Episode One Hundred and Twenty Seven: Belong; Snow Crashing (9); Humans As A Service

by danhon

0.0 Station Ident

45 minutes of near-silent swooshing in a black car from home to Manchester Airport and then suddenly being surrounded by something that’s best described as a mess of people. Not a throng of people, but more a spill, hastily mopped up, the equivalent of wads of paper towel attempting to force them into lines that go this way and that: departure check-in desks the week that school holidays start.

For all the talk of a manufactured normalcy field, this mass of people doesn’t look like anything particular normal is going on. Maybe in countries where air travel is more routine, more the fabric of everyday life because you’re country is just So Damn Big. But not at 4:30am in the morning in Manchester.

I’m still angry about the health insurance thing. Not least of which because I’m pseduo-helpless and for the first time, it feels like I’m reliant upon some sort of native guide. I neither have the time nor the inclination to spend a bottomless-pit’s worth of time and energy into researching what I need to know, and how much to pay, and whom to pay it to, and whether one plan or another is going to be worth it. Never mind the fact that a cursory glance at the plans available in Oregon don’t necessarily line up with what it is that I want, whether or not I wanted to pay money for it. But this is healthcare, after all.

1.0 Belong

It’s easy to make fun of Airbnb’s new logo, but on reflection, I’m with Brand New[1] – it’s a perfectly good logo, one that will be distinguishing for them. The objection is all the dross and wank that comes along with it: the naming of it and the frankly offensive long copy that tries to justify and explain what should instead be experienced[2].

For example, this type of description:

“We used to take belonging for granted. Cities used to be villages. Everyone knew each other, and everyone knew they had a place to call home. But after the mechanization and Industrial Revolution of the last century, those feelings of trust and belonging were displaced by mass-produced and impersonal travel experiences.  We also stopped trusting each other. And in doing so, we lost something essential about what it means to be a community. After all, our relationships with people will always be the most meaningful part of our lives. You just need to get to know them. That’s why Airbnb is returning us to a place where everyone can feel they belong.”

runs counter to anecdata of several friends I have of checking into an Airbnb and being categorically told by their hosts “if anyone in the building asks who you are, just say you’re relatives.” Because any sense of community and trust is first founded upon denial of identity, right?

I worry about the sheer amount of rhetoric involved and the language that’s used to say quite a lot when instead a little could do a better job. The audience for the blog post isn’t ever going to be that big compared to the eventual audience of the mark. If Airbnb succeeds in getting it in the windows of participants of the sharing economy – excuse me whilst I gag – then that’ll be a tremendous and recognisable achievement. But instead, the blog post is instead more of a dog whistle to those who’re more attuned to these kinds of things; who’re going to be more critical of it.

Does any of this matter? Maybe not, not in the long run at least. But I’m suspicious of brands that like to talk about themselves quite so much. Personally I would’ve preferred something a little less pretentious and a little more demonstrative. I don’t know what the takeup figures are for people creating their own Bélos (if that’s the right plural), but my suspicion is that the number is either a) low, or b) high and satirical. I know I certainly tried with the latter.

Part of this comes down to community management: Airbnb is saying that there exists a nascent community of sharers that they want to grow (and less explicitly, that they want to own). The premise of using Airbnb now is not simply that you want to find somewhere to stay, or that they’re taking advantages in terms of badly planned transient resident housing resource, but that in an increasingly mobile society, we all want somewhere that feels like home. Part of the delivery of that is through the “community” of users who let out their residences with Airbnb in the first place.

I don’t have access to Airbnb’s internal metrics about who’s doing what, but for those who’re following along with the outrage at the Facebook emotional contagion study from a few weeks ago, here’s how something like this works. Airbnb introduces a new metric by which guests rate their hosts: curation of the local neighborhood. At the same time, Airbnb lets its hosts know that they’re going to be rated by their guests on how well they introduce them to the local neighborhood. “Here’re some of the things you might want to do,” says Airbnb, like list the local coffee shops or, for other kinds of value-add, you might want to take them on a walking tour. This isn’t exactly great news for the opportunistic arbitrage spotter who’s buying up as many properties as they can to flip them into Airbnb properties to make a quick buck because hey: who wants to have to be a community manager and a social host instead of someone who’s just managing a portfolio of properties that fill themselves with pliant guests looking for a cheap place to stay in town when the hotels are doing a terrible job?

In other words: Airbnb believes that we should all be much nicer to each other, we should be happy to accept each other as guests in our neighbourhoods and we should be grateful to Airbnb for pointing out that until this moment when we graduated wholesale to accepting the sharing economy, we were living as savage brutes with no reason to be nice to each other or to introduce people to what’s around us.

This is pretty tangential, but what irks me about the way in which Uber talks about the Sharing Economy and how they’re playing such a large part in it is that most of the analysis revolves around their play in the Moves Everything  market in much the same way that Amazon is focussed on the Sells Everything market. Uber’s transcendence to the Moves Everything market relies upon a short-term play in the Share movement (and less so the sharing and more so the service infrastructure that allows them to manage demand), but at the same time, the vast amounts of capital now available to them are allowing them to implement a pretty classical price undercutting in all the markets that they care about. UberX is just going to get cheaper until it doesn’t have to anymore – presumably at which points the medallians, if they even still exist, will become worth something quite different.

No, Uber’s valuation should be more than the $18bn and the upside is stupendously massive: but let’s be clear, in the potential markets for Uber addressed in[3], these aren’t necessarily “sharing” models as much as they are rent-by-use models. It might feel like a linguistic sleight-of-hand, but you share things that you use. The other things that we talk about sharing – that are open to all – are known as *utilities*.

[1] New Logo and Identity for Airbnb by DesignStudio
[2] Belong Anywhere – Airbnb’s new mark and identity
[3] How to Miss by a Mile: An Alternative Look at Uber’s Potential Market Size

2.0 Snow Crashing (9)

We’ve hit chapter seven in our journey through Neal Stephenson’s Snow Crashing and this time we find ourselves in the Black Sun, in the Metaverse. (Last time, Y.T. had just ended up in The Clink due to inadequate provision for female guests at The Hoosegaw).

Stephenson is really going for the VR concept here with the Black Sun. We’ve got space (the Black Sun, a bar, is as big as a couple of football fields so curiously even in virtual space we’re still measuring distances with real-space sportsball terms), but just to fuck with us a bit, the tables are hovering in the air because why bother drawing legs when you’re simulating gravity in the first place. Everything in the Black Sun is matte black – because hey, why bother tracing out where all those rays are going to go, and also, I’m not sure if anyone’s actually done this before, but I bet it looks really weird and potentially nausea inducing. Readers with an Oculus Rift feel free to rez in a Black Sun-like environment and tell me if being surrounded by a mass of matte-black light absorbing material is indeed weird.

The Black Sun’s environment appears to be the equivalent of a mod – different rules apply here, so you’ve got collision detection and clipping; “only so many people can be here at once, and they can’t walk through each other. Everything is solid and opaque and realistic” – to the extent, of course, that the real world isn’t composed of solely matte surfaced objects that rest solidly in the air with no visible means of support. But yeah.

We get an introduction to the term Daemon – you’re reading this on something that has a bunch of them operating in the background right now, and if not right now, then they’ll wake up for a couple of microseconds and do some stuff and then go back to sleep. Don’t worry: they’re there. Anyway: now that we have a different sort of user interface, here’s an opportunity for us to treat daemons as “[serving] imaginary drinks to the patrons and run little errands for people”. I’m not sure if I’d still call these things daemons these days rather than scripted behaviour – but regardless, these are bits of pre-programmed behaviour, sometimes at the behest of a user.

We also get reference to cartoon-physics. Da5id has modded the environment so that people can get “hit on the head with giant mallets or crushed under plummeting safes before they’re ejected” which whilst not entirely describing cartoon physics the way we’re used to it now (ie exaggerated scroll and bounce, rubber banding) the way that Stephenson describes that cartoon physics evokes the same kind of effect – we expect people hit on the head with a giant mallet to comically compress and rebound before being thrown out, or for the safes that are dropped on them to amusingly squish them rather than implement a more accurate (and potentially computationally expensive) simulation of what happens when an avatar body gets hit with a few hundred pounds of force.

There’s a throwaway line here – that if your personal computer is infected with viruses “and attempts to spread them via the Black Sun, you had better keep one eye on the ceiling”. We don’t really get to see what the virus/malware ridden world of the Metaverse looks like – there certainly aren’t that many backdoors and not really that much hacking that goes on (other than, you know, the giant bit of bio/informational hacking with the Snow Crash virus itself). But we don’t really get to see what type of viral behaviour a host might exist in a virtual environment, which would’ve been interesting.

We do get a little bit of hacking here though: Hiro invokes Bigboard – by mumbling the word, which seems a bit weird that he has to say it out loud – and it’s a backdoor exploit that essentially lists all the current users of the Black Sun and displays them on a flat map in front of him. Bigboard is just another reflection of how Stephenson is kind-of-but-not-really treating space in the Metaverse like space in the real world. With most contemporary massively multiplayer games that are situated in a virtual environment, we still get a sort of chat channel/party line that lists everyone who’s in the immediate environment – World of Warcraft comes to mind. In the virtual worlds that we’ve built in the intervening time since the release of Snow Crash, it feels rare that you have to eyeball all the player characters to find out which ones they are, especially when you’re talking about a social setting.

By scanning through BigBoard we see the social focus of the Black Sun: movies, music and the Japanese, with the odd smattering of People Who’re So Powerful They’re (Practically) Sovereign Nations. This is also the first time we see Sushi K, who’ll pop up later on.

When Hiro heads over to Da5id’s table, there’s a line that’s resonated with a whole bunch of people: “Softare development, like professional sports, has a way of making thirty-year-old men feel decrepit.”

Now we meet The Girl. I’m not quite sure if Juanita – portrayed in the Black Sun as a low-rez black and white avatar and talking to Da5id – ever does talk to another woman in the book about something other than another guy (it’s possible she talks to Y.T. later on) but at the very least, Juanita has her shit together. She has her shit together in the way that a lot of young male geeks feel older women have their shit together – she’s outrageously smart and intimidating to Hiro and that’s part of what attracts him to her. Juanita and Hiro met in freshman class at Berkeley, and later when they’re both working at Black Sun Systems, inc. Stephenson writes perhaps one of the most insightful pre-Reddit diagnoses of casual sexism to have been committed to page:

“It was, of course, nothing more than sexism, the especially virulent type espoused by male techies who sincerely believe they are too smart to be sexists.”

In Snow Crash, Hiro’s the one who has to do the growing up. Bouncing around as an army brat from base to base, it’s only when he revisits his memory of Juanita and how he reacted (or rather, didn’t) to her when he met her at school, and then years later, in employment, that he realises that he’s the one whose expectations needed recalibrating.

Anyway. Juanita has a painting of her late grandmother in her office at the Black Sun – where Hiro and Juanita have their real jobs – and we learn that Juanita had gotten pregnant at a young age, that her grandmother had been able to intuit and tell; to “condense fact from the vapour of nuance” and that Juanita’s big deal, her epiphany, was that whilst writing yet another query interface for a DoD grant she remembered that moment with her grandmother and tries to work out how to present information in the form of faces – because that’s what humans are wired to recognise.

There’s a difference, we learn, between Juanita and Hiro. Juanita and her family know where they are – where they stand, with “a certitude that bordered on dementia”. Hiro, by contrast, never knew – never knew if he was black or Asian or plain Army, never knew whether he was rich or poor, educated or ignorant, talented or lucky or even where his home was until he moved to California, which was “about as specific as saying that you live in the Northern Hemisphere”.

We learn a lot more about the world of Snow Crash. That Da5id’s parents were Russian Jews from Brooklyn, that Da5id went to Stanford and then started a company when he graduated “with about as much fuss as Hiro’s dad used to exhibit when opening a new P.O. box when they moved.” The rest of the sketch of Da5id’s character is Valley stereotype – he’s certain of everything, so much so that Juanita eventually divorces him and Hiro leaves the company. Juanita’s circumstances allow her to hang on and to allow her stock to vest and cash out so she’s rich now (how things never change), whilst Hiro cashed out early to put his mom in a nice community in Korea.

Hiro might not have a lot in the real world, but he has a lot in the Metaverse and he looked after his mom. And at the moment, he seems perfectly fine about that.

Part of what makes this interesting is the context – this is the vision of the future that we thought we were getting, and simultaneously, it’s more or less the future being built right now – or at least tested – by startups like Oculus Rift and products like Sony’s Project Morpheus. There’s a lot here that’s being learned in terms of what sort of environments make sense to our old, physically evolved brains. In terms of the people building Oculus Rift, I’m at roughly 30k feet right now, but my guess is that the ratios haven’t changed that much. Juanita’s an outlier at Rift, and her female colleagues are much more likely to be working in outreach, marketing and HR than in areas such as helping people understand the nuances of communication – somewhat ironic, for a virtual reality company bought by the world’s biggest social network.

3.0 Humans As A Service

Part of what I wrote in yesterday’s episode was about this particularly engineer-driven strain of wanting to map human relationships to existing data models so that they can be manipulated, transformed and acted upon without any of that messy human stuff getting in the way. A reader wrote that – if we’re being extreme – there’s a strain of the Californian Ideology where people aren’t just uninteresting but that they’re terrifying – and even worse, unmediated people, people in your face where you’re terrified of some sort of social faux-pas where you weren’t able to look something up in time or you’re just afraid of being *embarassed* is the absolute worst.

In some respects, a Brit can point fun at California for this. This California, simultaneously the product of some of the worlds most connective, empowering and dehumanising technology is also the same touchy-feely hippy California that encourages you to get in touch with your feelings, to hug it out, to have exported trust falls and corporate team bonding exercises (okay, so perhaps these are fake social bonding exercises) to the rest of the world. This is the California that also wants to use technology to mediate the human experience, to tidy it all up?

We know that images can be manipulated, so it’s a staple of science fiction to drive that to eleven and to imagine worlds where realtime video of ourselves can be amped up, calmed down, massaged to help us promote the view of us that we want the recipient to see. Think of it as a realtime curated Facebook video feed – the cleaned up, Photoshopped version of ourselves that appears in FaceTime calls.

With all of this it feels like there’s an index – a score that can be assigned to all these different mediative technologies to purport to make human communication more efficient but instead produce unintended effects that may well just get in the way. So Graph Search when it’s used to provide a query interface to “stuff people like” is only as good as the information that’s at the nodes of the graph, and provides a false sense of security to those who’re using it (if those who’re using it trust that whatever is in the graph is true to begin with – and I’m not that sure how many people treat Facebook that way). But then there’s always the opportunity for leakage – Graph Search is just the high tech version of playground ask Sue to ask Steve to ask Jen what she likes so Amy can get her something she likes for Christmas, but instead goes horribly wrong because Steve gets it wrong and Amy buys the wrong thing and then Jen won’t speak to Amy anymore.

Of course, the human intermediary service that we’ve all been using forever has been: humans. Humans turn out to be pretty good (well, the least worst option, I suppose) at being able to intuit the needs, desires and emotional axes upon which other humans work. At least, the good ones do. And the best humans are the ones that also do the best introductions – the super-connectors. LinkedIn is human super-connector as a service, seeking to emulate that one woman you know with the rolodex who knows anyone who’s worth knowing, who’s always eager to introduce you to someone who’ll help you with what you do.

But of course, the reason why she works so well in this thought experiment is because she takes the time to know you, and to know you intimately. She knows what you’re into and what you’re not into, mainly because she’s been evolved over millions of years – just like every other human being – to be really good at modelling humans. And she’s one of the best. And her models naturally include what the receptors and emitters that you have, and everyone else in her network. She doesn’t need to send you cards on your work anniversary – well, she might – but you can be pretty sure that she doesn’t ask *you* to send anyone cards on *their* work anniversary either.

The inverse here is that an engineer can look at this state of affairs and say: gosh, there’s a whole lot of implict information out here that’s being wasted and this is a terribly inefficient system relying on a bunch of super-connectors and, let’s be honest, a fair amount of random brownian motion in society. No, this needs a bit of optimisation and potentially the inefficient information transfer drag that’s been holding us back can perhaps give way for us to finally sublime and live among the stars.

I belong to a relatively high traffic private mailing list that’s been going on for at least a decade, I feel, before I joined it, and that was at least a decade ago. One of its axioms is that the noise is the signal: ie the random crap that people talk about, the play, the banter and the back and forth are just as valuable as the purported “signal” that it’s easy to try and look for. Instead, all of these weak ties – the chats about where you find a plumber or what best boiler to get (this was before, of course, sites like The Wirecutter produced articles SEO’d to within an sub-pixel of their rendertimes entitled “This Is The Next Boiler You Should Get”). It’s hard for the engineers and product managers who’re designing our human intermediary products to explicitly design for noise. Everytime they see noise, they see something absolutely terrible – something that until recently was useless and something that might as well have been thrown away but for stupendously cheap disk space, but now, thanks to Big Data, you can throw a few social scientists and data scientists at it and potentially publish what you’ve found in the micro-interactions of your millions of users.

Some of these desires – the airport example – are less business problems to be solved, and more examinations of how reality and social interactions work and a desire to remake them to be better for the end-user. I suppose you could equate them to, a long time ago, someone gazing longingly across a stream that needed to be forded to provide access to the green pastures across the way and deciding to go out and build a bridge. But, I’d contend: not really. For starters, there’s the my-problem-is-your-problem worldview, the view that making something work “better” or more efficiently for a certain class of people will result in equivalent gains for others. But the Airport Introduction example relies on asymmetries, too. The socially shy individual who uses the service to find interesting people to talk to can only use it if everyone else who’s potentially interesting is broadcasting their interests. In other words, if only all of that private data about someone – and I’m using the term “data” guardedly – was accessible in some way. To the entire world.

I’m not sure what kind of hubris this is. On the one hand, you can say that those who have low-level social interaction anxieties and problems are at least taking things into their own hands and trying to make the world function better for them. You could try to make an equivalence with physical accessibility, and say that the more of the physical world that is accessible to those who are less physically able, the better and richer our societies. And I expect that most people (libertarian crazies notwithstanding) would be OK with that. So: what of remaking the world to make it easier to use for those people who just don’t like talking to other people? And then what of remaking the world in a fashion, through network effects, that mean that, well, you don’t have much of a choice? If someone in the right position, at the right time, decides that we’d all be better off knowing who everyone in the airport terminal was and what their interests were, do we get an opt-out? Do we get a straight opt-out, or just one if we happen to not want to use one of the world’s largest communication networks?

I sympathise, I really do. There are a great number of times when I find it less stressful to insert a screen or some sort of communications medium between myself and the person I want to talk to or interact with. And I see how the route of technology has always been to take that which is difficult or hard and to make it easier to do. But perhaps “interacting with people” is not the kind of technological problem that needs to be solved in such specific ways.

Notes! I EAT THEM. Also I like to know who you are, so you should introduce yourselves.