Sunday, 13 November 2016

Black Mirror - Season 3 - Episode Reviews and Discussion

I really like Charlie Brooker's Black Mirror anthologies. It seems exceptionally rare to get high concept sci-fi TV shows these days (good old Star Trek and X-Files long gone). Sure, the sci-fi setting is the Hollywood big budget movie genre default, but those are entirely different beasts: all explosions, or at least relatively little risk in terms of imagination. While space sci-fi TV series are generally quite campy, fairly low concept or very strung out with filler.

Series 3 has been up exclusively on Netflix since 21 Oct 2016.
All the Black Mirror episodes are near future fiction, explicitly examining technologies that might well soon exist and how that could change societal realities. So are generally 'hard sci-fi'. Many can also be taken as commentary on present day phenomena, particularly social media, examining our new normals via the safety of satire. As such I consider the show to be really valuable for potentially catalysing public discourse and personal comprehension of our rapidly changing world.

So I'm gratified that the show has been a success. In fact the performance of the first 2 series on Netflix was so good it convinced the new (self styled) global TV channel/network to steal the show entirely away from it's Channel 4 home (in the UK), commissioning this new season as one of their first directly funded projects.

This season perhaps wasn't quite plumb the depths of distaste or bleakness previously achieved. There episodes range greatly in overall mood, as well as themes. Humor also sometimes interjected with the grim. So overall very entertaining.

Episodic breakdowns below: me opinions and discussion on themes and related issues, with some spoilers further through...

Episode 1 - "Nosedive"

In 2016 we already quantify how much we like everything online. Amazon and Ebay sellers don't make sales unless they the top ranked seller, so they plead for customer reviews and quirm painfully under any non 5-star feedback. We seek approval from others on social media, or at least measure the worth of our creative output by how many likes, +1s, or upvotes it gets. The prestige alone is very motivating. But then we've long had more substantial metrics attached to our person: credit worthiness ratings, and the biggest one of all - bank balance.

So it shouldn't take much imagination to consider the marginal step in technology and society that would create a reality where everyone has a rating floating next to them in augmented reality, such that one need not even evaluate their physical appearance, accent, etc, to get a precise read on their place in society. Then why not have these rankings determined by each other? After all it should encourage and reward being an upstanding member of society...

This piece explores some downsides of that hypothetical development, while lampooning familiar idiosyncrasies of our current time. Like sharing photos of carefully presented, supposedly casual little meals, with twee tag lines. The millions who follow the fake, staged, perfect life styles portrayed by Instagram gurus.
"You have got a solid popularity arc here..." - Screenshot montage.
But really, it stays quite safe, offsetting the dystopian nightmare of this menace by sticking with a surreally clean, modernist, pastel coloured theme throughout. The walled off area guarded by private police, accessible only to high rated individuals (seen towards the end), is a little nod to how dark our present already is. Arguably worst, with the great popularity of gated communities (for example, in the US); society is divided broadly by race and neighbourhood of upbringing.

The story arc spells out how easy it might be to lose this sci-fi social ranking, with little hiccups easily magnified and snowballing into potential disaster. The impossible difficulty of climbing back up only gently referencing our horrendous problems with low and falling social mobility - unprivileged virtually doomed from birth.

It's anecdote about a small difference in rank potentially making a direct difference between life and death is a little contrived (relying as it does on a cutting edge medical treatment, when this society appears pretty much steady state). Again, this portrayal a kind of watering down of what we already have, with a 'postcode lottery' determining if you are in an NHS authority that can afford to provide you with certain essential (but costly) treatments. And lifestyle disparities between the most and the least salubrious parts of the UK already splitting life expectancies by over a decade.

In this Black Mirror, it's your peers who rank you. Quite logically, the ratings from persons with higher ranking are more heavily weighted in determining your score. This speaks to elitism here, or the self serving (and preserving) oligopoly, clustered together into a tight knit bunch by their wealth status. But the episode doesn't go further to contemplate, for example, political control: would the higher ranked have directly higher weighted voting clout in democratic elections? It doesn't make any account of celebrities, with the ultimate power that might, perhaps, beget. We seem to see a huge 'first mover' advantage with snowballing YouTube subscribers, for example. And we've been seeing a resurgence of almost political populism reminiscent of fascism, in various nation's politics.

But worst still, what if a ranking system is not perfect and neutral, as depicted? Presumably any real world system might be hacked, with insane scope for blackmail, or simply dispensing with troublesome people. What if the system isn't made by some doe eyed Silicon Valley startup with good intentions, but instead an incumbent ruling system?

There's been coverage and concern about a proposed 'social credit system' in China [BBC, Independant], quite possibly a major basis for this episode.  It sounds like the gentle (but not so subtle) hand of big brother, plying citizens towards good, or at least desired, behaviour. It's easy to vent self righteously about the poor human rights over there, with intonations from our native media aiding bias, it's probably made to sound more sinister than the reality perceived over there. And as usual, focusing in on the nature of this kind of distraction may blind us to our own plight in the West, or soften us to seemingly lesser, similar intrusions.

As with the present, our future is unlikely to contain just one monolithic rating score, in some simplified totalitarian state. Even though dominant players are likely (like Facebook now), a schizophrenic thicket of independent, co-existing schemes seems likely. Regulating them all would be nigh on impossible, against the force of demand for such metrics. With the nuanced dangers they pose, might it even be be worth adding additional human rights to guard against them specifically...?

Also in the mix of this conceptually dense episode:

  • Obsolescence in electric car charging connectors (above right). Unlike the well established petrochemical fuel types, (high speed) electric car charging tech seems likely to be brand specific and evolve rapidly, so preserving backwards and lateral compatibility might well be a nightmare in coming decades.
  • Exploitatively over-personalised advertising (above left - this could be your life in this wonderful house). Currently, personalised advertising done well is mostly a good thing, bridging some of the work of trying to find what you actually want. But what if the meta-data on each person is so detailed and accurate as to be able to reliably push all their psychological buttons unbearably? Advertisers could virtually force them into buying things they can't necessarily afford. 

Episode 2 - "Playtest"

Not so much to read into this one. It's kind of a one-trick pony, reminiscent of a "Goosebumps" book, "The Outer Limits" episode, or such - horror with a cute twist. A little too much shouting from the loudmouth Yankee protagonist for me.

At a stretch, you might claim it's a preemptive cautionary tale, warning of the potentially high prices to be paid for failing to abide by seemingly inconsequential warnings when using technology of increasing sophistication and biological integration.

Episode 3 - "Shut Up and Dance"

Surf the net like everything you ever do there will be seen by the whole world. And always cover up you web cam! How many hundreds of thousands of people could be blackmailed into taking pretty crazy actions, without question, under threat of revealing activities they thought were a secret between them and their computer?

Here, the hacking/phishing attacks and ensuing activities could plausibly have been orchestrated by a small group, or perhaps a very talented individual (quickly purchasing stolen credentials from a dark web site). The online vigilante justice seems to be coming from a Reddit/anonymous/4chan kind of perspective (as suggested by the troll face message - right). Witch hunts are something Reddit mods try hard to avoid, although the crowd sourcing for real world intel has been helpful in the past.

This is a warning, perhaps, of some of the extreme powers that might easily be wielded over others (irresponsibly) by rogue information security experts. [Spoiler coming.] Also, a moral being: don't EVER look at kiddie-porn! The story is like "Inglorious basterds" (2009) with that long building F U to Nazis replaced by an extravagant treatment for viewers of paedophilic images. And don't bother giving in to blackmailers, it'll all come out eventually anyway (sucker!).

But what if it were a superintelligent AI pulling the strings? Imagine the kind of computer systems that routinely deal with millions of complex search requests in parallel everyday, and all those smart phone assistants, all coordinated to a single end goal. You wouldn't be talking about one isolated bank robbery, it could be all the banks, everywhere, all at once. With that much coordinated intelligence, it could easily shut down entire cities or entire country's infrastructure. Bringing it down simultaneously via 10 thousand little cuts, that no one perpetrator could have anticipated would have caused THAT much damage, alone. At the extreme, that could involve irreparable damage to all of civilisation, no need for nukes; much more elegant.

Vernor Vinge's "Rainbow's End" is about an AI pulling clever social manipulation shenanigans, much like this, but in only one location. Stross's "Halting State" involves use of augmented reality to manipulate unsuspecting gamers. We already have droves of people wandering around in odd patterns, searching for Pokemon out in the world.

But to play things down a little: once an intelligence tipped it's hand, it would presumably stand a high risk of being shut off. And it's vanishingly unlikely a single, isolated super-intelligence will appear on it's own, out of nowhere. There should be a ramping up of incidents and a whole milieu of AI and counter AI action, with the thin end of these kind of attacks perpetrated merely by large teams of humans.

With an extreme AI-from-no-where case, regime change (of any/all countries) would be easy. Secretly rigging elections almost trivial. But big, co-ordinated (i.e. national) cyber forces could be doing similar things on a smaller, more ham fisted scale already. Mr Robot worked in a little of this
with the 'dark army' of Chinese hackers. But aside from the resulting scope of the grand actions undertaken in that show, the scale of the forces portrayed seemed way too small - big nations have tens of thousands of cyber security personnel. Particularly China with those policing it's great firewall. The acts that could be (probably are being) committed must already be a little mind boggling.

From CNN Money: 1/3 pro-Trump tweets are from bots.
It's already been kind of tense wondering what, if anything, was going to go down in conjunction with the US elections, with armies of sleeping Twitter bots poised. The active ones also playing a significant role skewing discourse (pictured, right). Legions of internet enabled webcams, routers and DVRs are now perpetually lurking, ready to systematically cripple the internet out of existence. Although thankfully they didn't make an appearance during voting, with limited attacks possibly coinciding with Julian Assange going dark and apparent fear for his wellbeing [BoingBoing]. I don't know if there's been any fiction that has really got to grips with the kind of cyber warfare that could break out in coming years... (Ops, I've gone tangential.)

Episode 4 - "San Junipero"

The most emotionally engaging of the episodes, it's pretty much a feel good story, with a refreshingly optimistic tint on futurism. A wonderfully unlikely love story... But I'll say no more because it's a really pleasing one to watch. Just give it a chance, because the sci-fi elements are extremely subtle to start with. A slower boiling mood piece.  Just enjoy the ride.

It has Mackenzie Davis, who appears to have been typecast as an 80s era geeky chick from South West USA (as in the 3 excellent series of "Halt and Catch Fire", so far).

The moral hazard here, if any, is that technology may well become so wonderful, so quickly, that legislation will be a little too restrictive of it: real paradise exists, but dying elderly are only allowed 5 hours per week!

 The elderly angle is reminiscent of the groups sharing dreams in "Inception", who spend much of their time 'asleep' together, and consider that second reality more compelling and real. Of course, in focusing very narrowly on the old and dying, this story glosses over the impact of a virtual heaven on earth would have on those who are supposed to be economically active. Apparently they still have jobs to do, like looking after the old folks, at the least.

Episode 5 - "Men Against Fire"

Yet another one featuring augmented reality. Although this time it goes deeper than smart contact lenses. Given that most new technologies tend to get used for porn or war, it's not unreasonable to contemplate how such tech might be used to help people kill other people. (The virtual sex gets dropped in as an implied part of that too.)

Of course, one prominent way that modern military has been doing that is via drone strikes. Although even that can be traumatic and problematic for the remote pilots, as depicted in "Eye in the Sky" (2016). My bet is generally that killer robots would be prevalent before this level of finesse is applied to multi-sensory implants in humans. But there are reasons and scenarios where such things might be used after that, regardless (like knives have never become entirely obsolete, for example).

It kind of follows the current military tech trajectory, where advances have served only to protect the friendly forces, with minimal consideration of collateral damage and no real attempt to avoid killing in the first place.

The episode is named for Samuel Lyman Atwood Marshall's 1947 book "Men Against Fire: The Problem of Battle Command". It apparently touted a WWII figure of "75% never fired at the enemy for the purpose of killing" [Wikipedia], which is closely echoed here. He supposedly found close to 100% fired in vietnam, too. But then, apparently he may have been fabricating his military analysis, so perhaps men (and women) aren't as adverse to killing other people as the comforting tale laid out in this fiction...
Women totally fine with fire!
(Madeline Brewer was also in the excellent
 season 1 of "Orange is the new Black".)

The episode is also great for equal rights, with a convincing female commander and friend/antagonist soldier. Good to show that women are just as vulnerable to be pressed into state approved murder.

There's the ugliness of manufactured consent. The army is sticking to the letter of the law in requiring verbal or signed consent from their soldiers, but they are either totally ignorant of what they are letting themselves in for, or psychologically tortured, blackmailed into it.

The motive for the genocide is genetic cleansing, like Nazi ethnic cleansing, but supposedly based on firmer, disease and behaviour metrics. But reality check: besides a few rare diseases, genetics tend to account for little more than 10% of disease incidence, the rest being environmental. Also, it's not unreasonable to think that there might be ways to simply fix these bad genes in vivo, given the advances that are currently being made with the CRISPA system, already.

But then I guess it would be a little too political to use typical real world reasons for mass murder, which tend to come down to arbitrary basis of race, nationality or religion. Having all the 'roaches' being, say, Muslims in a United States of President Trump...

Episode 6 - "Hated in the Nation"

At 90 minutes, this is basically a fully fledged feature film. Although on a modest budget of $40M from Netflix for 12 episodes [HollywoodReporter], this piece almost manages to pull off cinematic scope. The narrative has sweep and cadence, although it feels lost at the end and casting is stretched thin in places. For example, when one would be expecting swarms of cops and special forces all over the place there's just 5 actors arguing in a computer suite (with poor scripting).

Some of the opening shots, with the mechanoid bees, had me thinking of the book cover of Iain M Bank's "The State of the Art" (below). Then the police action itself, with it's info-tech lilt, started evoking (for me) Charles Stross's near future Scotland in "Halting State" or "Rule 34". In the latter novel, police are having to protect the public from dangerous online memes (or here, trending hashtags). Also the little detail of the self driving 4x4 chimed, for me, with a little scene in the first book (I think).
Right - Rectangular honey from these mechanical bees? (Nope.)
This episode seems to be the anthology's linchpin, tying several of the other stories into the same universe, via subtle little news snippets, ticker scrollers and a name drop (all of which I missed, heh). So maybe we'll see some of these characters again. This was only the first half dozen of the commissioned episodes, with another run (season 4) dues next year (2017). Thankfully not as long as the 2 years since the one-off "White Christmas Special" [mentioned in this previous blog].

The main plot mechanic here required a sizable pinch of salt from the viewer - solar powered, killer robot pollinators, burrowing into people's brains. Yeah...  And they're entirely self replicating and can send video feeds back to base while doing (on board?) facial recognition. Maybe some kind of new meta-material optics might make that last physically plausible. While obviously their surface area would provide insufficient light collection area to power flight.

The technology is way out of step with the nearly contemporary setting and seems further ahead than most the other tech we've seen in other episodes. I mean, the soldiers (whose's story seems to tie in later, chronologically) were using big, fat quadcopters for surveillance. Why not a canister full of these magic solar powered bees?

But it's not worth dwelling on that, since it's just a way to tie in the plot while giving a hat tip to the currently pressing decline of our pollinator species.  Although, the script repeatedly failed to convey the need for these cyber-bees as anything other than saving our eco-systems. Sure, that might be the public's current perception of this news, for a large part, but the core issue is that we'd be unable to grow most of the food crops that we currently rely on eating daily. It's actually a pretty terrifying slow motion disaster that well under way. It's largely because farms aren't eco-systems (with their barren expanses of mono-crops).

The core issue explored in episode 6 relates to contemporary troubles with tidal waves of online abuse that can crash down on relatively unsuspecting members of the public when a piece of media shows them (seemingly) doing or saying something both reprehensible and shareably poignant (such that it goes viral). Swarms of apparently humourless twitter hooligans then brashly spouting hatred and death threats at people they know only one (probably skewed) piece of information about. With no regard for the cumulative effect that might have on them.

As it happens, the (pretend) urination on a war memorial (featured in this episode) is based closely on a real event that I blogged about 6 years ago (2010). It received excessive media coverage due (I thought) to memetic selection. So too the 'purrminator' lady who randomly threw a cat in a wheelie bin.

Disproportionately vindictive retribution seems to be a leitmotif for Black Mirror - the unquenchable appetite of public vitriol. Pretty pertinent, given the appalling (and perhaps soon to be disastrous) US presidential election. With the population so polarised as to exist in two separate (filter bubbles).

What if that digital hate mail were to quite literally kill the recipient? I'd like to think that the vast majority of real users would refrain from using the particular "#DeathTo" tag, if they knew the consequences. Especially if it were made illegal and punishable. (Although I'm not sure how many Twitter accounts are untraceable to their owner...)

A literal cull of all the worst trolls out there. A bit heavy handed... But it's a somewhat interesting thought experiment: what would it take to tame that smothering tide of generally casual but crushing hatred?

If (or rather, when) Turing level AI can be widely deployed, the social internet could be vested with a kind of abuse storm immune system. I imagine either (undercover) vigilante-bots going after trolls, mirroring their hate back at them - instant karma. (Tthat would be pretty dismayingly censoral. I can imagine such tactics being deployed already by those controlling bots, to overwhelm those they dislike/disagree with.)

Preferably there would be a more nuanced, friendly psychological approach, aiming to diffuse or even council trolls. Blizzard's Overwatch recently started to filter out certain choice chat insults, replacing the posted text with amusing self-insults. A fun little joke, really, but perhaps enough to help would-be casual abusers to stop and think. It only catches a few specific phrases (e.g. "gg ez"), and although a decent future AI might be able to spot most slights, blanket censorship (of anything too critical) would be abhorrent.

The theme of government mass data collection gone wrong is a good and topical, too, given such legislation as that from Prime Minister May (from being home secretary): the "Investigatory Powers Bill" AKA "Snoopers Charter". Extending or legitimising our state run trawling of supposedly private communications and information, mandating ISPs track and record all websites visited by their users, etc. This is both cumbersome for the IT businesses and creates security vulnerabilities by having these storage repositories. Such big, juicy cachet's of the kind of private info those schemes dredge up are going to be prime targets for powerful enemy forces (foreign nation actors, criminal groups, etc). And government are rarely the best at securing computer systems themselves.

In this episode's plot it show how seemingly anonymous metadata might quite easily be matched up with real identities, en-mass.

Everything online seems to be more vulnerable now days; just a matter of time until any given account password gets stolen. (Biggest data breaches to date.) Of course, it doesn't help that the biggest spy agencies in the world have been deliberately inserting vulnerabilities into communications hardware and protocols so that they might peak in (opening the same holes to any old hacker). As revealed by some of Snowden's leaks.

But the scene dealing with this topic (above) was terribly awkward in terms of writing/acting/direction/production. There was an ill fitted assortment of swear words flopped out amid stilted clichés - pretty much just chewing out The Man for spying on everyone. There, and throughout, Benedict Wong felt the most awkward in terms of acting/casting - unconvincing and spineless as a lone representative of the spook agency. It's also a little shocking how much he appears to have aged (and gained weight) since he last appeared alongside Kelly Macdonald in "State of Play" (2003).

At 40 years old, Kelly herself has arguably aged too well for her role as a world weary police inspector. She barely looks that much older than her protegé, played by Faye Marsay (30), whom I'd not clocked at all, from her Game of Thrones role as the "Waife". (But then that shouldn't be surprising, since she's a master of disguise, heh.) But it is refreshing to have a feminine female lead in the Cracker-esq, detective role. And I was mostly convinced with her, to be honest. I've enjoyed the feel of the reversed gender biases in this series collection. Here a strong female as the main support too. Although she came out less convincingly written, with too much smug showing off by way of randomly dropping technical knowledge that was actually just vague prattle. Scripted as it was for the purpose of light info-dumps.

I was not enamoured with the closing out of the episode: following the perpetrator didn't really feel like it tied in. Seemingly fairly inconsequential, with the action having wrapped up, and it didn't even properly foreclose then. The only thing it seemed to be saying was that females can not only be technical wizards, but also take on crime fighting roles to ultra-dogged extreme, going rogue bad-ass to get justice.

Maybe the story will pick up in a later episode, or at least some of the characters will. I think I read rumours of such possibilities, and these police characters are perfectly setup to be involve in some other completely different plot (unlike many of the other central protagonists, who were consumed by their events, so to speak).

What I'd love to see in a future Black Mirror would be an intellectual treatment of mass unemployment from robots and AI, basic income, etc. We may have to wait another year to see what's in store. Although, in that time, a whole swathe of fascinatingly scary real world events are likely to have unfolded too, in the aftermath of the US election and with the playing out of Brexit (or perhaps not Brexit)...

No comments :

Post a Comment