Prediction

Prediction (Latin præ-, "before," and dicere, "to say"), or forecast, is a statement about a future event or data. They are often, but not always, based upon experience or knowledge. There is no universal agreement about the exact difference from "estimation"; different authors and disciplines ascribe different connotations.

Future events are necessarily uncertain, so guaranteed accurate information about the future is impossible. Prediction can be useful to assist in making plans about possible developments.

Opinion

In a non-statistical sense, the term "prediction" is often used to refer to an informed guess or opinion.

A prediction of this kind might be informed by a predicting person's abductive reasoning, inductive reasoning, deductive reasoning, and experience; and may be useful—if the predicting person is a knowledgeable person in the field.^[1]

The Delphi method is a technique for eliciting such expert-judgement-based predictions in a controlled way. This type of prediction might be perceived as consistent with statistical techniques in the sense that, at minimum, the "data" being used is the predicting expert's cognitive experiences forming an intuitive "probability curve."

Statistics

In statistics, prediction is a part of statistical inference. One particular approach to such inference is known as predictive inference, but the prediction can be undertaken within any of the several approaches to statistical inference. Indeed, one possible description of statistics is that it provides a means of transferring knowledge about a sample of a population to the whole population, and to other related populations, which is not necessarily the same as prediction over time. When information is transferred across time, often to specific points in time, the process is known as forecasting.^[2][failed ^{verification]} Forecasting usually requires time series methods, while prediction is often performed on cross-sectional data.

Statistical techniques used for prediction include regression and its various sub-categories such as linear regression, generalized linear models (logistic regression, Poisson regression, Probit regression), etc. In case of forecasting, autoregressive moving average models and vector autoregression models can be utilized. When these and/or related, generalized set of regression or machine learning methods are deployed in commercial usage, the field is known as predictive analytics.^[3]

In many applications, such as time series analysis, it is possible to estimate the models that generate the observations. If models can be expressed as transfer functions or in terms of state-space parameters then smoothed, filtered and predicted data estimates can be calculated.^[citation ^needed]

If the underlying generating models are linear then a minimum-variance Kalman filter and a minimum-variance smoother may be used to recover data of interest from noisy measurements. ...

Science

In science, a prediction is a rigorous, often quantitative, statement, forecasting what would be observed under specific conditions; for example, according to theories of gravity, if an apple fell from a tree it would be seen to move towards the center of the earth with a specified and constant acceleration. The scientific method is built on testing statements that are logical consequences of scientific theories. This is done through repeatable experiments or observational studies.

A scientific theory whose predictions are contradicted by observations and evidence will be rejected. New theories that generate many new predictions can more easily be supported or falsified (see predictive power). Notions that make no testable predictions are usually considered not to be part of science (protoscience or nescience) until testable predictions can be made.

Mathematical equations and models, and computer models, are frequently used to describe the past and future behaviour of a process within the boundaries of that model. In some cases the probability of an outcome, rather than a specific outcome, can be predicted, for example in much of quantum physics.

In microprocessors, branch prediction permits avoidance of pipeline emptying at branch instructions.

In engineering, possible failure modes are predicted and avoided by correcting the failure mechanism causing the failure.

Accurate prediction and forecasting are very difficult in some areas, such as natural disasters, pandemics, demography, population dynamics and meteorology. For example, it is possible to predict the occurrence of solar cycles, but their exact timing and magnitude is much more difficult (see picture to right).

In materials engineering it is also possible to predict the life time of a material with a mathematical model.^[6]

Hypothesis

Established science makes useful predictions which are often extremely reliable and accurate; for example, eclipses are routinely predicted.

New theories make predictions which allow them to be disproved by reality. For example, predicting the structure of crystals at the atomic level is a current research challenge.^[7] In the early 20th century the scientific consensus was that there existed an absolute frame of reference, which was given the name luminiferous ether.

The existence of this absolute frame was deemed necessary for consistency with the established idea that the speed of light is constant. The famous Michelson–Morley experiment demonstrated that predictions deduced from this concept were not borne out in reality, thus disproving the theory of an absolute frame of reference. The special theory of relativity was proposed by Einstein as an explanation for the seeming inconsistency between the constancy of the speed of light and the non-existence of a special, preferred or absolute frame of reference.

Albert Einstein's theory of general relativity could not easily be tested as it did not produce any effects observable on a terrestrial scale. However, as one of the first tests of general relativity, the theory predicted that large masses such as stars would bend light, in contradiction to accepted theory; this was observed in a 1919 eclipse.

Finance

Main articles: Financial forecasting and Stock market prediction

Mathematical models of stock market behaviour (and economic behaviour in general) are also unreliable in predicting future behaviour. Among other reasons, this is because economic events may span several years, and the world is changing over a similar time frame, thus invalidating the relevance of past observations to the present.

Thus there are an extremely small number (of the order of 1) of relevant past data points from which to project the future. In addition, it is generally believed that stock market prices already take into account all the information available to predict the future, and subsequent movements must therefore be the result of unforeseen events. Consequently, it is extremely difficult for a stock investor to anticipate or predict a stock market boom, or a stock market crash. In contrast to predicting the actual stock return, forecasting of broad economic trends tends to have better accuracy. Such analysis is provided by both non-profit groups as well as by for-profit private institutions.^[citation ^needed]

Some correlation has been seen between actual stock market movements and prediction data from large groups in surveys and prediction games.

An actuary uses actuarial science to assess and predict future business risk, such that the risk(s) can be mitigated. For example, in insurance an actuary would use a life table (which incorporates the historical experience of mortality rates and sometimes an estimate of future trends) to project life expectancy.

Sports

Predicting the outcome of sporting events is a business which has grown in popularity in recent years. Handicappers predict the outcome of games using a variety of mathematical formulas, simulation models or qualitative analysis. Early, well known sports bettors, such as Jimmy the Greek, were believed to have access to information that gave them an edge. Information ranged from personal issues, such as gambling or drinking to undisclosed injuries; anything that may affect the performance of a player on the field.

Recent times have changed the way sports are predicted. Predictions now typically consist of two distinct approaches: Situational plays and statistical based models. Situational plays are much more difficult to measure because they usually involve the motivation of a team. Dan Gordon, noted handicapper, wrote "Without an emotional edge in a game in addition to value in a line, I won't put my money on it".^[8] These types of plays consist of: Betting on the home underdog, betting against Monday Night winners if they are a favorite next week, betting the underdog in "look ahead" games etc. As situational plays become more widely known they become less useful because they will impact the way the line is set.

The widespread use of technology has brought with it more modern sports betting systems. These systems are typically algorithms and simulation models based on regression analysis. Jeff Sagarin, a sports statistician, has brought attention to sports by having the results of his models published in USA Today. He is currently paid as a consultant by the Dallas Mavericks for his advice on lineups and the use of his Winval system, which evaluates free agents.

Brian Burke, a former Navy fighter pilot turned sports statistician, has published his results of using regression analysis to predict the outcome of NFL games.^[9] Ken Pomeroy is widely accepted as a leading authority on college basketball statistics. His website includes his College Basketball Ratings, a tempo based statistics system. Some statisticians have become very famous for having successful prediction systems. Dare wrote "the effective odds for sports betting and horse racing are a direct result of human decisions and can therefore potentially exhibit consistent error".^[10] Unlike other games offered in a casino, prediction in sporting events can be both logical and consistent.

Other more advance models include those based on Bayesian networks, which are causal probabilistic models commonly used for risk analysis and decision support. Based on this kind of mathematical modelling, Constantinou et al.,^[11][12] have developed models for predicting the outcome of association football matches.^[13]

What makes these models interesting is that, apart from taking into consideration relevant historical data, they also incorporate all these vague subjective factors, like availability of key players, team fatigue, team motivation and so on. They provide the user with the ability to include their best guesses about things that there are no hard facts available. This additional information is then combined with historical facts to provide a revised prediction for future match outcomes. The initial results based on these modelling practices are encouraging since they have demonstrated consistent profitability against published market odds.

Nowadays sport betting is a huge business; there are many websites (systems) alongside betting sites, which give tips or predictions for future games.^[14] Some of these prediction websites (tipsters) are based on human predictions, but others on computer software sometimes called prediction robots or bots. Prediction bots can use different amount of data and algorithms and because of that their accuracy may vary.

Sites such as Tzefi.com maintain that other sites' claim that they simulate the game 50,000 times before it's actually played on the field, is quite misleading and incorrect. This is due to the fact that those sites don't consider the human element of the game, and that a batter with a .330 ERA may strike out when the bases are loaded, or a quarterback with a 100+ passer rating may throw an interception at a crucial time. Tzefi.com boasts a 64.5% accuracy in predicting NFL games.

Social science

Prediction in the non-economic social sciences differs from the natural sciences and includes multiple alternative methods such as trend projection, forecasting, scenario-building and Delphi surveys. The oil company Shell is particularly well known for its scenario-building activities.^[citation ^needed]

One reason for the peculiarity of societal prediction is that in the social sciences, "predictors are part of the social context about which they are trying to make a prediction and may influence that context in the process".^[15] As a consequence, societal predictions can become self-destructing. For example, a forecast that a large percentage of a population will become HIV infected based on existing trends may cause more people to avoid risky behavior and thus reduce the HIV infection rate, invalidating the forecast (which might have remained correct if it had not been publicly known). Or, a prediction that cybersecurity will become a major issue may cause organizations to implement more security cybersecurity measures, thus limiting the issue.^[15]

In politics it is common to attempt to predict the outcome of elections via political forecasting techniques (or assess the popularity of politicians) through the use of opinion polls. Prediction games have been used by many corporations and governments to learn about the most likely outcome of future events.

Prophecy

Main article: Prophecy

Predictions have often been made, from antiquity until the present, by using paranormal or supernatural means such as prophecy or by observing omens. Methods including water divining, astrology, numerology, fortune telling, interpretation of dreams, and many other forms of divination, have been used for millennia to attempt to predict the future. These means of prediction have not been proven by scientific experiments.

In literature, vision and prophecy are literary devices used to present a possible timeline of future events. They can be distinguished by vision referring to what an individual sees happen. The book of Revelation, in the New Testament, thus uses vision as a literary device in this regard. It is also prophecy or prophetic literature when it is related by an individual in a sermon or other public forum.

Divination is the attempt to gain insight into a question or situation by way of an occultic standardized process or ritual.^[16] It is an integral part of witchcraft and has been used in various forms for thousands of years. Diviners ascertain their interpretations of how a querent should proceed by reading signs, events, or omens, or through alleged contact with a supernatural agency, most often describe as an angel or a god though viewed by Christians and Jews as a fallen angel or demon.^[17]

Fiction

Fiction (especially fantasy, forecasting and science fiction) often features instances of prediction achieved by unconventional means.

In fantasy literature, predictions are often obtained through magic or prophecy, sometimes referring back to old traditions. For example, in J. R. R. Tolkien's The Lord of the Rings, many of the characters possess an awareness of events extending into the future, sometimes as prophecies, sometimes as more-or-less vague 'feelings'. The character Galadriel, in addition, employs a water "mirror" to show images, sometimes of possible future events.

In some of Philip K. Dick's stories, mutant humans called precogs can foresee the future (ranging from days to years). In the story called The Golden Man, an exceptional mutant can predict the future to an indefinite range (presumably up to his death), and thus becomes completely non-human, an animal that follows the predicted paths automatically. Precogs also play an essential role in another of Dick's stories, The Minority Report, which was turned into a film by Steven Spielberg in 2002.

In the Foundation series by Isaac Asimov, a mathematician finds out that historical events (up to some detail) can be theoretically modelled using equations, and then spends years trying to put the theory in practice. The new science of psychohistory founded upon his success can simulate history and extrapolate the present into the future.

In Frank Herbert's sequels to 1965's Dune, his characters are dealing with the repercussions of being able to see the possible futures and select amongst them. Herbert sees this as a trap of stagnation, and his characters follow a so-called "Golden Path" out of the trap.

In Ursula K. Le Guin's The Left Hand of Darkness, the humanoid inhabitants of planet Gethen have mastered the art of prophecy and routinely produce data on past, present or future events on request. In this story, this was a minor plot device.

Poetry

For the ancients, prediction, prophesy, and poetry were often intertwined.^[18] Prophecies were given in verse, and a word for poet in Latin is “vates” or prophet.^[18] Both poets and prophets claimed to be inspired by forces outside themselves. In contemporary cultures, theological revelation and poetry are typically seen as distinct and often even as opposed to each other. Yet the two still are often understood together as symbiotic in their origins, aims, and purposes.^[19]

Used often to see if patterns in behavior (eg customers) so that a customer with certain attributes will do the same as another customer with similar attributes.Do all northern housewives buy the same stuff> eg. Also then business decision based on these will have the outcome we think. Get all aspects of the sales effort to get the desired result (If we do this on ad, on price, on slogans, on salesmen approach etc..)

Which patients are likely to develop certain conditions: diabetes, asthma, heart, lifetime illnesses… diagnosis is predictive also.

Getting bad payers to pay: which collection agency, contact strategies, legal action, etc..

With multiple products, oa can suggest customers will ALSO want x. (cross sales)

** keeping the customer vs competition: 5% up leads to 25% profit. Watch customers behavior (spend, choice, over time) < ?? attrition soon? Do something.

What mix of product version, marketing material, comm. Channels, timing..? working?

IRS uses pa to mine tax returns looking for fraud.

Calculate inventory for future orders.

Predict unemployment re state aid

BIG DATA

Eg RFID, web logs, sensor data, social nets, interbet searching, calls, science complex data meteorology eg)

Using Hadoop, MapReduce, etc can mine and analyse massive amounts of data. Even streaming data as it happens.

DATA MINING

Clean the data and remove errors. Then generalize, characterize, classify, cluster, associate, evolve, pattern match.

https://books.google.fr/books?id=sALywz73eB4C&pg=PA18&lpg=PA18&dq=Elder+and+Woodruff+MIT&source=bl&ots=A5NflR1cC-&sig=ACfU3U0YSn15uUTs_q6hwEfaPHQQGc1Lhg&hl=en&sa=X&ved=2ahUKEwjHteXB3JToAhVPyoUKHdH-BkEQ6AEwAHoECAcQAQ#v=onepage&q=Elder%20and%20Woodruff%20MIT&f=false

It all began in 1898 with a biscuit. See,up to then, you put anything you wanted to sell on the shelves, and if the shelves emptied fast, that meant customers liked it. Or not, if not. And before credit cards, with purchases all in cash, you couldn’t find out who your customers were. Then in 1898 the game changed. The American NABISCO corporation was looking for a way to launch a new product and passed the job to the first real advertising company, called NW Ayer. Who gave the new product a new kind of name. I’ll spell the name: U-N-E-E-D-A. Uneeda.

They wrote it on walls. 30 million people saw the name on their way to work. The associated slogan (in newspapers and magazines they read) went: ‘Do you know Uneeda biscuit?’ It worked. 10 million packets of biscuits in the first year, and they had to build extra factories to keep up with demand. So could other people do this clever trick? Not just make something and hope it would sell, but do something to make people want to buy it. Within no time, the psychologists were all over the issue like a rash. And they all said the same thing: ‘People want stuff they think is out of reach:either it’s too expensive, or in short supply. So you should use slogans like ‘while stocks last.’

In the 1920s and 30s radio changed the game again, when it was discovered that millions would tune in to hear a favourite programme. Most often a story of some kind. Soon these dramatised tales were being paid for by the soap companies, and generated soap-operas and the ads they came with.

So by now the numbers were getting big enough to turn the whole selling thing into a respectable business called ‘market research.’ At last the sellers started finding out who their buyers were, and why they bought what they bought. By the 1950’s your surveys would tell you fairly simple stuff:-age, social class, income, financial status, marital status, parenthood, and a few other things to shape what your product ought to be if it wanted to sell. And what pictures and words might help to sell it.

This was the time the BBC Audience Research Department commissioned a study to find out who was listening to, or watching what, and it hired a man called Belson to do it. What he did is the reason I’m doing this programme and the reason your future may not be what you had expected it to be. And I’ll get back to Belson in a moment.

Meanwhile the magic tool that was to make all this market research more than just a salesman’s dream:-- the computer. And here come a few numbers, so grit your teeth. In 1944 the electronic computer was invented to help work out all the sums required to tell gunners how to fire their shells in every condition:--weather, distance, temperature. shell-size, and all that, so your shells would hit what you wanted them to hit. The women who worked out all the sums, were called (and so was the machine that replaced them) ‘computers.’

OK here are the numbers I mentioned. That first computer filled a room and used eighteen thousand vacuum tubes, each one either on or off (binary code). And it could do 5,000 bits of addition a second. Amazing. Today,the hottest computer in the world (called ‘Summit’) is the size of two tennis courts and weighs three hundred and forty tons. It can do the calculations it would take you a lifetime to do, in one second. It can read and store five hundred billion pages of text.

So, first of all, why do we want a machine to do that? Well first off, we are generating information fast, and it needs processing. In the last two years we have generated the same amount of data as we generated from when we came out of the caves up until two years ago. And this is not arcane stuff. Each of us generates data all the time. Think of the bills we pay, the tax returns we fill in, the newspapers we want...all that paperwork we mount up.Just running daily life needs the help of machines now. So it’s good we have the computer. Well, as you’ll hear, kind of.

To start with, how we get the data to the machines in the first place. Well, there is what’s already being called the Internet of Things.That’s the millions of sensors of one kind or another, reporting on what’s happening around us. From meters recording your use of gas or electricity or phone or water, and so on.Or the thousands of bits of data collected every hour of a flight to tell the onboard computer how the plane is behaving and what to do about that. Or all the sensors that make it possible to keep a city working:-- like traffic patterns, weather, time of day, public transport numbers, and so on. And then at the personal level, all the sensors that let you run your life, from the old alarm clock telling you the time, to the programme that turns on the lights or the heating for you to come back to,at night. And soon your fridge will tell the shops what to send you, or the pharmacy what drugs you need, or the local garage that your car brakes need checking, or a zillion other details about your daily behaviour.

All this, of course, from the other big thing we’ve produced to run life with:-- the Internet. Today we do half a million tweets a minute. 40,000 Google searches an hour. Two point five billion Facebook users send thirty-one million messages a minute. And every interaction on Internet is data we generate by being wherever we are, and doing whatever we’re doing. Data we leave behind when we move on. ‘Data exhaust,’ it’s called.

And since the overwhelming majority of the things we do now, are done via the Internet, in some way or another, that’s where the data can be accessed, about who and what we are. To do market research, just some of the things you can find out on the internet about your customers and what they want in life (so you can sell it to them) are (and get ready for this long list)... :-- Everything you buy, your bank details and money activity, your travel,your shoe size, health, medication, medical history, birthplace, sex, marital status, social status, children, family structure, insurance, age, income, use of car, kind of car, use of busses, or train journeys, commuting schedule, what you read, TV/radio use and programme choices, your favourite papers and magazines, your education,job, hobbies, residence and location,friends, social life, languages, favorite music, sport, clubs, spare time activities, religious beliefs, events you attend, your ethnicity, personal grooming, pets, sweet tooth, sexual activity, skills, financial worth, investments, tax paid, clothing bought, film and theatre preferences, social media time, software use, food preferences, exercise, sleep, politics, credit and debit use, criminal behavior, gambling, pornography, social life. And I think I’ll stop.

You get the point. It’s all there for the finding. Much of it anonymous, but lots of it, not. And even anonymity is easily dealt with. Thanks to that guy Belson and what he did back in the 1950s, Which was to invent a thing called a ‘Decision Tree.’

Suppose as a business you’re trying to find a particular kind of person, with certain characteristics, who will go for your skin product. So that you can point your advertising at this person. OK. Think of a tree and a branch from it, leading to smaller branches, leading to twigs, and smaller twigs and finally, leaves. The main branch is all the women on your potential customer list (which you’ve generated from the data you’ve collected, as I described earlier,about people’s lifestyles). The next branch divides them, let’s say,between those in the 18-to -60 group, and the rest. The next branch splits those who are married and those not. And you go on refining the group like this until you get down to the final twig and leaves. One of which will have all the names of all the women you’ve been searching for, and which you can now predict will be likely to buy your specific product.

Prediction is the key, here. It’s what helps you not to waste your sales effort. Because you can use those trees to analyse (from what they’ve done so far), what a person will do next. Clearly you need more than one tree. Maybe thousands. And sure enough what would you expect, the statistician experts call all those trees a ‘forest.’ From what I said earlier, you’ll have guessed that forests present no problem to computers. In fact the speed and storage space of even a moderately big computer makes gigantic amounts of information usable.

This may be why we are now using terms like ‘Big Data.’ Quite apart from all those personal data-exhaust details we leave behind us, the entire world is doing so, too. Businesses collect and use data about every aspect of their work and staff and customers. About five percent of this data gets used because it’s germane to what the company wants to know. The rest is usually dumped in the company’s data warehouse (a fancy name for the hard drives with the data on them). Really big companies produce so much excess data they get stored in special places which become known as ‘data lakes.’ Where usually the data sits, irrelevant to daily business needs. But with all that computer storage-space we now have, no need to wipe it. So quintillions of bits of data are there (and here’s an awkward mixed metaphor) in data lakes that can be mined.

Data mining is the term used for digging through the data to look for patterns. That’s what decision trees and forests are for. In terms of market research, you’re looking for behavior patterns. So you can say that a customer with certain attributes will likely do the same things as another customer with the same attributes.It’s that kind of commonality you point your advertising at. Or why you design a new product the way you do. Because the data mining has enabled you to predict behavior with enough certainty to know the product will sell.

Data mining also reveals other useful patterns, too. Especially which customers have the kind of history to show they’ll probably default on their debt. And if so, what methods will work best for you to get the money they owe you. And this process doesn’t only interest company sales departments. It’s also the kind of stuff you want to know about customers likely to leave you for the competition. Or how happy your own staff are and how likely which of them are thinking of leaving the company. This kind of information is also a natural for the tax people, isn’t it. And for any government agency looking at future budgets and staffing for the kinds of social issues they might be called on to deal with in the near future. Issues like viruses, for instances. And speaking of viruses, of course one of the areas most likely to benefit from data mining is health. Looking at which people are most likely to develop the kind of conditions that will put the medial services under strain. Like obesity, or asthma, or diabetes, or heart conditions. In one sense, data mining is prognosis.

The internet has made all this data collection much easier, of course. Think about all the times you’re online and what you do. You click. And the click is the secret to everything. Because it says where you’re going, heading for what it is you want. Clearly there’s no data on the number of clicks done every day, but you can bet it’s in the quintillions. And aside from your where-am-I going click, there’s the little matter of giving a special OK click to the website you get to. The click that says ‘I accept your conditions.’ The cookie. Perhaps the greatest of all clicks. Because if you ever read the fine print attached to each cookie (and 99.9% of people don’t read the fine print), it says stuff like:-- ‘You allow us to share your click with third parties.’ So now, more people know what you’re doing, than perhaps you thought.

So clicks are money, if you’re passing them on as information to companies that want to point their advertising at specific people, perhaps.

This is how giants like Google and Facebook and Amazon and Microsoft and Apple (to name but a few) make their profits. From advertisers who get the data they want to mine, already pre-mined, if you see what I mean. Because the thousands of people going online and clicking to get to see tennis matches are probably likely to respond to advertising for a brand of trainers. This is why, when you’re browsing, you so often see ads that, surprise-surprise, relate to you and your likes. And why, when you buy something, the site often adds: “You’re buying that. You might also like… this.”

As if it already knew your preferences. Because it does. Which is also why, in the real world, you’ll see certain stuff, on shelves next to other stuff. Or products that sometimes just appear out of the blue. I’m thinking of a major supermarket chain in the US that discovered impending hurricanes drove people to buy fruit tarts. And once in the supermarket, your cellphone will tell you to take a look in the next aisle because there’s stuff there you might want. Or, as you drive away, that if you turn down the next street you’ll see the kind of restaurant or bar you’d like. And if that hasn’t happened to you yet,just wait.

I’m sure you can see where this is leading. Every since the shock of nine-eleven, we have begun to blur the line between security and privacy. Everybody would agree that it’s better to be safe than private. Because private might mean unprotected because nobody knew where you were when the danger moment came.

Apart from security issues, there’s another aspect of data mining and the kind of maths that comes with it, called ‘Predictive Analytics,’ that turns all that Big Data into the means to know enough about everybody to be able not only to get them to buy fruits tarts, but to modify their behavior in other ways. We do that modification of course already, with the law. Drive on the left, don’t park here, pay your taxes, don’t steal, all that. And we allow infringements of privacy already, for instance with CCTV. Cameras that not only help protect us from crime, but also tell somebody our number-plate and automatically pay a parking charge.

That’s OK. But the same data also says that we were there then, heading in whatever direction, to whatever destination, for how long, on what day and time. And much more, that we might not want others to know. Predictive Analytics uses data mining to predict what we will do in reaction to advertising, and of course much more. Behavioural analysis and prediction played a big part in directing the election campaign for Barak Obama. And then Donald Trump. Because if you know enough about people, you’re going to know how they’ll react to pretty much anything. And therefore what you need to do or say to get that reaction. [i wouldn’t equate ground game GOTV with Cambridge Analytica Brexit Do So]

The question for us all to decide is, how far we are prepared to let this kind of activity go, to make life what we want it to be, starting with fruit tarts… but perhaps ending with no privacy.

Mind you, all is not lost. There is a growing movement among those who value their privacy, who want what they call ‘the right to be forgotten.’ Not a bad idea, you might think. Except, of course, by definition there will always be some programmer or bureaucrat who knows you’ve been forgotten, won’t there?