Intermittent Mechanism

For the past couple of months, I have been hard at work at a new “Let’s Study,” the most ambitious so far. It’s for Half-Life 2, and I foresee it being spread over seven parts. Part One: Linearity is now posted.

There’s a lot of material in this particular Let’s Study adapted from the first chapter of my dissertation, as well as material I developed when teaching the Half-Life franchise in class (including this lesson from my “Comparative Media Poetics” course). My first “Let’s Study” was just a playthrough with some commentary and a bit of b-roll; for this particular series I’m really leaning in to the video essay format more, trying to create shareable versions of what are basically class lectures, or conference presentations. This particular series is still geared very much toward a general audience, but I’m using it as prep for potential future adaptations of dissertation material into video essay format for submission to a genuine peer-reviewed academic video essay journal.

As usual, the script is below the fold. Part two coming soon!

Hello, and welcome to “Let’s Study Half-Life 2,” a hybrid let’s play and video essay, covering one of my favorite games.

Half-Life 2 is a single-player, linear first-person shooter. In my opinion, it’s not only one of the best single-player, linear first-person shooters ever made, but also one of the best single-player games ever made of any kind. It is a master class in using visual and auditory cues to direct players where to go. Its spaces are set up in such a way as to avoid player frustration, while simultaneously leveraging player curiosity to hide its own narrow linearity.

There aren’t nearly as many big-budget, single-player, linear first-person shooters being made today as there were in the past. But even if that genre of video game eventually dies away forever, I think there will still be reasons to look back at Half-Life 2. I think the lessons it can teach us about staging narrative action in a virtual world can be carried forward, not only into other genres of videogames, but also to emerging media forms like VR cinema.

First things first, though: what do I mean by single-player, linear first-person shooter? I don’t want to get pedantic, but I also think it’s a good idea to define one’s terms upfront … especially when it gives you a chance to delve into some fun cross-media history.

1.1 The Guided Tour

Okay, so: single-player, linear, first-person shooter.

Single-player is easy enough: it means that Half-Life 2 is about the experience one player has while interacting with a computer. And a first-person shooter is a game about shooting, presented from a first-person perspective. Your player-character isn’t visible onscreen, except maybe as an arm. Or, occasionally, a foot or two.

That leaves “linear.” This one’s a bit tricker. Basically, the games in the Half-Life franchise restrict player movement to a relatively narrow (if often windy) path. These games provide a “guided tour” through their landscapes, one that dictates that landmarks and set pieces will be arrived at in a certain order by the player.

The borders of what exactly qualifies a first-person shooter as “linear” are a bit fuzzier, and more subjective. Sure, there are the good folks over at TVTropes, who have tried to exhaustively define a five-point scale of “linearity” versus “openness.” But the categorization of games is mudded by the fact that “linear” is something of a dirty world in some circles of gaming enthusiasts. Some people are very vocal in their preference for open-world games, or endlessly repayable rougelikes, or wide-open social Petri dishes, and can be very quick to label anything less free-form with the derisive label “on-rails.”

1.2 On-Rails

Where does the term “on-rails” come from, anyway? And how did it become an insult?

I’m not entirely sure the etymology, but my best guess is that it emerged from a combination of two traditions. One is the history of the tabletop roll-playing game. Roll-playing enthusiasts want a game master that can build an elaborate world, dynamically keeping up and adapting on-the-fly to adjust the story in any direction they want to go. But since GMs are humans, a lot of them can’t exhaustively prepare for such contingencies. They just want to tell one single story. And so they’ll find elaborate ways to force players down a single path, barreling down a single track that they have set in advance. This is known in roll-playing communities as railroading.

The other etymological connection is to the videogame genre of the rail shooter. This is one that I’m going to go into more detail on, since it’s actually relevant to the history of shooter games to which the Half-Life franchise belongs.

Many early shooters tended to implement some sort of scrolling—either horizontal, or vertical. Sometimes, only the enemies scrolls forward, giving the impression of an approaching armada. Other times, the entire landscape scrolls, giving the impression that the vehicle players are controlling is itself moving through space. In the mid-1980s, the rail shooter broke with this scrolling tradition, and instead offered the the impression of penetrating movement “into” the space of the screen.

Early instances of the genre, such as SEGA’s Space Harrier from 1985, appeared years before the widespread adoption of polygonal 3D rendering in games. Space Harrier uses sprite scaling—the rapid re-sizing of multiple two-dimensional pictorial elements according to the laws of linear perspective—to create the impression of forward penetrating motion. Even in the absence of polygonal graphics, rail shooters set themselves apart from their arcade peers by offering the more dynamic visual impression of hurtling forward into a fantastic landscape.

Maintaining this impression of forward movement simply through the perspectival re-scaling of two-dimensional environmental objects, however, required a restriction of players’ possible viewing angles upon the scene. This is where the “rail” component of “rail shooter” comes in: The path of z-axis motion within classic rail shooters is strictly enforced by the game itself, with user input relegated to only moving the player avatar along the x and y axes of the screen. Penetrating movement was an unalterable fact of rail shooters; players were placed on a set track, endowed with a set velocity, and found themselves simply along for the ride.

During the early 1990s, the shooter genre saw a rapid generic splintering and dilution, as developers sought to attract new audiences on new platforms. On the one hand, you had the Japanese arcade scene. Arcades rely on unique hardware experiences to pull players out of their homes. This need for flashy hardware drove a transition from third-person, joystick-controlled rail shooters such as Space Harrier to first-person, light-gun controlled rail shooters such as Virtua Cop and Time Crisis. Light gun games offered a certain tactile thrill. But the plastic gun carried a trade-off: players couldn’t fully control their player’s movement, and as a result were beholden to the rail.

Meanwhile, something completely different was happening in the PC scene in North America. Largely due to the innovations of id Software, home computers had emerged as a viable platform for action games. Mice and keyboards aren’t nearly as flashy as light guns. But id’s breakthroughs with Quake demonstrated that they could provide a lean and precise control scheme for 3D movement. Quake’s single-player campaign is a series of levels that the player progresses through in a set order, from beginning to end. In that sense, it’s linear. But within any given level, players have a much greater freedom of movement than contemporaneous arcade shooters allowed. Quake is a tech demo for then-cutting-edge real-time polygonal 3D rendering. It’s all about marveling at your computer’s ability to render a space you could authentically get lost in. Although it shares a first-person point of view with Virtua Cop, and both games involve shooting, you’d otherwise be hard-pressed to find two games that are more different, in terms of the experience they offer of moving through space.

The Quake franchise kept expanding on this less-linear, less-on-strict-rails approach. In 1999, Quake III: Arena went so far as to drop the single-player campaign altogether. There was no linear progression through a set of levels. There were only levels. Big and twisty levels. Levels that weren’t designed around one player moving “forward” from a set “beginning” to a set “end,” but were about accommodating the quick, competitive movement of multiple players. Quake III: Arena isn’t about moving from a beginning to an ending. It’s about correctly pinpointing where on the map other players are headed, cutting them off, and taking them out. It encourages a very different sort of spacial navigation and awareness, and it deposits us completely out of the realm of linear games.

1.3 Rails and stories

We can draw a lineage from a multiplayer-only first-person shooter like Quake III to more recent games such as Overwatch, Playerunknown’s Battlegrounds, and the first Titanfall game. But the Half-Life franchise isn’t a part of this lineage. As we’ve established, Half-Life’s lineage is that of the single-player, linear first-person shooter. It’s a very a different evolutionary offshoot from Quake, one much more focused on spectacle and storytelling.

Ken Birdwell, one of the developers of the original Half-Life, wrote a famous postmortem called “The Cabal,” where he outlines Valve’s design process for Half-Life. Birdwell’s postmortem reveals a lot about the original Half-Life’s cultural touchstones—the way it borrowed storytelling logic, pacing, and spectacle from Hollywood filmmaking. At one point, he describes the overall tone and pacing the design team was going for as “Die Hard meets Evil Dead.” And he also reveals how heavily wedded story-heavy first-person shooters are to the forward rush down a pre-determined track.

“All content is distance based, not time based,” Birdwell writes of the pacing of Half-Life. “If the players are in the mood for more action, all they need to do is move forward and within a few seconds something will happen.” This description trades in the language of player volition and preference. But if you look at it closely, there is no escaping the fact that the only way to progress through a game like Half-Life is to move in the direction the designers have designated as “forward.”

I mean, going back to Aristotle, there’s this idea that narratives must have a set beginning, middle, and end, arrived at in the correct order. So it’s no surprise that, in a story-heavy first-person shooter, the logic of barreling forward down a designer-set path would re-emerge. Half-Life isn’t literally a rail shooter. It doesn’t automatically determine the player-character’s z-axis movement, as something like Space Harrier does. But this treatment of game space as a narrative delivery vehicle has ensured the continued persistence of pushing forward as a primary spatial logic.

And Valve is quite aware of this—in fact, they’ve been historically quite cheeky about it. The first Half-Life begins on a literal rail, with players taking control of Gordon Freeman as he rides a commuter tram to get to his workplace within the vast Black Mesa Research Facility compound. Later, there’s a chapter explicitly called “On a Rail.” Half-Life 2 also begins on a train (although we only spend a few seconds on it). Half-Life 2: Episode One doesn’t begin on a train, but it ends on one … and then Half-Life 2: Episode 2 begins on that same train. And then there’s this joke in Portal 2, which fully acknowledges Valve’s continued commitment to linear first-person experiences, even in an age where players were increasingly getting taken in by the “freedom” rhetoric of open-world games….

Valve is cocksure. They like winking at the fact that their games’ experiential routes are as locked down as a theme park ride. And by and large they get away with designing their single-player games in this way, because they’re very good at making them. (Or, at least, they were, back when they made games.) There are plenty of players who don’t like the tradition of linear, story-based first-person shooters. In the early 2010s, especially, it was easy to find players bemoaning how far the simple linear corridors of Call of Duty had fallen from the large, open, maze-like exploration of DOOM and Quake. But no matter how many legitimate gripes there are to be made about this particular branch of the first-person shooter family tree, I nevertheless think that Valve’s work in the Half-Life franchise merits our attention. As I said at the outset, I think that Half-Life 2 can teach us lessons that are applicable within a wide range of contexts. That’s because it shows a masterful grasp of staging action in three-dimensional space.

1.4 Where to Look & Where to Go

What do I mean by “staging action in three-dimensional space”? Basically it boils down to two questions. If you’re playing a first-person game, in which you have full control of the camera, you’re frequently going to have two questions. One is, “where do I look”? Where is the relevant piece of action in this scene? Is there something that’s about to shoot at me? Or is already shooting at me? The other is “where do I go”? Ken Birdwell told me that all I need to do is move forward and within a few seconds something will happen. But which way is forward? What if I got turned around? This isn’t a 2D platformer, where all I have to do is go right.

In the first Half-Life, Valve simplifies things, so that, during the game’s opening moments, players only have to worry about one of these questions. I mean, you can move, sort of, a little, within the confines of the Black Mesa Transit System tram car. But, overall, Gordon Freeman’s movement is mostly dictated by the rail. So the first chapter instead becomes a tutorial on how to look. Look at all the things you can … um, look at! Are those … vending machines? Is this a cafeteria? Missiles, and … helicopters, and technology … science!

There are even things that I can look at, at the exclusion of other things. This world stretches around me in 360 degrees. I can’t see in 360 degrees. That means, if I look at this robot cleaning up nuclear waste—which I’m actually encouraged to do, because the voice coming over the loudspeaker is talking about regular tests for radiation exposure—I will miss the man in the blue suit, adjusting his tie. This seems really unimportant—but it’s not. This character is the G-Man. He is of central importance to the Half-Life franchise, and this is his first appearance, ever. And we’re allowed to miss it! Valve wants us to look at certain things, but they’re never going to force our eyes there.

This has certain resonances with the history of cinema, and the history of film criticism & theory. Often, cinema uses editing to draw our attention to certain visual elements of a scene. But not always. The French film critic André Bazin was really fascinated by films that minimized editing, that used fairly long takes, often staged in depth, with wide-angle lenses that not only allowed a lot of space into the frame, but also ensured that things stretching into the background would still be in focus. Bazin loved scene shot in this way, with various actions staged simultaneously in different parts of the frame, in different layers of depth. He thought that it provided viewers with some amount of “individual choice,” allowing us to gaze upon events and make decisions on where to look as we would in real life. More recently, cinema researcher such as Tim J. Smith have rigged up apparatuses for tracking viewer’s eye motions when watching films, breaking down how expert directors use lighting, motion, and focus to direct viewers’ gaze to the most relevant parts of the frame. And then you have films like Mad Max: Fury Road, the makers of which have been very upfront about using techniques such as centered framing to keep viewers’ visual attention in the right place during rapid editing of actions scenes.

As invested as Half-Life is in visual storytelling, its opening shows that there are limits to what sort of tricks the game can borrow from cinema, given the fact that its action has to be staged in 360 degrees. On a certain surface level, the opening of Half-Life bears a lot of similarities toward something like the 1906 Biograph film Hold-up of the Rocky Mountain Express. Hold-up of the Rocky Mountain Express was an evolutionary outgrowth of the phantom ride film genre, which was sort of the rail shooter of the early cinema era. 1980s-era rail shooters wanted to astonish players with their pseudo-3D y-axis penetration into space. Phantom ride films were built around the same pleasure. Motion pictures were new. People hadn’t ever had the mediated experience of simply moving forward in space before. They didn’t even quite have the language to describe it, yet—one critic, fumbling for words and ending up somewhere poetic, said that in these films “an unseen energy swallows up space and flings itself into the distance.” These films established some of the basic vocabulary of the moving image. They’re part of the reason why we still refer to forward or backward camera movement as “tracking” shots (even if the apparatuses we use to move cameras in this way have evolved over the past century and change). They established the technological thrill of the feeling of bodily movement. And I daresay that that’s a thrill that’s still very much with us in visual entertainment.

Hold Up of the Rocky Mountain Express is a twist on this formula. Just as Half-Life’s storytelling ambitions made it distinct from a generation of rail shooters that had come before it, Hold Up’s storytelling ambitions make it distinct from a previous generation of phantom ride films. Hold Up wants to give us a phantom ride, but it wants to tell a story, too. At a certain point, bandits rob the train. As they make their getaway, the camera never moves from its position on the front of the train. As we chug forward, our eyes have to move across the screen, as a series of events occurs across diverging visual arcs. A separate getaway vehicle rides up alongside the hand cart that the bandits have commandeered. They rush over to it, leaving behind the train, and some railroad employees still on the cart. They ride alongside the track, on a separate ridge, firing upon us all the while. Finally, their road re-converges with the track at a railway crossing, where they are apprehended. And all of this takes place over the course of a single long take.

Half Life’s opening train ride gives us similar divergent arcs of movement. But there’s a complication, because we’re also in control of the camera. Unlike in Rocky Mountain Express, Valve can’t just frame everything in a wide shot, and call it a day. Since the action stretches around the player’s avatar in 360 degrees, the possibility alway exists that they’ll look somewhere other than the action is, and miss something.

There are tools available for developers who want to absolutely avoid this possibility. Some games will have moments where they wrest camera control away from the player completely, forcibly pointing the view directly where the action is before depositing it, somewhat awkwardly, back into your control. There are ways to do this fairly elegantly in third-person games, but it’s almost always an awkward kludge in first-person games.

Another option is to direct the camera, but make it optional. Naughty Dog is fond of this in their Uncharted series. Every now and then you’ll hear a little chime, and you’ll see a button prompt. And if you push the button, the camera will briefly make some sort of movement that’s out of your control, pointing toward some relevant bit of scenery. Often, Nathan Drake or one of his compatriots will make some sort of comment, leading up to this. But if you want to ignore it, you can, and eventually the prompt will just fade away.

Nintendo has established an especially-charming way to direct player attention in 3D third-person games, built entirely on character animation. So, in this moment in The Legend of Zelda: The Wind Waker, I’m unsure of where to go. It doesn’t seem like Link can cross that gap, and the camera’s not being any help. But if I just swing the camera around, and look where Link is looking, I can follow his gaze, and look up. And lo and behold, there’s a little tree branch there, that I can use my grappling hook on. And, again: just another quick example of this principle in action. If I’m unsure of where to go, Link’s eyeline is always a very good indicator.  Just pop back into first person, look up, and lo and behold I can see something that I wasn’t quite clearly seeing in the third person point-of-view. And, again, it helps me out.

But Valve never employs any of these tricks in the Half-Life series. Although movement might occasionally be restricted when Gordon is pinned down by something, Valve never takes away player control of the camera. Their games aren’t big on player freedom. But Valve are very committed to giving their player the freedom to look around. Valve has confidence in their ability to get us to look where we should be looking, but they’re also comfortable with the possibility of us missing things. When it comes to relevant story points, they’re big on redundancy. It’s okay if we miss the G-Man here, because we’ll have plenty of other chances to see him throughout the franchise.

So that’s “where do I look.” Next up: “where do I go?” We step off the tram, and … hold up. This is a video about Half-Life 2, not about the original Half-Life. It’s about time we get to it as a case study, don’t you think?

So let’s fast forward all the way to the end of Half-Life. At the end of Half-Life, our protagonist, Gordon Freeman, seemingly successfully closes a dimensional rift accidentally opened at the Black Mesa Research Facility, stymieing an alien invasion. At the moment of his victory, he is whisked away by a character known as the G-Man, a mysterious Man-in-Black type figure that keen-eyed players might have caught glimpses of earlier in the game. Despite his name, the G-Man doesn’t seem to be a government agent—at least, not a representative of any Earth government. During the game’s brief closing sequence, he wields power over time and space, giving us a tour of alien worlds as we ride in a facsimile of the tram car we were on at the beginning of the game. He then offers Freeman employment, with the canonical ending being that he accepts it.

Half-Life 2 begins with the G-Man waking Freeman up from an unspecified amount of time spent in some sort of stasis. The environment that Freeman wakes up in is totally unexpected. Valve chose to use Half-Life 2 as an entirely new canvas, throwing players into a drastically different setting basically no exposition.

We begin, as before, on a train. This time around, though, we won’t spend very long on it.

1.5 Six Tools for Herding

Okay, so this is the final section of this particular video, and it consist entirely of a close look at the City 17 train station.

So, we’re thrown into this entirely new setting, without textual exposition. This opening chapter is going to basically serve the same purpose of the two opening chapters of Half-Life, “Black Mesa Inbound” and “Anomalous Materials.” It is a combat-free exploration of space, using a tour through a space as exposition. But Valve have dumped the literal rails, this time around. We’re not starting on a vehicle. Right away, we’re in full control of character movement. And in these opening moments, Valve sketches out its basic vocabulary for directing our attention and movement in three-dimensional space.

The term of art for this type of direction is “herding.” And just in the opening moments of the train station, we’re going to see Valve employing no fewer than six tools for herding.

First, sound. As we step off the train, there’s a really impressive rush as the sound design changes. We hear a sudden reverberation murmur of idling locomotives, the room tone of a large concrete enclosure—enough to give us the subtle impression of air pressure changes. And then, of course, we hear Doctor Breen’s announcements over the screen, which draws our eyes up toward him. And this is the first exposition we get in this place: we are in City 17, and the human race is under the control of something Breen is euphemistically referring to as “our benefactors.”

Okay, so—simple bit of attention directing via sound. Now—where do we go?

One way to tell is through NPC movement, Valve’s second tool for herding players. This is pretty simple: If we’re wondering where to go, just follow everyone else. And in this opening scene, everyone else heads to the left.

But what if everyone else is already gone? What if we spent too long watching Breen’s video, or playing around with the hovering photo-drone, and we missed that initial NPC stampede? We’re in luck, because Valve likes redundant cueing. And in addition to using NPC movement here, they also use lighting. As we leave the train car, the sound of the Breen announcement screen gets us to look to the left. And then, there at the end of the platform, there’s a wall. And the light is angled on that wall such that the left half is illuminated. It’s a small thing, but it’s exactly the type of unconscious cuing that can get us to look left, and then eventually go left.

I want to pause just for a moment, right in the middle of our six tools for herding, and ask a question: what if a player doesn’t follow the script—either because they don’t read the cues, or they deliberately ignore them? Well, we meet a dead end. If we try to go right, we find that we can’t. We hit up against a chain link fence. Beyond it, we can see a Vortigaunt slave, forced to do clean-up work. So we just get a brief glimpse of world-building, showing us an alien species that players of the first game would be familiar with, establishing the continuity of this story and hinting at the trans-species social relationships of this startling new setting.

This is a tactic that Valve is fond of, and is worth highlighting: False choices you don’t know you’re making. In each of the spaces they create, Valve’s level designers have a clear, linear path that they want players to trace. But they also like creating a sense of expansiveness. They like cultivating the idea that the player could have gone other places, but just happened to go the most interesting and pertinent route.

So, first, they’ll give us the initial illusion of a choice between competing routes. Then they’ll add tiny cues like the lighting, so that most players will go a certain direction for unconscious reasons they’re not even entirely aware of. Players who do this will happen upon the next interesting scene. This contributes to a feeling of constant discovery: because of how things are staged, the space seems larger than it actually is, and more full of interesting expositive action than it actually is.

But, backing up, if players ignore the unconscious-level cues, and actually do exercise some choice-making, they’ll see that the choice was false: there’s only one way to go, in actually. But these more adventurous players are rewarded in their own way. The chain-link fence acknowledges Valve’s constriction of this space. But we also get the little world-building flourish of the Vortigaunt. It tempers our annoyance at being railroaded.

Anyway, onward. We witness a few scripted sequences on our way forward, filling in some exposition that the game’s immediate opening lacked. City 17 exists in some sort of dystopian, totalitarian future. People are being relocated, detained, and disappeared on a regular basis. The cops wear scary masks that distort their voices.

After this woman talks to us, we move onto our fourth tool for herding: subtle architectural clues.

The terminal we’re about to head into is basically donut-shaped. [Overhead view] The exit exit we’re heading to is on the right. If players were go immediately to the right, they’d get to their eventual destination more quickly. Valve isn’t going to prevent players from doing this, but they definitely would prefer that players take a clockwise tour around this space, so that they can overhear NPC dialogue.

So how to they suggest this route to us? Simple. They just add a gate to this entrance. This gate doesn’t actually prevent us from going to the right, by any means. But it visually dissuades us from doing so. The area to the right isn’t blocked off, but it looks blocked off, barred and in shadow. So, we take the long way around, which gives us an opportunity to interact with NPCs, and this whole space becomes an “exposition machine.” We also give these two characters time to hit their marks, so that by the time we reach them we can overhear a conversation between them about Dr. Breen. If we didn’t take the counter-clockwise tour, we could have easily beaten them here. (Just as an example, here’s what that looks like.)

As we navigate through this twisted line of chain-link fences, the paranoid tension ratchets up. We see a character who is being forcibly pulled out of line, amidst protests. And we get directly addressed, as this masked guard barks at us. The guards even push us if we don’t comply with our orders. And as the security camera snaps our pictures, the gate in front of us slides shut.

Now we get to our fifth tool for herding: overt architectural cues. This one’s pretty self-explanatory. Our progress is being literally blocked by gates, here. These hostile guards are acting as Valve’s own avatars in this space. They’re sort of an in-game acknowledgement of Valve’s own dictatorial relationship to player movement. If we get too uppity, try to wander around of our own accord, then they’re going to close the gates, put up fences, and re-assert the rail.

But even at their most dictatorial, Valve still gives us things to be distracted by, and to wonder about. So here, as we trudge down this constricted corridor, we can catch a quick glimpse of the passenger from earlier, still protesting. Again: an extra bit of world-building, adding to the paranoid tension of this scene.

Once inside our own interrogation room, we bear witness to a bit of intrigue. And I’m not going to show this entire scene, just for the sake of time. But we’re introduced to a friendly character … and we briefly make it outside (but not really) … and we have this little tutorial for physics movement….

Finally, passing through the station’s grand concourse, we arrive at the double doors that open onto the main plaza of City 17. And when we throw these doors open, we encounter the sixth and final tool in Valve’s player-herding toolbox: birds.

Yes, birds.

Almost immediately upon entering this new and more open space, a bird flies from the right edge of the frame, in towards the upper portion of the center of the frame. This is scripted. If you play this sequence over and over again, it will always happen, every single time. The arc of the bird’s flight is meant to draw our visual attention to the center of the frame: first to the obelisk in the town square, adorned with another Breen propaganda screen, and, beyond that, the impossibly tall building known as the Citadel, dominating the City 17 skyline.

Valve uses birds to introduce quick movement into visual scenes. It’s well-known that quick flashes of movement can draw the attention of the human eye—this is one of the rules that’s been established by the research I mentioned before by Tim J. Smith. It’s not foolproof, by any means. I’ve done some informal play testing myself, and seen players that don’t look at what Valve is trying their best to point at. It’s not something that players are physiologically compelled to do, by any means. But it’s useful enough as a visual tactic that Valve has used it again … and again … and again.

And that is it for this first part. I’m going to leave you with the parting image of the Citadel, perched on the City 17 horizon. That’s where we will pick things up in part 2. Thanks for watching!