design myself, and then Josh Knapp and I did the item design and research together. He had a huge amount of knowledge on Norse mythology, and together we translated that into game mechanics for the various gods, artifacts, myths, and creatures that that culture had. I was more familiar with Greek mythology as it was, but Josh was really critical for research in that area, too, which saved me a ton of time.
The premise of Skyward Collapse was to make a “god game” in general, where you are playing as kind of an overarching god bossing around smaller gods (well, if Zeus and Odin are “smaller”). But the real twist with this game was that I wanted to make a game that was single-player but which would encourage the player to engage in brinkmanship with themselves.
The song used in the above trailer is sung by Pablo, and my goodness does he knock it out of the park. So often are people asking for his wife Hunter to sing, but they’re both just insanely talented.
The timing of this release was really fortuitous, and also kind of surreal. There had not been many god games around for a decade prior to this one, and so it was easy to get attention for it. But then immediately after this one came out, a bunch of other god games also released (they were already inwork before I even started this one, but I had no way of knowing that and also would not have cared). The result was that this did really well for a while, and then as more god games came out, it faded from any sort of prominence or memory at all.
There was some quote, I think it might have been a tweet, where a journalist said it was “like playing Chess against yourself, and you’re always a huge asshole.” Which I found hilarious and apt. I think it was Brenna Hillier for VG247, but that quote isn’t in her actual review. The game encourages you to play hard against yourself, playing both sides, and thus… make life progressively harder because of your own ambition rather than any third party enemy. I really love the concept there, and I’ve never seen it elsewhere.
I was particularly pleased with reviews that were talking about how this was the most interesting take on the genre since 1989’s Populous, which was the first game of the genre. The creator of the genre was coming out with his own new take on the genre later that same year, and VentureBeat predicted that he would have a tough act to follow now that Skyward Collapse existed. That’s not exactly how things played out, but it sure did feel good to hear at the time.
This game earned back its development costs in the first month, which was a new record for us for sure. A big factor in that is just how short the development time was, and how cheap it was to make. Since it was selling well, we made a DLC for it called Nihon no Mura (“Japanese Village”), which introduced Japan as a third faction that played very differently, and added a new optional victory path that involved building villages of a different sort with any faction. This came out in August of 2013.
The DLC didn’t sell all that well, and once Reus came out from another developer, interest in Skyward Collapse faded away in general. The margins were low, and people seemed plenty satisfied with what was there for what it was, so it was time to move on to something else.
I have to say, I actually really like games like this, even if they don’t have a major place in the memory of gamers. It was inexpensive and experimental, some people got a lot of enjoyment out of it, and there was no expectation of long-term support or development from me. Everyone had a good time, there were no strings attached, and it was just an exercise in being creative and exploring the boundaries of game design.
Vengeance of The Machine
AI War was proving to be way more evergreen than I ever could have imagined. Thanks to steady DLC and free features and ongoing support, it was earning a steady (or rising) amount every year for the fourth year in a row. When Exodus didn’t pan out, Keith and I both agreed that it made way more sense for his solo project to instead be another AI War expansion.
Vengeance of The Machine was focused less on anything super high-concept, and instead just had a lot of interesting new content that could be combined with existing content to make things harder or more interesting. It was quite popular. We didn’t know it at the time, but this was the next-to-last expansion for AI War.
We had been releasing new collected-editions with each new DLC, and this one was called the Four Year Anniversary Edition. I don’t recall the details of what each other one was called, but this one was particularly profitable, as the Alien Edition had been.
Our sixth and final product release for 2013 was this turn-based tactics game, in October. Keith and I split the design duties this time in a way we never had before, and that caused some interpersonal static if I recall. We worked it out like adults and decided to be more clear with what role each of us would have in future projects, and we never had that problem again. But it was a learning experience on how to co-design without frustrating your co-designer.
Tom Chick tells me that the name for this game is absolutely horrible, and that’s one of the reasons it didn’t sell as well as it should have. I think he has a point, when it comes to the clickthrough rates that we see on Steam; the metrics do back him up to an extent. Essentially, his point was that the name just sounds really negative, because who wants to owe dues, etc. His point is more nuanced than that, but that’s the basic idea.
I think that art was another thing that really held this one back. In order to make the levels readable, a lot of the art (which had originally been done in a very colorful fashion) had to be dimmed way down. This made the screenshots look extremely dark and kind of uninteresting. The game doesn’t actually feel quite so dark when you’re playing it, and this game keeps coming up as a cult classic that people wish I would make a sequel to or remake in general. But the wider market was kind of ambivalent.
Speaking of vocals in trailers, this is one where Hunter Vega is absolutely slaying it. We don’t usually do anime-style vocals, but for this game that was the goal.
This one still hasn’t earned back the money we spent to make it, even though it was a compressed time schedule; we put a lot of artist and programmer and design and music hours into it in a short period of time. And it just really went very unnoticed by the market as a whole.
Blue was of course working on this one, and then Cath Langwagen was the other main artist on this one. The two of them were often the core of our art team for quite a while after that. Cath had also worked on Valley 2 and a bit of Skyward Collapse, but she and her partner were traveling the world at that time and so she only had so many hours available until this project.
This game had the most voice acting in it out of any of our titles up until now, and that was also fun (and not that expensive). This was also a fun game to work on because the premise is that each enemy is intentionally buggy, but in a predictable way. So it winds up making levels into something of a puzzle, because you can anticipate the timing of each glitch that each enemy has, and use that against them.
The idea for that came from the kind of predictable patterns that enemies in a game like Mario have; a lot of times those patterns aren’t rational if they’re really trying to kill Mario, but when you layer them together they make for interesting levels to navigate. I figured that translating that to a tactics game, and making the patterns even LESS rational would be fun, challenging, and also funny. We played up the humor quite a bit, which mostly went over well (some voice lines repeating too much notwithstanding).
Making intentionally-stupid enemies as a way to create challenge was really fun, and again something kind of novel (in tactics, anyway — more or less every platformer game functions this way). It’s very much the opposite of the general ethos that we have toward AI in our games, so it was really fun to be able to do that and show how much of a challenge the game could still be.
I like dinosaurs a lot. Part of why I like chickens is that they are very clearly dinosaurs-turned-birds if you hang around with them much. At any rate, both Valley 1 and 2 had let you transform into a velociraptor as a special ability, and later I would try to make a raptor-themed game.
After Bionic Dues had wrapped, the next project that I started working on with Blue and Cath (at that point it was just the three of us on it) was a grand strategy game called Cretaceous. The premise was that this was a dinosaur society where they had evolved into modern sentient beings. In some ways kind of inspired by the Repitites in Chrono Trigger, but not in a prehistoric setting.
I really enjoy the game Risk, most specifically Lord of the Rings Trilogy Risk, and so this game was going to be sort of a dinosaur-themed combination of a citybuilder and Risk. It didn’t really work from a game design standpoint, and the art also really wasn’t working when we tried to accomplish the sort of things I had in my head. Some things just don’t translate to the screen in a visually compelling way, even if the individual pieces of art are very well done (which they were).
So I punted, and we started working on something else after less than a month on this.
There’s an iOS game called Mega Mall Story that I was really enamored with for a while. You’re basically managing a mall, and seeing it grow and customers come in, etc. You maintain the shops and see how things go, etc. It’s simple, it’s effective, and it works really well with side-view 2D pixel art.
I wanted to do my own take on that concept, but set in a space station. This would be called Starport 28, and there would be I think 15 different races of aliens coming through who all had different interests and needs, etc. It was to be a side view like Mega Mall Story, but not pixel art.
Honestly, the thing that killed this one was not so much game design but visuals. In 2D, there was just no way to attractively get what I wanted to happen. Blue and Cath did wonderful art, but it was my job to bring all of that together and try to make it make sense in a game where you could construct things as you went. It just didn’t work. As a game in 3D, this is the sort of thing I would handle from an isometric camera angle or similar, and I think it could work really well.
At any rate, I knew that if we couldn’t make this look great, then it was never going to sell. I decided to punt on that concept as well.
We started on a new project, this time with Keith also on board, and this was to be a game that was vaguely like FTL: Faster Than Light, but with a lot of differences. I honestly don’t remember the details too well, but I remember the art quite well. It was striking, but again very dark and suffering from the limitations of what we could do in 2D.
We got the battles working in a vaguely FTL-like fashion (no crew inside or things like that; we were going to be focusing elsewhere), but it just was not any fun.
A bunch of ideas swirled around in this time period, but mainly this project was going nowhere fast. There was the kernel of something interesting here, but it just wasn’t coming together.
2014 Starts, And The Last Federation Coalesces
Sometimes you’re just feeling around for a design, when you have a general idea what you want to do. I have a document with literally dozens of unused game ideas in them that I find interesting, and so I’m never afraid to just drop one idea if it doesn’t work. Well, depending on the circumstances.
In this particular case, I knew I wanted to make another space game, and I wanted it to have a simulation-game style to it at least in part, and I also knew I did not want it to be remotely an RTS or grand strategy, because I didn’t want it to compete with AI War.
We started experimenting with some realtime directly-controlled ship combat. There was an old PC game from the 90s called Star Control that had mixed realtime combat with an overarching plot that was kind of strategic, and I had not played it for 25+ years but had some fond memories of it. What we quickly found with the realtime combat was that it was fun, but required a completely different skillset from the strategic/simulation half of the game.
There are people who like action space games, and people who like strategic/simulation space games, but trying to court the intersection of ONLY people who like both of those things seemed profoundly stupid. Plus, the action combat was trending towards the feel of a bullet-hell/SHMUP game, and those are even more niche (though I am a fan).
When I was originally working on AI War back in November of 2008, there was a brief period of a week or so where the game was turn-based. Turn-based free-flowing combat in space!? The idea was kind of cool, but in practice didn’t work when you had a bunch of ships. We already had the concept to just control a single ship (that could spawn other ships that you don’t control, if needed) in this game, so that solved that problem. We resurrected the idea of making the combat turn-based, and that actually worked incredibly well.
Suddenly this was a turn-based SHMUP, when it came to combat. That also meant that we could lean really hard into making it even more SHMUP-like, where bullets are just going everywhere and you have to avoid them. Rather than this being a game of reflexes, this would be a game of planning and preparation. That’s a compatible interest for someone who wants to play a simulation/strategic game.
This also finally solved one of my big gripes about making games in space, namely that there is no terrain in space. That’s kind of a long story, and there are blog posts I’ve written in depth about it, but the general idea is that you need some sort of variegation to the environment to make an interesting RTS. AI War had its own way of dealing with that, which I was only somewhat happy with, but this new game was going to make terrain out of bullets, which made me very happy indeed.
Finishing The Last Federation
I should stress that at this point, in early 2014, we were starting to really feel the pinch financially. We had started and discarded three-ish games in a row, and four overall in the last 12 months. Bionic Dues was a poor earner, Skyward Collapse had started strong but was fading, the Valley games were fading a bit but still decent, AI War was still strong but was starting to fade, and Tidalis and Shattered Haven were just total losses for most purposes.
Income was indeed up in 2013 compared to any prior year, but expenses were also up since Blue and Cath had both become fulltime, and we had more contractors working with us, and so on. Our income was growing every year, but our expenses were growing faster than that as I chased higher production values and more stability.
Again, the irony is that this led to less stability than if I’d gone with a different approach, but my options were limited at the time since I’m not a 2D artist. If I had thought more about it, then I could have kept with my 3D-to-2D approach that I used with the original Valley 1, but I still had a lot of fear based on how people had reacted to that. Plus I really cared about all the people I was working with, and I had zero intention of just laying off someone from a dream job just because the situation looked like it might become inconvenient for me in the future.
Early 2014 was… a time where I really overdid it. I didn’t feel it in a personal sense, but my marriage certainly did. I worked 7 days a week, 12 hours a day, from February 15th of that year through sometime in May, with zero days off. If you’re keeping notes, this is not a good thing to do to your spouse, especially when you have a four-year-old. I still made time every day for my son, but it was pretty much just work, eat, do things with him, eat, work, sleep, repeat. Don’t… do this sort of thing. I am fortunate that I am able to do this sort of thing without collapsing in on myself, but it’s not good for anyone around you even if that’s true.
That does speak to the financial situation, though. There was always the option to just lay some people off and things would be fine, but I couldn’t countenance that when I had any other options. Again, if you’re taking notes: working an insane schedule and making life very unpleasant for your spouse is not a reasonable option to take in order to avoid layoffs.
The other side of the coin is that I really could sense that there was something really cool here. We had picked nine of the most interesting races from Starport 28 and turned those into the races for The Last Federation, as it was now being called. The game came together and…
Arcen’s Last Smash Hit: The Last Federation
Well, I mean I thought it had potential, but goodness. This was far and away more than I expected. The Last Federation released in April 2014 (my ongoing insane-o hours into May were extra unwarranted but born out of anxiety about things that needed to be fixed on launch). The game sold into the six figures in the first month, and I didn’t want that momentum to drop because of something like bugs not being fixed in a timely manner.
Over the next months it continue to rack up insane amounts of revenue. Overall the income for Arcen in 2014 jumped to $702k up from something like $410k in 2013. The Last Federation was officially a hit, and the fastest-selling title that we’d ever had at this point. It made up about half of the 2014 income, and AI War still made up another quarter of our revenue.
One of the best things about this game was that it had finally delivered on one of my major career goals: procedural storytelling that leads to meaningful stories that people want to tell. I am an enormous fan of the story of Boatmurdered from Dwarf Fortress, and a lot of Arcen’s fanbase overlaps with the fanbase for Dwarf Fortress. I had always wanted to create a game where people could tell a crazy story like that in a compelling way, based on things that happened to them and only them.
And it happened, in the form of The Biscuit Federation:
That’s right. My almost-friend, a man I greatly admired, John Bain himself not only told an amazing story from The Last Federation, but then it was also turned into a two-part animated short.
At this point, I honestly felt like I could retire happy. I had hit my major career goals up to that point, and I had a ton of money. Granted, I also had a lot of staff who were burning through that money if I just sat around, so just sitting around was not an option. You always have to keep moving, quick quick, fast fast, when you have staff. So I wasn’t literally thinking I could retire, and I also had no desire to retire, but I was able to… retire from a certain career goal. That procedural storytelling goal had been hit, and it was probably time to find a new goal and focus on that.
Unfortunately, that’s not what I actually did: instead, I doubled down on the procedural storytelling idea in Stars Beyond Reach, but that’s a story for part four. But before we get to that, there’s still two more products that came out in 2014 — definitely part of why revenue was so high that year.
Destroyer Of Worlds
This was it: the final AI War expansion, and basically the end of long-term support for the game. The game was five years old at this point, and given the engine constraints (my engine on top of Unity, not Unity itself), neither Keith nor I felt like we had much more to add to this franchise. It was done.
This last DLC had some parts that people really loved, and other parts that were less popular, like the Exodian Blade. It was released in August of 2014, the absolute worst time of year to release, but we didn’t really care. AI War was evergreen, and it really just didn’t matter.
The first expansion for The Last Federation also launched at the end of 2014, this one in mid-November. That’s also an incredibly risky time of year to launch a title, but again it really didn’t seem to matter at this point in time. I don’t think we lost any money by launching these two DLCs when we did. Any other time, and other franchises, and sure that would have been super stupid.
Anyhow, this expansion added the ability to betray your normal goals (a peaceful Federation of races) and instead go on a war of conquest as the last Hydral. This expansion also added the ability to play in an Invasion mode, where a mysterious new race called the Obscura would come invading the solar system and messing up… everything. Both modes were quite popular, and the Obscura themselves were an exciting addition to the Arcenverse. You haven’t seen the last of them (though it may be a few years into the 2020s before I get to more of their story).
I also brought on an individual by the handle of Misery to work on the SHMUP elements of this expansion, and he really delivered. He would later return to contract again on Starward Rogue. Misery was a longtime Arcen fan and community member, and a ludicrously skilled gamer. We often had to add extra difficulty levels, which we invariably called Misery Difficulty, just for him. He’s a master of the SHMUP genre in general, and has an incredible eye for certain kinds of level design, so he was an obvious choice for both projects.
Looking At The Franchises
I don’t remember what the earnings were for AI War at this particular point in time, but it’s up over $2m now. I would guess that it was around $1.5m at the time. That’s quite a long tail.
For perspective, as of this writing, The Last Federation is only up to something like $800k, and AI War 2 is closing in on that a bit. The Valley games are somewhere in the $600k range.
The year 2014 was incredibly triumphant, and incredibly lucrative. Unfortunately, it was the last truly profitable year Arcen would have from then until (sometime in the future of now, hopefully). And The Last Federation was the last game I made that has turned any sort of profit. I hope to see that turn around with our next title (which is planned but unannounced as of early 2022), and maybe even eventually with AI War 2, but AI War 2 has a long road ahead of it to hit that point (because it’s been so expensive to make).
Before we get into the doom and gloom that pervades the next few years (don’t worry, there’s hope, especially towards the end), it’s worth at least basking in the success of 2014. I didn’t really do that at the time, because I was more concerned with how high our burn rate was and what to do to make use of all these talented people in the best possible way. But it was a really great year for Arcen.
Part four (oops, and now 5) will cover most of the second half of my career-so-far, which I can best describe as a sense of endlessly falling down the stairs for half a decade. I promise, it gets better toward the end.