I didn’t get around to mentioning it in my end-of-2017 retrospective, but one of the big stories of the year was that Marc Laidlaw, the lead writer of the Half-Life series, published his own story outline for the game that Could Have Been But Never Was. Laidlaw had been with Valve for 18 years before departed the company back in 2016. The story he published is ostensibly what was planned for Half-Life 3.
Having read the story synopsis, I have to say it felt just right. This feels exactly like the sort of story I’d expect from the series. Outside of Valve everyone had guesses, fan theories, fan fiction, and suggestions for what could / should happen in Half-Life 3, but none of them quite hit the mark the way this did. Like I said during my Mass Effect series:
Sir Terry Pratchett was an amazing talent. But if J. K. Rowling had hired him in 2002 to help her pump out Harry Potter books twice as fast, it would have fundamentally changed the tone of the series. Different creative people come up with different ideas, and this will give the new work a different texture. And even if it's an improvement – even if you want to argue that Pratchett-Potter books are better than Rowling-Potter books, the new books will still feel ill-fitting and alien to people who fell in love with the originals.
Amazingly enough, it turns out Marc Laidlaw is really good at writing fiction in the style of Marc Laidlaw, so this unofficial ending to the story rings true for me. This takes the edge off of never getting a follow-up to the cliffhanger ending of Half-Life 2: Episode 2. We at least have an answer to the question of “Where was the author going with all of this?”
Of course, this doesn’t ease the annoyance of never getting another Half-Life game.
Was Half-Life Really all That Good?
A common refrain I hear these days is that Half-Life isn’t as good as you remember it. Supposedly everyone is looking at the game through rose-colored glasses, warped by childhood nostalgia. I think this is sort of true in the case of the original Half-Life, but I’m convinced that Half-Life 2 holds up just fine today. Sure, Half-Life 2 has its critics – all games do. But I think if you liked the game in 2004 you’ll probably still like it here in 2018.
Half-Life 1 is somewhat marred by the one-two punch of jumping puzzles and dodgy roller-skate movement physics. The difficulty is really uneven and the final chapter is legendarily unpopular. It was incredible for a shooter of its generation, but if a modern game had these flaws it would end up savaged by the audience. In contrast, there’s nothing particularly clunky or outdated in Half-Life 2. (Aside from the graphics. And even those look pretty dang good for a 14 year old game.) So understand that when I defend the legacy of the franchise, I’m mostly talking about Half-Life 2 and it’s episodes.
Having said that, I can understand why people of a certain age might say Half-Life 2 is overrated due to nostalgia. Part of becoming an adult is looking back through the stuff you loved as a child and sorting the things that were genuinely great from the things that only seemed great because you didn’t know any better. But I don’t think the love for Half-Life 2 is based (entirely) on starry-eyed childhood nostalgia. I played the original in 1999 when I was 28 years old, and I played Half-Life 2 when I was 33. I wasn’t some child who didn’t have a frame of reference for quality. I was fairly well-seasoned as a gamer, I paid for my own games, and I had a pretty good handle on figuring out if a game was working or not.
Rather than blaming nostalgia, I think this game was the product of hard work and good timing. I think that time period represented a unique period in the evolution of PC gaming, when the development cost / benefit curve was “just right”.
The gaming audience was big enough to support ambitious “big budget” games, but development houses were still small and independent enough that they could take chances and try lots of new things. Graphics were just good enough that we could portray environments that were recognizable as whatever they were supposed to be, but those environments weren’t too expensive to build. The internet was ubiquitous enough and robust enough that it could be used to support enthusiast communities, but it wasn’t widespread enough to support phone-home DRMOther than Steam itself. Which was bad for 2004 but seems downright gentle compared to what came just a few years later.. The technology was good enough to make cool stylized cutscenes, but game developers hadn’t yet embraced a world of stilted photorealism, expensive motion-capture, awkward lip-sync, and cringy dialog in an attempt to be more like Hollywood and less like game developers.
I’m not saying that the years from 1998 to 2004 was this perfect time of flawless games. There was a lot of crap then, just like now. But our fond memories of those times aren’t just nostalgia. There really were some unique gems back then. If you look at the best of the time period you’ll see games that couldn’t have existed sooner because the technology wasn’t ready, and they couldn’t exist today because they’d be too expensive to make using today’s standards in visual fidelity. Even here in 2017, there’s still a gap between the $100,000 indie budgets and the $60 million blockbusters.
So What Was So Good About Half-Life Anyway?
Technically, anyone is free to make “another Half-Life game” in the sense that anyone could make a game in the same style. If Crytek wanted, they could make “Decay Time”, a sci-fi shooter about Gary Wildman, an aerospace engineer wearing Hazmat power armor and fighting off alien invaders called the Harvesters. Or whatever. I mean, they sorta tried. The thing is, none of those things are really the core ingredients that made the series such a hit. And if we try to drill down and figure out what the supposed “core ingredients” of Half-Life are, I imagine everyone will have a different answer.
But let’s try to do it anyway, since that’s sort of the point of this site.
Maybe testers would be confused about where they were supposed to be going. The brute-force solution most game developers would use would be to clutter up the HUD with giant glowing waypoint markers, holding the player’s hand and telling them where to go and what to do. In contrast, Valve would change the lighting to draw your eye in the desired direction, tweak some dialog to clarify your goals, and use pickup items like health and ammo to draw you towards your destination.
Maybe testers would be confused about what they were supposed to be doing to progress because they didn’t understand how to shut off a particular machine. The modern gamedev would throw the directions up on screen and leave it there until the task was done. Something like, “Shut down the force field at the security console.” Or maybe they would give you a buddy character to badger you with repeating dialog. Maybe they would design the security console to be a comically oversized red button (which would of course clash with the photorealistic art style) so you can’t miss it. In contrast, Valve would make it so you could “shut down” the security console using the existing language of the game. If they’ve already established that shooting a BLUE THING will cut power, then they can just make BLUE THING part of the security console and the player will immediately know what they’re supposed to do without the need for text prompts, dialog, or any other tools that make the player feel like they’re being micromanaged.
If testers seem to be wandering off and bumping into the edges of the play area, Valve doesn’t hit them with an immersion-breaking message telling them, “YOU ARE LEAVING THE MISSION AREA”. Instead they redesign the layout to draw players towards the action. Rather than yelling at the player for leaving the world, the designers put up visually plausible barriers so they can’t leave the world.
The thing about this dedication to polish is that it’s completely invisible to the player. They’re not sitting there saying to themselves, “Man, Valve sure did a good job of drawing my attention to this doorway!” This is the kind of thing you don’t even notice until it’s gone.
In a lot of ways, it’s a slower version of the 4-step design philosophy used in games like Super Mario 3D World: The designer introduces a new tool or mechanic like grenades, the crowbar, or the gravity gun. At the same time, they place you in a situation where it’s pretty obvious what you’re supposed to do, and they allow you to experiment with the tool in a safe situation. Often the situation is designed so that you have to demonstrate understanding of the concept before you can proceed. (Such as having wooden planks barring the way that require you to use the crowbar you just picked up.) A little later you’ll encounter the obstacle again, but now you’ll need to use your new tool under pressure. Then the game designer will test your mastery of the tool by having you use it in more complicated situations or against more extreme dangers. Then finally, you’ll run into situations where the now-familiar mechanic can be used in new ways. (Like using the rocket launcher to fight striders, or using the gravity gun to throw grenades back at the enemy.)
They don’t perfectly adhere to this formula, of course. You have to learn to use the gravity gun to flip the car while under attack. And while the designer allowed the player to lean how to pilot the airboat in a safe arena before moving on, the player needs to learn to drive the car while being hounded by antlions. Still, the design philosophy is there and you can see it at work throughout the game.
Maybe the designer intended the game to deliver powerful emotional moments like Last of Us, but either the writer wasn’t given the leeway to establish and develop the characters and world well enough to build an emotional connection with the player, or the writer simply lacked the skill and had to resort to brute-force tools (like the death of Some Kidd) in the pursuit of drama.
At the other extreme you have games that just see the story as an excuse to string together set-piece battles, but the writer never got the memo. So you wind up with a game where the story didn’t matter to the team, but it still has a bunch of blustering, cliche, overly-long cutscenes for some reason. It’s a gameplay-focused game where the gameplay stops for a non-interactive cutscene, which is the worst of both worlds.
Half-Life 2 was a game with a story that wasn’t really about the story. It never stole control from the player for a cutscene. At the same time, it didn’t treat the story as some half-assed afterthought used to string action sequences together with contrivances. The story was there to create and sustain a mood, and it did that beautifully.
On the other hand, the fact that the story is told through gameplay means you can’t skip the talky bits. This usually isn’t a problem, except for the three or four moments in the game where everything stops for an extended conversation. It’s nice to be free to walk around during these bits, but it would also be nice if there was a convenient way to skip to the next bit of action. The player can load up the next chapterThe really long conversations are usually isolated into chapters to facilitate this. but that’s an inelegant solutionLong scenes can’t always be skipped this way. Even when you can skip to the gameplay by starting the next chapter, you’re reset to the default ammo loadout chosen by the developer. This negates whatever ammo hoarding / squandering you’ve been doing.. It’s not perfect, but it’s preferable to the way a lot of shooters handle things.
Valve is Done Making Games
Lots of people are making shooters, but nobody made shooters quite like Valve. Oh, they try, but it’s harder than it looks. But now Valve is out of the shooter business, so not even Valve can make Valve-styled shooters.
Erik Wolpaw – writer of Portal – left Valve in early 2017. Chet Faliszek (Left 4 Dead writer) left a few months later. Marc Laidlaw is gone. As far as anyone can tell, Valve no longer employs any writers. That means no more Portal. No more Half-Life. Heck, it means no more Team Fortress 2 shorts and comics.
I’m sad we’ll never get Half-Life 3, but I’m grateful Laidlaw gave us some closure. I’m kind of curious why Valve never gave us the same courtesy. For years they promised, “Yes, we’re working on it. Yes, it’s coming. Be patient.” And I’m reasonably sure that was true when they said it. But based on what ex-Valve employees have been saying, it’s been ages since it was true. It’s been years since anyone was working on Half-Life 3. Even if Valve miraculously decided to resume development, it’s been so long that they would probably have to start over. And anyway, they no longer employ the required people to make it happen.
Most of us have concluded that the game is dead, so I have no idea why Valve has been reluctant to make it official. There’s nothing to be gained by pretending the game might still happen, and publicly canceling the game might allow people to move on and end the frustration of the poor fans who have been waiting and hoping for over a decade.
And please Valve, don’t hand the license off to a random third-party studio. The only thing worse than not getting Half-Life 3 is getting a Half-Life 3 that plays like Call of Duty.
 Other than Steam itself. Which was bad for 2004 but seems downright gentle compared to what came just a few years later.
 The really long conversations are usually isolated into chapters to facilitate this.
 Long scenes can’t always be skipped this way. Even when you can skip to the gameplay by starting the next chapter, you’re reset to the default ammo loadout chosen by the developer. This negates whatever ammo hoarding / squandering you’ve been doing.
Project Button Masher
I teach myself music composition by imitating the style of various videogame soundtracks. How did it turn out? Listen for yourself.
Trusting the System
How do you know the rules of the game are what the game claims? More importantly, how do the DEVELOPERS know?
Good Robot Dev Blog
An ongoing series where I work on making a 2D action game from scratch.
Final Fantasy X
A game about the ghost of an underwater football player who travels through time to save the world from a tick that controls kaiju satan. Really.
A programming project where I set out to make a Minecraft-style world so I can experiment with Octree data.