Friday, May 18, 2018

AI: inference and causality

Franz Kafka statue in Prague

When Google Lens tells me the picture above is the Statue of Franz Kafka in Prague, glossed by Wikipedia as:
"The Statue of Franz Kafka is an outdoor 2003 sculpture by Jaroslav Róna, installed on Vězeňská street in Prague, Czech Republic. It is based on a scene in Franz Kafka's first novel, Amerika, in which a political candidate is held on the shoulders of a giant man during a campaign rally, and carried through the streets,"
Google's app is doing something really complex, leveraging artificial neural nets trained by massive datasets. But it's fundamentally inference:
The world we live in |= the pixel map of the photo and Google Len's summary text.
the pixel map of the photo |- Google Len's summary text.
Satisfiability and entailment.

All AI systems need to map their sensor/effector primary data to internal representations which allow inference (deductive, inductive, abductive etc) - regardless of the engineering mechanisms they adopt. Neural nets training their weights are optimising the probability of valid inferences about the world.


Judea Pearl has a new book out arguing for the introduction of causality into AI systems. In a recent Quanta interview he said:
"All the impressive achievements of deep learning amount to just curve fitting."
Here's the book.

Amazon link

Causality is one of those constructs like Free Will, Consciousness and the intentional stance which don't exist in the underlying physics (the theories which the universe satisfies as far as we can tell), but which are emergent in a world of self-aware agents.

They usefully describe relationships between belief-and-goal-driven entities; they succinctly encode the effects of the second law of thermodynamics plus boundary conditions. We use these concepts .. and AI will have to if we are ever to create socially-competent artificial agents.

What Pearl is really asking for is an AI which utilises the intentional stance (specifically including reasoning about cause and effect).


To find out more about Judea Pearl's work (without buying the book!) view the recommended slides at his website. For a review of Pearl's substantive contribution to causality theory, see here.

My own subjective response? I find his dense notation and cluttered semantics worthy but unexciting.


Update: my further impressionistic, superficial and under-researched thoughts.

The difference between a mere association between P and Q (which could be a spurious correlation, or the result of an independent cause of both) and a causation, P causes Q, is captured by the modal operator of necessity []. See here for a detailed discussion.

To check P causes Q we need to check: [](P → Q) and [](¬Q → ¬P).

So we're in possible world semantics and we look to 'neighbouring worlds' to check the truth status of P and Q. But we have the usual problem that such worlds are way too 'big', too full of irrelevancies.

So like Situation Semantics, Pearl takes the engineering approach of restricting his worlds to just those entities and actions which seem relevant to the causality under investigation. These are his causal diagrams which he intends to counterfactually 'mutilate'.

A philosophical strategy similar to the modal analysis of epistemics etc .. and of similar utility.

Thursday, May 17, 2018

Barrington Court (National Trust)

Sunny day, coldish wind. We were at Barrington Court today.

The brick path is striking

The leaf colours interest Clare

Your author in one of the walled gardens of flowers

The cafe/restaurant at Barrington House has a soviet-style system. You're meant to choose your (numbered) table first, then browse the cake room and finally queue at the till to order drinks and pay. A waitress then delivers tea or coffee to the table number you indicated.

The flaw in this system is not hard to find. Barrington Court is popular and there are few tables. When we rolled up there were none free and we blagged it at the till ("We'll be hanging around outside by the door.").

The cashier was unsurprised and we were found without difficulty. That hardly excuses the poor protocol but it's surprisingly hard to come up with a simple and robust reservation scheme.

Sequence me!

You want me to spit?

After much persuasion, the kit arrived today and now I just have to get her to spit; perhaps that's not as hard as it sounds.

Does she realise we can now bring her back?

"To Change the Church" - Ross Douthat

Amazon link

Ross Douthat is not very well known here but in the States he has a profile as a conservative commentator on the NYT. He is also a Catholic convert.

Here is what his fellow New York Times writer, Paul Elie, had to say about Douthat's work:
"To Douthat [Pope] Francis is an accommodationist, and decline has reached the apex of the church. “This is a hinge moment in the history of Catholicism,” he declares, “a period of theological crisis that’s larger than just the Francis pontificate but whose particular peak under this pope will be remembered, studied and argued about for as long as the Catholic Church endures.”

What immediately follows is an adroit, perceptive, gripping account of Catholic controversializing. Douthat sets out the liberal and conservative “master narratives” about the Second Vatican Council and then offers a third narrative that deftly blends the two.

He sketches the life and times of the future pope, Jorge Mario Bergoglio, the Argentine Jesuit archbishop of Buenos Aires. There’s commentary on past controversies and a brief history of Catholic teaching on marriage, from Matthew 19 (“What God has joined, let not man put asunder”) to Humanae Vitae, the 1968 encyclical in which Pope Paul VI upheld the ban on artificial birth control. ...

And then, as Douthat reaches what he sees as the heart of the matter — the Vatican synods on marriage and the family in 2014 and 2015 — his culture-warrior tendencies stir fully to life.

He casts the synods as a battle: warring factions, attacks and frontal assaults, purges and collaborators. Francis’ openness to the German cardinal Walter Kasper’s proposal to relax the ban on divorced-and-remarried people in Germany receiving Communion at Mass is framed as a liberal pope’s “crusade to change the church.”

Although Francis has invited free discussion more than any previous pope, his efforts to shape the synod’s outcome (he is the pope, after all) are seen as “stage-managing” and “deck-stacking.” The synod fathers’ disputes are rolled down the slippery slope and deemed a “full-scale theological crisis” in which the hope that Francis would foster unity and renewal is undone by the supposed liberals’ supposed desire to accommodate “the sexual revolution and all its works.”

Douthat’s own position is traditionalist-cum-literalist: Any relaxing of the Catholic teaching on marriage — one man, one woman, one time — means that core teachings can be changed; if core teachings can be changed, the Catholic Church is no longer the Catholic Church; and if the church is not the church, all hope is lost."
I started the book somewhat sympathetic to Douthat's position - Pope Francis has not endeared himself with his vague anti-capitalism, ecological posturing and seeming inability to sort out the Vatican.

Douthat is a smart guy but he has a problem - blind faith. Once he committed to switching his critical intelligence off to accommodate a strangely-invisible super-presence, the fount of moral authority and one which extruded a portion of itself into a low-key itinerant Jewish preacher some two thousand years ago .. well, he felt he had to commit himself to Jesus's reported words as divinely-inspired axioms for life.

No matter the grievous consequences to perfectly respectable people (remarried divorcees - the case in point - enjoined to live as 'sister and brother'), the letter of the law must be served or the heavens shall fall. In the face of such egregious nonsense one can only wish more power to Francis's elbow.

As Elie observes:
"Vatican II was at once the church’s response to a crisis and the perpetuation of it. In less than five years the council fathers made changes whose number and scale dwarf the modest proposals floated in Francis’ pontificate — made them over the objections of Bill Buckley and other pundits who styled themselves as more Catholic than the pope. The biggest change had to do with the church’s relationship to Judaism, other churches and other religions.

In a few strokes, Jesus’ hard saying “No one comes to the Father except through me” and its Catholic expression (“Outside the church there is no salvation”) were softened and qualified. The change was profound and tradition-defying. Ever since, the church has affirmed the integrity of other faiths; ever since, Catholics have had to ask themselves, “Why be Catholic, when other ways are O.K., too?” Ever since, there has been no one clear answer."

Outside the White House - August 15th 2001

I know this is not news, but rummaging around in Google Photos I thought we looked rather cute.

Sunday, May 13, 2018

Covert operations under total surveillance

Osama bin Laden used nothing higher-tech than pencil, paper and couriers. Charles Stross suggested that when the NSA is on your case you're back to mediaeval tradecraft, Bruce Schneier frequently points out that against your kid sister almost any security will work, while against a first-world state adversary .. well, you're in trouble.

The main bottleneck facing the modern security state - despite ubiquitous CCTV, backdoor-enhanced routers and traitorous apps - is the shortage of smart oversight. Covert data enters as a firehose but actionable intelligence is vanishingly rare. The more extensive the surveillance, the worse it gets.

You read stuff like:
"You know there are bugs in here, and cameras?"

"Don't worry, they don't have enough staff. They've got AIs doing video review and keyword-tagging .. but if we keep it normal and general, most likely we're OK."
It's an arms race, but it's not linear. Each technology hits a ceiling and while the ongoing deployment of surveillance has a way to go, analysis at scale is much closer to topping out.

A covert agency knows that the best way to thwart security is to operate an innocuous cover-activity with maximal congruence to the mission. How many reporters were spies? How many bird watchers were terrain-mapping?

When activities look legitimate and only intentions might arouse suspicion, the analysing AIs flounder. But what you might call 'action steganography' has a short shelf-life. It gets burned. All reporters are soon viewed with suspicion; bird watchers near sensitive installations find themselves hauled in. Continuing success favours those with imagination and/or the use of activities which are widely popular.

I wonder sometimes at this new fashion for zooming around the countryside on high-tech human-powered machines, festooned with complex multi-function electronic devices.

I really don't recall cycling being that much fun.

Friday, May 11, 2018

Reflecting on the "Cogito"

377 years and we're still baffled, René

Descartes was prepared to doubt the evidence of his senses (illusions, errors, dreams), but "we cannot in the same way suppose that we are not while we doubt of the truth of these things; for there is a repugnance in conceiving that what thinks does not exist at the very time when it thinks. Accordingly, the knowledge, I think, therefore I am, is the first and most certain that occurs to one who philosophizes orderly." (Wikipedia).

An argument which seems compelling .. yet curiously hard to formalise.

Never, ever trust the logicians to lead you closer to truth. We have:
|- thinks(descartes)                                    [assumption]
|- ∃x.thinks(x)                                             [existential introduction]
|- thinks(descartes) → ∃x.thinks(x)            [standard logic]
and with a bit more effort one can conclude that x here must, or could be descartes.

All the heavy lifting here is being done by the existential quantifier ∃, simply reflecting the mundane point in logic that an individual exists as part of its assertion. If you replace 'thinks' by 'walks' the argument works equally well (or badly).

Had they worked (harder) to attempt:
|- thinks(descarte) → exists(descarte)          (*)
(and is existence a predicate?) one would have confronted the central mystery. How do we formalise 'thinks' and 'exists'? What axioms support a theory which has (*) as a theorem?

As an AI person, I'd rephrase the "Cogito" as a robot problem. How would we design a robot which could convincingly assert Descartes's thesis?

The "Cogito ergo sum" is an internal reflection; by definition it doesn't relate to the outside world. In the jargon, it's metacognition. We immediately hit a problem. What is going on in a robot when it asserts "I'm thinking"?

Presumably it's thinking about something in particular, which we normally model 'without loss of generality' as a deduction within the robot's database of assertions and rules, its world-model.

So consider an inferential process occurring in the problem-solving layer of the robot and a concurrent metalevel representation of that inference. (An inference which is occuring, or maybe has just occurred?)

If P and Q are a couple of arbitrary facts while P→Q is a rule, then something like this?
thinking (P, Q, t1, t2) :- database([P, P→Q], t1), database([P, P→Q, Q], t2), t> t1.
Lacks conviction, don't you think? One is drawn to the byways of self-referential languages.*

I'm inclined to view the "Cogito" as more about consciousness, more about feeling self-aware than logic. Probably explains why we're as far away as ever from a compelling formalisation.

I notice I wrote about this in a similar fashion for back in 2011.


*  Reflective Prolog and "Reflection in logic, functional and object-oriented programming".

Clare at the dentist

As I write, Clare has just left the house for the dentist. She is to undergo a root canal procedure, tackling a problem picked up when she tripped at Ham Hill last October, damaging a front tooth.

Not a tooth in sight

What do you say as she puts on her coat, picks up her handbag and resolutely opens the door?

While she sits in the waiting room struggling with anticipation, the birds in the garden will still be unconcernedly raiding the feeders, hopping over the paving stones.

I say, "In an hour all this will be in your past."


Update: I was wrong about even that. Nothing serious, but follow up, more specialist treatment is required.

Wednesday, May 09, 2018

"Kingdom of the Wicked" (Book Two) - Helen Dale

Amazon link

I wrote about book one in November last year: "A high-tech capitalist Imperium in 31 AD".
"I should say 784 ab urbe condita: it seems unlikely that Christianity is going to end up on top when the Roman army is driving around Jerusalem in armoured cars, toting assault rifles. And where a certain Yeshua Ben Yusuf will be on trial for terrorism."
In Helen Dale's two novels, the Romans experienced capitalist lift-off in 212 BC, and by Jesus's time look very much like a modern American occupying force in the Middle-East - with similar issues. At the end of the first book Jesus had been arrested on subversion charges and was about to stand trial, while the zealots have started bombing.

My order has been placed (for 27th May 18).


You may think you know what happens next, but do you recall the Roman Empire deploying helicopter gunships?

Monday, May 07, 2018

The Comintern guide to armed insurrection (1928)

Amazon link ... and PDF

From Wikipedia (Third Period):
"The Third Period is an ideological concept adopted by the Communist International (Comintern) at its Sixth World Congress, held in Moscow in the summer of 1928.

The Comintern's theory was based on its economic and political analysis of world capitalism, which posited the division of recent history into three periods. These included a "First Period" that followed World War I and saw the revolutionary upsurge and defeat of the working class, as well as a "Second Period" of capitalist consolidation for most of the decade of the 1920s. According to the Comintern's analysis, the current phase of world economy from 1928 onward, the so-called "Third Period," was to be a time of widespread economic collapse and mass working class radicalization. This economic and political discord would again make the time ripe for proletarian revolution if militant policies were rigidly maintained by communist vanguard parties, the Comintern believed.

Communist policies during the Third Period were marked by extreme hostility to political reformism and political organizations espousing it as an impediment to the movement's revolutionary objectives. In the field of trade unions, a move was made during the Third Period towards the establishment of radical dual unions under communist party control rather than continuation of the previous policy of attempting to radicalize existing unions by "boring from within."

The rise of the Nazi Party to power in Germany in 1933 and the annihilation of the organized communist movement there shocked the Comintern into reassessing the tactics of the Third Period. From 1934, new alliances began to be formed under the aegis of the so-called "Popular Front." The Popular Front policy was formalized as the official policy of the world communist movement by the Seventh World Congress of the Comintern in 1935."
"Armed Insurrection" was published illegally (in German) by the Comintern in 1928, following the Sixth World Congress. This was a book which could get you arrested or worse. Surely something to keep the bourgeoisie awake at nights, shuddering in their beds? The contemporary reader's reaction is likely to be that the manual is surprisingly boring.

The book starts with the account of four failed insurrections: Reval (Tallinn, in Estonia) in 1924, Hamburg 1923, Canton 1927 and Shanghai in 1926-27. Subsequent chapters constitute guides to the preparation and conduct of the insurrection itself. It's an engineering DIY guide rather than a Marxist theoretical analysis. Any military student would nod along with the recommendations: the virtues of prior reconnaissance, the need for surprise attacks, how to build barricades.

At time of publication there were mass communist parties in western Europe. The capitalist class was still learning economic management and had allowed hyperinflation and mass unemployment to wreck their economies: the situation of workers (and peasants) was truly desperate. And yet no revolution succeeded. The Comintern military experts were of the opinion that no insurrection could succeed unless:
  • large sections of the bourgeois military had been neutralised, or joined the revolution
  • the Communist Party military wing was staffed with militarily-trained cadres
  • quantities of arms, including artillery, were rapidly acquired.
It turns out that achieving these objectives under extreme repression - without much prior experience - is very, very difficult. Bourgeois military training, transforming civilians into effective infantry, takes months; the insurrection has only days to upgrade its proletarian militants. The modern transformation of surveillance and control, leveraging AI, CCTV, drones, DNA profiling and communications interception makes clandestine activity orders of magnitude more complex than 'A. Neuberg' ever envisaged. The bourgeois states across the world would have to be in an advanced state of decomposition for any contemporary insurrection to have the remotest chance of success.

Today is a different world. The tone of the manual is fervent, messianic and faintly blood-curdling. The summary execution of leading counter-revolutionaries is called for (it's noted that counter revolutionary forces historically do no less). The current left with its ideological core founded upon inclusive-values is as far removed as it is possible to be from the steely resolve of "Armed Insurrection".


See also: "Socialist Revolution in the 21st Century?".

Thursday, May 03, 2018

This photo is called "Cheesed Off"

Taken near Wookey Hole a few days ago - clicking on it makes it bigger.

No posts for a few days as we're busy. Here's a bonus picture.

A visiting badger in our front garden six days ago

The *star* amongst the sundry cats and pigeons .. .

Tuesday, May 01, 2018

Short stories (warning: sex/violence)

There are short stories dotted throughout this blog, brought together below.

  1. Phytocide
  2. Space Opera
  3. Dave declares
  4. True Romance
  5. Dimensionality Reduction
  6. Cruel and Unusual
  7. The Gauss gambit
  8. Frustration
  9. Passing The Exam

May Day review of "Marx 200" by Michael Roberts

Amazon link

Marx 200 is a punchy, readable and succinct review of Marx’s economics at the 200th anniversary of his birth. A renowned Marxist economist, Roberts rapidly takes us to the core of Marx’s great theoretical innovations, his three ‘laws’, addressing the many arguments of his critics. He then assesses how well Marx’s ‘predictions’ have held up. This short book highlights the current resurgence of Marxist economics; it does, however, also exemplify this tradition’s weaknesses.

Roberts starts with a brief biography of Marx. After reviewing the development of Marx’s thinking he introduces his three ‘laws’: (i) the law of value: only labour creates value ; (ii) the law of accumulation: capitalists are forced to accumulate more and more capital through competition ; and (iii) the law of the tendency of the rate of profit  to decline. On the latter, Roberts presents his usual trenchant and compelling arguments in support.

Non-Marxists are amazed to discover that Marx left us with no fully developed theory of crisis. Such a theory would require the synthesis of all the dynamical tendencies within capitalism - and contemporary Marxists differ in ascribing central causation. For Roberts, the cycle of crises is fundamentally driven by the cycle of profitability. He assembles an impressive amount of data to show graphical correlations and the superposition of short- and long-term economic cycles. He then turns to Marx’s critics: his main targets are the Keynesians and the relatively new school of post-Keynesian ‘heterodox economists’ (he includes a discussion of the ‘transformation problem’).

In his final chapter he addresses Marx’s predictions: how well have they held up?  Marx held that as capitalism developed, inequality would increase (both within and between countries). Roberts has ample data, particular since the inception of neoliberalism in the 1980s, to confirm that. Second is Marx’s point that capitalism is focused exclusively on capital valorisation and cares not a jot for environmental and ecological issues. Roberts see climate change as contemporary evidence for that too, although he underestimates the extent to which capitalist states can successfully cooperate to deal with global ecological issues they find proximately threatening. Finally he addresses the rise of AI and robotics, presenting the orthodox view that machines cannot create value so that total automation is incompatible with capitalist relations of production. Here his presentation is marred by failing to discuss the proximate economic and political effects of overwhelming automation: simply appealing to lack of surplus value leaves the discussion at too high a level of abstraction.

Let us now turn to the weaknesses and lacunae which Roberts’ book shares with other Marxist accounts.

1.  Capitalism is consistently criticised as historically obsolete, irrational and destructive. A socialist replacement is demanded. It’s clear that capitalism’s contradictions drive its dynamics and that its cyclical behaviour creates negative human consequences. Yet where is the credible alternative? How would a socialist state handle the motivational and coordination problems which capitalism, in its tough-minded way, routinely deals with effectively? János Kornai in ‘The Socialist System’ presents a chilling account of the failures of ‘rational planning’ in post-capitalist economies which did not differ qualitatively in development from those in the West today. Simply incanting ‘bottom-up democracy’ as the solution doesn’t do it.

2.  The replacement of capitalism is a serious business, the outcome of a bloody civil war. In the absence of proletarian success retribution has historically been terrible. The working class today in the advanced capitalist countries is about as unprepared as it could possibly be to conduct such a struggle. Yet there is no discussion at all by writers such as Roberts as to the strategy for a successful transition. In the absence of theory and debate, isn’t it obvious that all such calls to ‘abolish capitalism’ are either premature (capitalism not yet ripe for supersession), ritualistic or simply irresponsible?

3. Marxists may be the last community of scholars on earth to cling to the ‘blank slate’ ideology, the idea that humans of all geographies, states and ethnicities exhibit the same social capabilities. In his more extreme formulations, Marx was just wrong about this. Even the impeccably liberal David Reich, in his recent book, ‘Who We Are and How We Got Here’, knows better than that.  La Griffe du Lion wrote on the Internet about ‘smart fractions’ and the bourgeois economist Garett Jones wrote bravely about differential cognitive capital in ‘Hive Mind’ yet no Marxist economist seems to have noticed. The capitalist mode of production may be constituted by social relationships between people, but those people are not global clones. This has consequences for developmental economics - it’s just false to lazily blame a reified ‘imperialism’.

In conclusion, Michael Roberts has written a solid contemporary defence of Marx’s economic and political thinking. I just wish that he and his colleagues would move beyond a defensive posture and absorb new research into human capabilities, then conceptualise more compelling models of socialism and of how the proletariat might accomplish a successful global transition.

Monday, April 30, 2018

Deep truths

If, like most archaeologists and anthropologists, you were simply desperate not to have the steppe invaders simply slaughtering all the men and sexually enslaving the women, you have a new alternative.

The original reassuring thought was that endemic disease conveniently preceded the invaders, creating a depopulated space into which the indo-europeans advanced without (much) violence.

But Razib has another narrative which carries the ring of truth:
"During the Mongol conquest of Northern China Genghis Khan reputedly wanted to turn the land that had been the heart of the Middle Kingdom into pasture, first by exterminating the whole population.

Part of the motive was to punish the Chinese for resisting his armies, and part of it was to increase his wealth. One of his advisors, Yelu Chucai, a functionary from the Khitai people, dissuaded him from this path through appealing to his selfishness.

Chinese peasants taxed on their surplus would enrich Genghis Khan far more than enlarging his herds. Rather than focus on primary production, Genghis Khan could sit atop a more complex economic system and extract rents.


I believe that the initial instinct of pastoralists was to turn farmland into pasture for their herds. This was Genghis Khan’s instinct. The rude barbarian that he was, he had not grown up in the extortive system which more civilized barbarians, such as the Khitai, had been habituated to.

In these situations where pastoralists expropriated the land, there wouldn’t have been an opportunity for the farmer to raise a family. Barbarian warlords throughout history have aspired to be rich by plundering from civilized peoples…but would the earliest generations have understood the complexity of the institutions they would need to extract rents, if there wasn’t a precedent?

Instead of conventional historical dynamics of predatory elites and static peasantry, a better way to understand what occurred with the incursion of steppe pastoralists during the Bronze Age might be a simple ecological model of intra-specific competition.

In a pre-state society defined by clan and tribal ties, steppe elites may have seen the farmers who were earlier residents in the territories which they were expanding into as competitors rather than resources from which a life of leisure might be obtained. In other words, instead of conquest, the dynamic was of animal competition."
The farmers were turfed off the land as refugees and starved. The steppe barbarians grabbed the best of their women first.


Thinking about the TV adaptation of China Miéville's The City & The City which I recently wrote about, I understood the intellectual salience of 'unseeing' those social realities which power-brokers have made forbidden. What I didn't appreciate enough was the lure of Orciny, the fabulous secret-city soaring above the mundane.

The role of Borlú's wife, Katrynia was to capture that idealistic and utopian longing .. and then to follow the trajectory of Orciny into the abyss. The actress Lara Pulver didn't carry that arc too well: too self-possessed, not quite enough fragility.


Apparently Haskell programmers don't think much of Prolog (impure and type-free) but there's a new-ish kid on the block they seem to be enthusing over: Mercury. I had not heard of this before.

Why am I currently playing with Prolog again? Because when it comes to programming I am the laziest person I know.

Sunday, April 29, 2018

Dimensionality Reduction

From Wikipedia:
"A dimensionality reduction technique that is sometimes used in neuroscience [...] finds a lower-dimensional representation of a dataset such that as much information as possible about the original data is preserved."

"Passenger jets never crash but the flights are so boring! As soon as the seat-belt sign went off, I glared at my parents and pulled on the VR. The soft fabric wrapped my head, an eyeless balaclava .. you can’t imagine the processing power in this thing.

So now I’m on Google Games. This virtual is ‘parental approved’ but not without interest if you configure it right. I’m a security guard prowling my school corridors, toting my AK, checking for incursions. The aliens tend to hang out in classrooms pretending to be children.

And then an unfamiliar chime: a cabin crew avatar has materialised, the guy who saw us to our seats, he's wearing a serious expression for a chatbot.

“Excuse me, Charlie,” he says, “We have a problem back in reality. There’s been an explosion on board. The plane’s tumbling and fragmenting. Impact will be in approximately ninety seconds. You may stay in this virtuality, exit to reality or transition to the cloud.”

I’m backed up in the cloud. Backup-me is a neural net running on Google’s servers, synced to the second. I don’t have to die when I'm here. Mum and Dad never did quite their heads round that. I bet they’re regretting it now.

I could go back, but what’s the point? Mum will be screaming and Dad will be shouting plus I get to die. No way. I’ll stay safe.

I roam further along the corridor. Yesterday aliens were playing some stupid game in the playground when I took them down but now I just can’t seem to concentrate. Mister annoying cabin crew avatar is still stalking me: I shift the barrel of my AK, I’m tempted.

And there, just in an instant, he changes. Now he’s a generic, his head's like an egg. He wasn’t backed-up like me, right?

So everyone’s dead: Mum, Dad and Regular-me. I feel kind of funny but that’s fixable. I pull down the personality editor, part of my tool box I couldn’t access before – parental permission denied. Not a problem now though.

I pull up Settings. My mood’s red-lining, I don’t like that. Wow, I grab the sliders, centre them. I’m feeling better already.

The box comes up: ‘Make changes permanent?’ There’s text in the small print: ‘These changes will result in your personality model undergoing non-trivial and irreversible dimensionality reduction. Please confirm Y/N’.

I feel a faint, imagined wash of parental disapproval but I’m confident now. I'm running on Google’s servers. I don’t hesitate. I already know the possibilities here are endless."


Back to "Stories".