No Man’s Sky is a computer game that came out etc.

Brendan Caldwell’s RPS piece takes a very strong line and is well worth reading. In it, he argues that the pre-release marketing for No Man’s Sky was “wildly unrepresentative” of the finished game and later implies that Hello Games are part of a larger trend which is “poisoning the industry with faux-gameplay trailers and hyper-ambitious promises”.

Here’s three radical suggestions:

Sean Murray is not a liar

No Man’s Sky is, in all meaningful senses, the game that was unveiled in 2013

The game’s marketing does not represent a treacherous tumble into Trumpian “truth-bending”

To clarify, I don’t mean to suggest that Brendan is calling Sean a liar: his article is more thoughtful than that. I’ve seen this in countless other places — it’s almost become a meme — so it’s worth dismissing before we move on to far more nuanced arguments.

While I believe that Hello Games have been treated unfairly, I certainly don’t consider their approach to be perfect. My own PR and marketing efforts could always be better: placed under the same immense pressure, I’m sure I would falter as well.

Also, do not think for one second I am questioning the fact that developers should strive for accuracy and honesty when marketing their games. That is never in question and we should condemn any developer for genuinely deceitful behaviour.

I want to explore some specific claims and provide some context. Let’s blast off into the thrilling vacuum of the beyond!

Too Many Man

Close to the release of No Man’s Sky, two Twitch streamers managed to locate one other at a specific point in the universe. Like ancient mariners on a voyage of discovery, they used the sky to get their bearings. To their great consternation, they could not see each other. The sky led to no man: it was…the no man’s sky.

How could this be? Surely Sean Murray had previously confirmed multiplayer on multiple occasions?

In this video, TotalBiscuit called him on an “outright lie”, due to the following quote in a Eurogamer interview:

…[the player] will be attacked by AI, potentially — very rarely — other players, things like that if they cross paths with them

Here’s another interview with Sean Murray from later in the game’s development:

Multiplayer for the game, we’ve always said, is not really a big focus. If you want an MMO or a deathmatch game, then there’s loads of other games that cater for that really well. What we want…is a sense of other people being in that universe.

So what will happen reasonably often is going to a planet and finding out that someone else has been there before you. You see some traces of them: creatures that they’ve named…

But actually going to a planet and a player being in the same space at the same time is incredibly rare. And it’s something that, depending on how many people play the game, might not even happen, basically. If it does, we want people to have a sense of that.

Bear with me on our interstellar journey, we are approaching our destination.

Now, here’s a video compilation of Murray responding to journalists talking about multiplayer, pithily entitled “Sean Murray lying about multiplayer for No Man’s Sky”:

A quick summary:

He doesn’t refute a claim that the game is “massively-multiplayer”

He says you can run into other people

“You could encounter other players: the reality is, the likelihood of that is tiny”

He agrees that you can “play with your friends”

He responds to a suggestion that you could grief other people with “a little bit”

Finally, here’s a relatively exhaustive list of his comments on multiplayer, if you would like more of the same.

Approaching structure. Please buckle in.

Let’s say you’ve only seen, perhaps, the reveal trailer and one or two of these video interviews. You haven’t read the more detailed quotes above. In my opinion, based entirely on what Sean Murray has said, it would still not be reasonable for you to assume that the game will definitely have fully-fledged multiplayer at launch. There’s evidence for some sort of interaction, but absolutely none for a comprehensive affair. Placed in the context of the game as a whole, the likelihood decreases even further.

So, it’s not reasonable to believe this, but is there legerdemain in play? I think not: someone isn’t lying if they believe what they’re saying to be true.

I am convinced that Sean Murray, to the best of his ability, thought that the light multiplayer features he had planned (such as players being able to see each other in passing) would make it into the final version of the game. That might seem deluded to you; however, the guy envisaged an infinite universe where every planet looks like a sci-fi book cover, so perhaps you shouldn’t talk to him about plausibility and actuality.

For one thing, it makes absolutely no sense to lie: there’s no benefit. If you were going to make up multiplayer features, why wouldn’t you claim the game had full combat? What would be the point of inventing these small subtle enigmatic things which were liable to disappoint people? It’s vastly more likely that a man who spends his life trying to reach the stars has slightly over-reached.

I have complete confidence that we will see — when Hello Games have had time to deal with the current vast demands on their time — further information emerge which puts that issue to rest. There were features on the slate: they were either cut, deactivated on the server or not implemented close to launch.

If Sean Murray believes these things when he’s saying them, what right have I to suggest that his audience has to be skeptical? Simple: we can afford a degree of objectivity which isn’t available to him.

Let’s look one more time at what he’s actually said in the videos:

There is some way in which you could call the game “massively multiplayer” ie many people can play simultaneously in the same space

There is some form of direct interaction with other players

So, we infer that this type of multiplayer is clearly a minor peripheral feature. It is reasonable to assume that minor features will get cut shortly before launch; developers are not always able to announce these for reasons I will go into later. If I don’t really care about multiplayer that much, I stop here: it surely doesn’t bother me.

If I’m truly heavily invested in those ideas — perhaps my deepest fantasy is to shoot Alex Hayes in the head with a laser while jumping off a stone structure that looks like a hippo — the next thing I do, as a reasonable person, is seek out more evidence. I would now look at more considered interviews where Murray has been asked a direct question and given time to expand on the topic. Here’s the Eurogamer quote again:

…[the player] will be attacked by AI, potentially — very rarely — other players, things like that if they cross paths with them

“Potentially — very rarely”. If I am someone who is heavily invested in games, I know that, when a developer says “potentially — very rarely”, it means “this feature is a prime candidate to be cut during development.”

Is the developer being disingenuous? No, because this is a known transaction. We have to, at some point, assume a baseline level of intelligence in our audience otherwise all communication would fail.

Again, Sean Murray is a visionary: his job is to imagine insane fantasies and motivate other people to help him construct them. If you walk around all day spouting qualifiers, nobody will believe you and you won’t believe yourself. If you say, “we are going this way”, people will follow you. That kind of talk can spill over into interviews, which is dangerous.

But! He didn’t say those things to the press or the public when he had a chance to choose his words carefully. He said, “potentially — very rarely”. He did this on multiple occasions:

Multiplayer for the game, we’ve always said, is not really a big focus. If you want an MMO or a deathmatch game, then there’s loads of other games that cater for that really well.

It’s “not a big focus”, which again means it could well be cut. It’s “not an MMO”: that likely means I should reframe the earlier very cursory response to someone shouting “massively multiplayer” in Sean Murray’s face.

So where is the focus, when it comes to these features? Here it is:

So what will happen reasonably often is going to a planet and finding out that someone else has been there before you. You see some traces of them: creatures that they’ve named…

And lo and behold, which multiplayer features made the final cut? The ones he explicitly mentioned as being the focus: players being able to name in-game objects. The other stuff essentially does not matter: it’s not important to the game, no matter how much people have projected onto it.

With that clarified, we can now ask this question: why not be more explicit?

TotalBiscuit suggested it was…

“…bizarre that he would be so evasive on this particular subject rather than simply stating a straight yes or no answer”

The answer to this comes up time and time again in Murray’s interviews. Here’s one from Gamespot:

I guess we’ve always downplayed multiplayer because it’s not really a multiplayer game. Actually, the experience is reasonably solitary. But we want you to feel like you’re playing in a shared universe, and I think it’s important to have those moments.

Murray wants to keep it enigmatic: he wants a slight insinuation of another player’s presence.

Here’s a passage from the same Eurogamer interview I quoted earlier that illustrates what happens when he is explicit:

“You can see the terrain is destructible, which is something we haven’t shown people before…”

So you can terraform?

“We’re not sure how much we want to say because I don’t want it to become the thing. Even showing this I’m nervous, because people will think, ‘Oh, right — it’s like Minecraft’, and then it becomes that. ‘It’s the Minecraft in space game!’

He’s not being more explicit because, frankly, he is the fucking creator of the game. It’s not his job to define and codify and package what he’s doing: that is the job of people who come after him.

He confirms a feature: immediately, things spin out of control. Destructible terrain, to the interviewer, directly implies “terraforming”: potentially an extremely complex set of actions which have huge implications about the entire course of the game. Murray immediately backs off, becomes a little defensive, retreats to meta-discussion about the difficulty of communication.

Have you ever seen an interview with a musician where the interviewer tries to discuss genre or pin them down on the specifics of their process? 99% of the time the musician will say “I don’t really think of it in those terms”.

“What were you actually doing?”

“Relaxing,”

Garbage in, garbage out. I once interviewed a well-known producer and asked him how he went about starting a track. He began his answer with, “I consider myself to be a fisherman of sound…”

I’ve always had a love-hate relationship with this approach. If you pinned me down, I’d probably tell you that creative people should be a bit more self-aware and communicative, a bit more giving to the members of the audience who have genuine questions. I’ve been repeatedly battered with disdain for this opinion by so many creative people that I’ve now given up on it. Sorry.

Is it pretentious? Yes, but so is the process of creating anything which is more than functional. Without pretentiousness there would be no culture.

Murray’s sole role isn’t just that of a creator, but characterising his more enigmatic statements as “bizarre” or “evasive” completely misses the point.

Also, he does give a direct “yes or no answer” in this interview, where he’s compelled by the rapid-fire format to do so…

Interviewer: Will you be able to play with your friends?

Murray: Yes

This only serves to get him into even more trouble! He doesn’t want to play the game, and as such he can’t win.

His deepest desire is for the game to be enigmatic; he wants to express ideas through the game itself, not in interviews. In the same interview he talks extensively about the answers he “wants to give” and the ones he feels he has to in order to satisfy those who are asking questions.

People love enigmas so much that they want to grab them and set them in stone. Ambiguity is seen as empty territory to be immediately colonised without stopping to admire its natural beauty along the way.

So, beaten down by the process, on the day before launch he makes the following statement:

He finally gives up on the whole thing: people want certainty because apparently the most important thing is a “well-informed consumer”, not the ability to approach a creative work on its own terms. He’s reduced to being “super clear”.

The game then comes out and he is excoriated for lying.

Between Two Trenches

Is it fair or realistic to ask Sean Murray to perfectly deflect every comment about multiplayer in high-pressure interviews? No.

Is it fair or realistic to ask the audience to read between the lines and predict cuts? I think so, if the correct caveats are employed. I’ll discuss the nature of developing this sort of game in a little bit, but first let’s talk about PR.

Murray didn’t want to play the game of traditional games industry PR, just as he didn’t want No Man’s Sky to have a “classic motivation” for the player: these decisions helped create an enormous success. They also caused a few mistakes.

These mistakes do not equate to him telling lies, but I acknowledge them anyway for the sake of completeness.

People will ignore caveats. If you see this…

potentially — very rarely — other players

…and you are desperately hoping for multiplayer, you’re going to focus entirely on the “other players”. However, that does not mean you should stop using caveats. Developers need to employ them constantly, apparently to the point of absurdity, to satisfy this sort of detective work. If you have to say “to the best of my knowledge, at the present time” before literally any statement you make, you should do so. This is what the audience demands.

More realistically, I think book-ending every interview with, “this is where we are right now, everything is subject to change” is worth doing, even if it gets cut out or ignored. I think the only time we’re really buying into “post-truth” is if we give up on these formalities: they help keep you grounded more than anything else.

Even though it should be implicit that interviews mid-development come with the caveats, they need to be restated. Some people do not do implicit.

Ironically, if you hate the machinations of PR, you need to have more clarity about your message going into an interview. I would frame the whole thing like this:

“This is not a multiplayer game. You may have some sense of encountering other players, but it will most likely be quite indirect and probably not involve combat. You can see planets and creatures that other people have named, and that will add to your solo experience.”

I think that message could have been established earlier, despite the problems with that which I will go on to discuss.

I also would be on the lookout for “gotchas” where people throw loaded phrases like “massively multiplayer” at me, but this is very hard to do in certain formats of interviews. He managed to catch the “so there’s terraforming”, for example, but missed others like this: it’s understandable in the context of a live interview, but still a mistake. It does not constitute lying.

There’s no need to elaborate further: this is a straightforward idea. Game developers do things like call the Wii a piece of shit in a public talk and then express surprise when a journalist quotes them on it (sorry Chris, you are awesome!): I suspect the audience for such stories does not want this to change. If you want developers to speak directly, they will make mistakes: you can have this or well-honed PR; your choice.

Blue Sky Thinking

I’ve spent a long time looking at the idea that Sean Murray has lied about a specific factual issue. It’s pretty tedious, but it matters to me because I think this sort of claim is unjust and should be stopped.

Some people, like my esteemed colleague Ian at Mode 7, believe this issue is a total irrelevance and that claims of deceit are just an external manifestation of a general feeling of disappointment: cut off one shrieking head and two more will grow in its place.

Calling someone a liar just because you feel bad is never justified. If you say things you don’t mean when you are angry, you are doing it wrong: anger should be a lightning rod for truth.

But let’s look at the big picture. What happened to “the game we were sold in 2013”? I’d suggest that it was recently released!

Trailers are not promises.

The closer a game gets to launch, the greater the correlation between trailer and raw gameplay footage should be. But trailers are not promises.

If you want to know what a game is really like, wait for the late-stage footage. If you’re unsure, or a specific aspect of a game matters to you above all else, wait for a review. If you want to get a general sense of the developer’s ideas, watch a trailer.

The absolute gold standard for announcement trailers is, in my opinion, that features should be:

Plausible

Practical

In-engine

On the schedule

Defended against cuts

You cannot possibly have anything more. I think the above trailer meets these criteria, whereas some of those for Colonial Marines and BioShock Infinite did not.

Why can’t trailers be perfectly accurate from the get-go? Games like No Man’s Sky are developed initially for their visual impact: let’s call this “aesthetic development.” If you need external funding, or to kick off a huge publicity campaign; you do aesthetic development. There is no alternative.

If you want to play big budget games which are both visually and conceptually ambitious, you need make peace with this process. There is virtually no developer in the world who could put up all of the cash for a long period of front-loaded gameplay iteration on a game this visually advanced: everything has to be geared towards attention and funding; not core gameplay and accurate feature lists.

We were very lucky with Frozen Synapse: we didn’t need money to make it and we didn’t need to put out trailers immediately: things were very different back then. The game looked like this for probably two years:

Surprisingly, we didn’t get a call from Stephen Colbert.

When we put out our first trailer, I could have told you in a great deal of detail what would have been in the final version. Hello Games didn’t even know which genre their game was in when their announcement emerged in 2013.

The fact that NMS looked so good at the outset tells you all you need to know: it was wildly ambitious with few hard decisions made about the core gameplay. If you really need to fill in the blanks, there was a lot of additional information available to the public. Hello were a fairly well-known developer; their size and relationship with Sony was made explicit. It was clear this was a “moon-shot”; it was also clear that the responses and expectations were unprecedented.

Is it a huge stretch to suggest, then, that fixating on specific moments in early trailers is inherently ridiculous? It’s there to give a sense, an impression: very dangerous when a huge audience insists on taking everything literally.

Having said that all that, I do understand why people see something like the fleet combat, or the animal behaviour in a later trailer, or the crashed freighter and say, “I want that.” These things are very cool ideas. Sometimes, the more you desire something, the more you believe you deserve it; that it has been promised. But this is a false assumption and trailers are not promises.

Look, I don’t just understand: I empathise. One of my favourite games of all time is Wing Commander Privateer: it eschews bland “numbers-in-space” vastness for a small galaxy with a host of baked content packed with personality. It sprinkles in random encounters and events to keep you on your toes, but is essentially a very crafted experience. I adore that game: I want to replicate parts of my experience with it wherever possible.

So sure, I felt disappointed when No Man’s Sky didn’t turn out to be like that. But at no point did I feel misled: Sean Murray made some off-hand comment about trading but the composition and nature of the game became manifestly obvious over time. I watched a late gameplay video and saw the survival stuff: my trading game didn’t exist. I didn’t feel the need to blame either myself or Hello Games for those feelings: I didn’t want to accuse anyone. Why do I feel so alone here?

Do Mans Sky in the Game?

In his video which I linked previously, TotalBiscuit uses the term “desperation genre” to refer to genres which are so poorly specified that they encompass other genres and vast numbers of features, to the point where every aspect is necessarily simplified or impoverished. Ian uses the term “maximalism” when he talks about this sort of thing.

I remember some very early interviews on Frozen Synapse. Interviewers would constantly ask me questions like “can I dive behind cover?” or “can I use flash-bangs?” This baffled me initially: we largely wanted to talk about aesthetics and gameplay. Frozen Synapse has a very tight, limited feature-set: it was an attempt to make something elegant in the face of bloated, over-complex strategy games; surely it was implied that the mechanics were bare-bones?

Surely, it doesn’t matter if you can use flashbangs or not: what matters is how the gameplay is constructed. You don’t look at a game of Go and start asking if the pieces can fly off in helicopters.

I’ve since learned that this is a hugely naive view: nobody really cares about gameplay until a game is playable. They care about three things:

Does it look cool?

What can I do?

Does it feel like something I enjoy already?

That’s it. Only game design nerds and people working on the game care about its mechanics in any way, right up until the moment they get to play it.

“The things you do in a game which are not explicitly game mechanics” doesn’t have a common term. I’ve heard “verbs” and “affordances”; suggestions from my Twitter followers included:

Peripheral/vestigial interactions

Optional

Playvarication

Actvities? deeds? accomplishments/ conducts?

Interactive activities, or “Interactivities” as I may call them from now on

Actions

Russian Roulette Decision Making

I like “doings” because it’s a silly word, so doings it shall be.

In this context, if you work in game development for long enough, you start to fantasise about interviews where you can answer yes to every question about doings. If people really want doings, why can’t I provide them? You want to this to happen:

What can I do in the game?

Anything.

That’s the logical conclusion: the “omnigame”, a game comprised entirely of doings.

The more discrete doings in a game, the shallower each doing will be: that’s necessarily true. If you attached each doing to a game mechanic, the game would become hugely unwieldy. So some — like shooting and mining in NMS — become essentially the same action, causing a game to feel repetitive.

People want doings so much that omnigames have taken over: maximalism is here to stay. Look at any Top Sellers chart for pay-once games: Rusts and DayZ’s and Arks and many, many others continue to dominate. This isn’t just a trend, or a genre: it’s a paradigm.

Let’s not forget that No Man’s Sky is a stratospherically successful game in commercial terms: it’s currently dominating the charts on both PC and console. You can’t argue that this is because an audience has been somehow misled or manipulated: reviews and Let’s Plays are out.

The entire game is to make something which has infinite potential and creates infinite discussion. The audience has been ramming this in the face of every developer who looks at their game on launch day and sees DayZ still hovering above it. Hello have won that game: why are we complaining at them for having done so?

That’s all fine and people enjoy those experiences. However, there are some innate drawbacks to the omnigame. One is prioritisation, and No Man’s Sky definitely suffers from this. In a game with survival and space combat and ground combat and animals and trading and AI baddies and exploration and resource management, it is completely impossible to know where to place emphasis.

That means anything can be cast as a minor feature and cut at any time. Get invested in trading? Minor feature: it’s up for grabs. Animal behaviour? Up for grabs. This was completely transparent and obvious virtually from the get-go: if a game doesn’t declare its mechanics upfront, it’s because the developers haven’t figured them out yet. If they’re aiming for a vast scope, anything goes.

If we don’t like this situation then we have to stop focussing on potential and “what you can do” in a game; we have to start prizing gameplay above all else. I submit that this is a foolish suggestion; it’ll never happen and everyone has to accept the current situation.

The Moon on a String

People have become so obsessed with what they feel is missing from No Man’s Sky that they’re ignoring the 18 quintillion planets staring them in the face.

Here’s some things I thought were probably bullshit in the first trailer that I have directly experienced within two hours of gameplay, or seen via people tweeting about the game:

Cool sea creatures

Seamless transition from sea to land

Awesome looking foliage and trees

Varied-yet-stylised colour palette for environments

Formation of fighters flying over my head on a planet which can then be followed into fucking space come on are you serious

Seamless transition from land to ship, then from planet to atmosphere

Big trading stations visible from planet actually there in space

Flying through a destructible asteroid (for some reason I really believed this would be impossible)

Large-scale creatures (no, I’ve not seen sandworms yet but I am a True Sandworm Believer: they might be out there!)

That’s a genuinely astonishing list given that this is the very first trailer. Remember that trailers are not promises and that aesthetic development means that accuracy about features is impossible. Add to this the challenge of the omnigame; where the number of possible features tends towards infinity.

I’d go so far as to suggest that Hello Games have done an incredible job of delivering on their initial promises. Let me take one technical aspect about which I can speak with some limited authority.

After seeing the creatures in the trailer for the first time I thought, “Doing audio for this is going to be a nightmare: there is no way the audio can live up to this level of random creature generation.” Up pops the incredible Paul Weir, who collaborated with Sandy White to produce one of the best procedural audio systems I’ve ever heard. You can see it demonstrated here:

I saw a lecture on proc gen audio for games a few years ago and the guy was able to make a very rough engine sound, as well as something which sounded a bit like a helicopter if you weren’t paying much attention. Paul Weir’s work, however, is a brilliantly neat deployment of physical modelling synthesis coupled with some very clever data management. It absolutely trounces Mr Helicopter— it’s light years ahead and sounds unique.

The trailers are not “wildly unrepresentative”: they represent exactly the attitude with which Hello Games approached development.

Wells of Infinite Sadness

So, have Hello Games poisoned the well with their approach to developing and discussing the game? Are we plunging into a “post-fact” games industry?

It seems nonsensical to claim this when extensive Let’s Plays of every game come out either on launch day or before launch; when the internet allows people to share their impressions of a finished product seconds after playing it. If I turn on Twitch, I can watch someone playing the game right now in a completely unfiltered way for as long as I want. We can’t see what Donald Trump would be like in the White House and then decide whether or not to vote for him.

In actuality, we have a supersaturation of facts about games and this has driven developers to try almost anything to cut through the noise.

Could it be that the audience wants hype? The fact that nobody plays demos any more, coupled with vast pre-order numbers, seem to indicate that people increasingly want to make a commitment pre-launch: they have a need to buy into the excitement. Otherwise, why not just wait?

We can’t ask developers to provide this service— to give us everything prior to launch — and then be upset when they subsequently fail to deliver. We have to approach marketing material, and certainly early trailers, with some degree of critical judgement. If we care about games to the extent that we want to consume vast amounts of information about them prior to launch, isn’t it our responsibility to do that appropriately?

If the audience can manage this, how should developers conduct themselves?

I definitely think Hello Games could have handled some aspects of communication better: I’m sure they would agree. Everyone in game development could do more to communicate accurately and in a timely manner: it’s literally a daily struggle. There’s work to be done and nobody is above criticism.

It’s clear that some things were cut from the game at a late stage and these were communicated poorly. I’d call on publishers and devs to do more to ensure that fans and potential customers can be notified in situations like this. Much as I’ve harped about approaching games as creative works, they are also products and consumer rights need to be respected.

Things can change: both Sony and Microsoft have undergone nothing less than a revolution in the way they deal with developers in the last few years. We now have Early Access coming to consoles; we are a universe away from Nintendo’s clampdown on unlicensed developers in the late 80’s. We have to keep pushing so that communication channels can be opened further.

It’s not satisfactory, though, to hand-wave the intense difficulty of dev communications on a title like this with statements which amount to “sometimes there might be legal issues, I guess”. There are myriad reasons why cuts can’t be communicated well:

You don’t know you have to cut something until very close to release; there isn’t time to do a press release due to needing to coordinate with platform-holders

You have promised or half-promised a feature to a publisher and you don’t know if you can implement it, so you have to keep quiet about it until the last minute

You realise that a very small feature makes a portion of the game rubbish and you make the difficult decision to cut it; having previously over-invested, a loud but small facet of game’s community will massively misinterpret the cutting of this feature and cause a giant fuss about it, overshadowing the rest of your message and ruining your launch

I’ll pause here: no developer ever wants to make this call. I’ve luckily never had to: it’s extremely difficult from a PR standpoint. You’re damned if you do, and damned if you don’t, but the best policy is to be direct and honest as soon as possible. It doesn’t always happen. These are not excuses but I’d like to ask anyone what they would honestly do in this situation — piss some people off or save the launch?

Something causes a horrendous bug and you find this as you go for final TRC submission; you just don’t know if you can fix it

There just isn’t fucking time to talk about some problems while you’re trying to fix them; you forget and by then it’s too late

Your NDA means you can’t talk about a specific technical or commercial facet which explains a cut, or you’re not sure if it means that and you don’t have time to talk to lawyers about it

You don’t have enough servers at launch; something goes wrong with whitelisting; literally nobody on the planet can ever get launch servers completely right

You had to cut something because a publisher pushed you to work on something else; you can never say so in public

You fully intend to have something at release but then have to cut it off server-side due to load or other problems

Again, I could go on, but I hopefully don’t need to.

Developers should fight against these issues and try to break down further walls: customers deserve it and we can all work harder.