🖍️Alan Turing - Collected Quotes

"A computer would deserve to be called intelligent if it could deceive a human into believing that it was human." (Alan Turing, "Computing Machinery and Intelligence", Mind Vol. 59, 1950)

"If one wants to make a machine mimic the behaviour of the human computer in some complex operation one has to ask him how it is done, and then translate the answer into the form of an instruction table. Constructing instruction tables is usually described as 'programming'." (Alan Turing, "Computing Machinery and Intelligence", Mind Vol. 59, 1950)

"It is unnecessary to design various new machines to do various computing processes. They can all be done with one digital computer, suitably programmed for each case." (Alan Turing, "Computing Machinery and Intelligence", Mind Vol. 59, 1950)

"The idea behind digital computers may be explained by saying that these machines are intended to carry out any operations which could be done by a human computer.” (Alan Turing, “Computing Machinery and Intelligence”, Mind Vol. 59, 1950)

"The original question, 'Can machines think?:, I believe too meaningless to deserve discussion. Nevertheless I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted." (Alan M. Turing, 1950) 

"The view that machines cannot give rise to surprises is due, I believe, to a fallacy to which philosophers and mathematicians are particularly subject. This is the assumption that as soon as a fact is presented to a mind all consequences of that fact spring into the mind simultaneously with it. It is a very useful assumption under many circumstances, but one too easily forgets that it is false. A natural consequence of doing so is that one then assumes that there is no virtue in the mere working out of consequences from data and general principles." (Alan Turing, "Computing Machinery and Intelligence", Mind Vol. 59, 1950)

"This model will be a simplification and an idealization, and consequently a falsification. It is to be hoped that the features retained for discussion are those of greatest importance in the present state of knowledge." (Alan M Turing, "The Chemical Basis of Morphogenesis" , Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences, Vol. 237 (641), 1952) 

"Almost everyone now acknowledges that theory and experiment, model making, theory construction and linguistics all go together, and that the successful development of a science of behavior depends upon a ‘total approach’ in which, given that the computer ‘is the only large-scale universal model’ that we possess, ‘we may expect to follow the prescription of Simon and construct our models - or most of them - in the form of computer programs’." (Alan M Turing)

"Science is a differential equation. Religion is a boundary condition." (Alan M Turing)

"The whole thinking process is rather mysterious to us, but I believe that the attempt to make a thinking machine will help us greatly in finding out how we think ourselves." (Alan M Turing)

"We do not need to have an infinity of different machines doing different jobs. A single one will suffice. The engineering problem of producing various machines for various jobs is replaced by the office work of "programming" the universal machine to do these jobs." (Alan M Turing)

🖍️David R Cox - Collected Quotes

"Exact truth of a null hypothesis is very unlikely except in a genuine uniformity trial." (David R Cox, "Some problems connected with statistical inference", Annals of Mathematical Statistics 29, 1958) 

"Assumptions that we make, such as those concerning the form of the population sampled, are always untrue." (David R Cox, "Some problems connected with statistical inference", Annals of Mathematical Statistics 29, 1958) 

"Overemphasis on tests of significance at the expense especially of interval estimation has long been condemned." (David R Cox, "The role of significance tests", Scandanavian Journal of Statistics 4, 1977) 

"There are considerable dangers in overemphasizing the role of significance tests in the interpretation of data." (David R Cox, "The role of significance tests", Scandanavian Journal of Statistics 4, 1977) 

"In any particular application, graphical or other informal analysis may show that consistency or inconsistency with H0 is so clear cut that explicit calculation of p is unnecessary." (David R Cox, "The role of significance tests", Scandanavian Journal of Statistics 4, 1977) 

"The central point is that statistical significance is quite different from scientific significance and that therefore estimation [...] of the magnitude of effects is in general essential regardless of whether statistically significant departure from the null hypothesis is achieved." (David R Cox, "The role of significance tests", Scandanavian Journal of Statistics 4, 1977) 

"At a simpler level, some elementary but important suggestions for the clarity of graphs are as follows: (i) the axes should be clearly labelled with the names of the variables and the units of measurement; (ii) scale breaks should be used for false origins; (iii) comparison of related diagrams should be made easy, for example by using identical scales of measurement and placing diagrams side by side; (iv) scales should be arranged so that systematic and approximately linear relations are plotted at roughly 45° to the x-axis; (v) legends should make diagrams as nearly self-explanatory, i.e. independent of the text, as is feasible; (vi) interpretation should not be prejudiced by the technique of presentation, for example by superimposing thick smooth curves on scatter diagrams of points faintly reproduced." (David R Cox,"Some Remarks on the Role in Statistics of Graphical Methods", Applied Statistics 27 (1), 1978)

"Most graphs used in the analysis of data consist of points arising in effect from distinct individuals, although there are certainly other possibilities, such as the use of lines dual to points. In many cases of exploratory analysis, however, the display of supplementary information attached to some or all of the points will be crucial for successful interpretation. The primary co-ordinate axes should, of course, be chosen to express the main dependence explicitly, if not initially certainly in the final presentation of conclusions." (David R Cox,"Some Remarks on the Role in Statistics of Graphical Methods", Applied Statistics 27 (1), 1978)

"So far as is feasible, diagrams should be planned so that (a) departures from "standard" conditions should be revealed as departures from linearity, or departures from totally random scatter, or as departures of contours from circular form; (b) different points should have approximately independent errors; (c) points should have approximately equal errors, preferably known and indicated, or, if equal errors cannot be achieved, major differences in the precision of individual points should be indicated, at least roughly; (d) individual points should have clearcut interpretation; (e) variables plotted should have clearcut physical interpretation; (f) any non-linear transformations applied should not accentuate uninteresting ranges; (g) any reasonable invariance should be exploited." (David R Cox,"Some Remarks on the Role in Statistics of Graphical Methods", Applied Statistics 27 (1), 1978)

"There are two general decisions to be made when displaying supplementary information, the first concerning the amount of such information and the second the precise technique to be used. The amount of supplementary information that it is sensible to show depends on the number of points. The possibility of showing such information only for relatively extreme points and possibly for a sample of the more central points should be considered when the number of points is large; thus in a probability plot of contrasts from a large factorial experiment it may be enough to label only the more extreme values." (David R Cox,"Some Remarks on the Role in Statistics of Graphical Methods", Applied Statistics 27 (1), 1978)

"It is very bad practice to summarise an important investigation solely by a value of P." (David R Cox, "Statistical significance tests", British Journal of Clinical Pharmacology 14, 1982) 

"The criterion for publication should be the achievement of reasonable precision and not whether a significant effect has been found." (David R Cox, "Statistical significance tests", British Journal of Clinical Pharmacology 14, 1982) 

"The continued very extensive use of significance tests is alarming." (David R Cox, "Some general aspects of the theory of statistics", International Statistical Review 54, 1986) 

"It has been widely felt, probably for thirty years and more, that significance tests are overemphasized and often misused and that more emphasis should be put on estimation and prediction. While such a shift of emphasis does seem to be occurring, for example in medical statistics, the continued very extensive use of significance tests is on the one hand alarming and on the other evidence that they are aimed, even if imperfectly, at some widely felt need." (David R Cox, "Some general aspects of the theory of statistics", International Statistical Review 54, 1986) 

"Most real life statistical problems have one or more nonstandard features. There are no routine statistical questions; only questionable statistical routines." (David R Cox)

🖍️John H Johnson - Collected Quotes

"A correlation is simply a bivariate relationship - a fancy way of saying that there is a relationship between two ('bi') variables ('variate'). And a bivariate relationship doesn’t prove that one thing caused the other. Think of it this way: you can observe that two things appear to be related statistically, but that doesn’t tell you the answer to any of the questions you might really care about - why is there a relationship and what does it mean to us as a consumer of data?" (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"A good chart can tell a story about the data, helping you understand relationships among data so you can make better decisions. The wrong chart can make a royal mess out of even the best data set." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Although some people use them interchangeably, probability and odds are not the same and people often misuse the terms. Probability is the likelihood that an outcome will occur. The odds of something happening, statistically speaking, is the ratio of favorable outcomes to unfavorable outcomes." (John H Johnson & Mike Gluck, "Everydata: The misibivarinformation hidden in the little data you consume every day", 2016)

"[…] average isn’t something that should be considered in isolation. Your average is only as good as the data that supports it. If your sample isn’t representative of the full population, if you cherry- picked the data, or if there are other issues with your data, your average may be misleading." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Big data is sexy. It makes the headlines. […] But, as you’ve seen already, it’s the little data - the small bits and bytes of data that you’re bombarded with in your everyday life - that often has a huge effect on your health, your wallet, your job, your relationships, and so much more, every single day. From food labels to weather forecasts, your bank account to your doctor’s office, everydata is all around you." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Confirmation bias can affect nearly every aspect of the way you look at data, from sampling and observation to forecasting - so it’s something  to keep in mind anytime you’re interpreting data. When it comes to correlation versus causation, confirmation bias is one reason that some people ignore omitted variables - because they’re making the jump from correlation to causation based on preconceptions, not the actual evidence." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Essentially, magnitude is the size of the effect. It’s a way to determine if the results are meaningful. Without magnitude, it’s hard to get a sense of how much something matters. […] the magnitude of an effect can change, depending on the relationship." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"First, you need to think about whether the universe of data that is being studied or collected is representative of the underlying population. […] Second, you need to consider what you are analyzing in the data that has been collected - are you analyzing all of the data, or only part of it? […] You always have to ask - can you accurately extend your findings from the sample to the general population? That’s called external validity - when you can extend the results from your sample to draw meaningful conclusions about the full population." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Forecasting is difficult because we don’t know everything about how the world works. There are unforeseen events. Unknown processes. Random occurrences. People are unpredictable, and things don’t always stay the same. The data you’re studying can change - as can your understanding of the underlying process." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Having a large sample size doesn’t guarantee better results if it’s the wrong large sample." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"If the underlying data isn’t sampled accurately, it’s like building a house on a foundation that’s missing a few chunks of concrete. Maybe it won’t matter. But if the missing concrete is in the wrong spot - or if there is too much concrete missing - the whole house can come falling down." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"If you’re looking at an average, you are - by definition - studying a specific sample set. If you’re comparing averages, and those averages come from different sample sets, the differences in the sample sets may well be manifested in the averages. Remember, an average is only as good as the underlying data." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"If your conclusions change dramatically by excluding a data point, then that data point is a strong candidate to be an outlier. In a good statistical model, you would expect that you can drop a data point without seeing a substantive difference in the results. It’s something to think about when looking for outliers." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"In the real world, statistical issues rarely exist in isolation. You’re going to come across cases where there’s more than one problem with the data. For example, just because you identify some sampling errors doesn’t mean there aren’t also issues with cherry picking and correlations and averages and forecasts - or simply more sampling issues, for that matter. Some cases may have no statistical issues, some may have dozens. But you need to keep your eyes open in order to spot them all." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Just as with aggregated data, an average is a summary statistic that can tell you something about the data - but it is only one metric, and oftentimes a deceiving one at that. By taking all of the data and boiling it down to one value, an average (and other summary statistics) may imply that all of the underlying data is the same, even when it’s not." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Keep in mind that a weighted average may be different than a simple (non- weighted) average because a weighted average - by definition - counts certain data points more heavily. When you’re thinking about an average, try to determine if it’s a simple average or a weighted average. If it’s weighted, ask yourself how it’s being weighted, and see which data points count more than others." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"[…] remember that, as with many statistical issues, sampling in and of itself is not a good or a bad thing. Sampling is a powerful tool that allows us to learn something, when looking at the full population is not feasible (or simply isn’t the preferred option). And you shouldn’t be misled to think that you always should use all the data. In fact, using a sample of data can be incredibly helpful." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Statistical significance is a concept used by scientists and researchers to set an objective standard that can be used to determine whether or not a particular relationship 'statistically' exists in the data. Scientists test for statistical significance to distinguish between whether an observed effect is present in the data (given a high degree of probability), or just due to chance. It is important to note that finding a statistically significant relationship tells us nothing about whether a relationship is a simple correlation or a causal one, and it also can’t tell us anything about whether some omitted factor is driving the result." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Statistical significance refers to the probability that something is true. It’s a measure of how probable it is that the effect we’re seeing is real (rather than due to chance occurrence), which is why it’s typically measured with a p-value. P, in this case, stands for probability. If you accept p-values as a measure of statistical significance, then the lower your p-value is, the less likely it is that the results you’re seeing are due to chance alone." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"This idea of looking for answers is related to confirmation bias, which is the tendency to interpret data in a way that reinforces your preconceptions. With confirmation bias, you aren’t just looking for an answer - you’re looking for a specific answer." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"The more uncertainty there is in your sample, the more uncertainty there will be in your forecast. A prediction is only as good as the information that goes into it, and in statistics, we call the basis for our forecasts a model. The model represents all the inputs - the factors you determine will predict the future outcomes, the underlying sample data you rely upon, and the relationship you apply mathematically. In other words, the model captures how you think various factors relate to one another." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"The process of making statistical conclusions about the data is called drawing an inference. In any statistical analysis, if you’re going to draw an inference, the goal is to make sure you have the right data to answer the question you are asking." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"The strength of an average is that it takes all the values in your data set and simplifies them down to a single number. This strength, however, is also the great danger of an average. If every data point is exactly the same (picture a row of identical bricks) then an average may, in fact, accurately reflect something about each one. But if your population isn’t similar along many key dimensions - and many data sets aren’t - then the average will likely obscure data points that are above or below the average, or parts of the data set that look different from the average. […] Another way that averages can mislead is that they typically only capture one aspect of the data." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"The tricky part is that there aren’t really any hard- and- fast rules when it comes to identifying outliers. Some economists say an outlier is anything that’s a certain distance away from the mean, but in practice it’s fairly subjective and open to interpretation. That’s why statisticians spend so much time looking at data on a case-by-case basis to determine what is - and isn’t - an outlier." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

"Using a sample to estimate results in the full population is common in data analysis. But you have to be careful, because even small mistakes can quickly become big ones, given that each observation represents many others. There are also many factors you need to consider if you want to make sure your inferences are accurate." (John H Johnson & Mike Gluck, "Everydata: The misinformation hidden in the little data you consume every day", 2016)

🖍️Victor Cohn - Collected Quotes

"Different problems require different methods, different numbers. One of the most basic questions in science is: Is the study designed in a way that will allow the researchers to answer the questions that they want answered?" (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"If the group is large enough, even very small differences can become statistically significant." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"In common language and ordinary logic, a low likelihood of chance alone calling the shots means 'it’s close to certain'. A strong likelihood that chance could have ruled means 'it almost certainly can’t be'." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"Most importantly, much of statistics involves clear thinking rather than numbers. And much, at least much of the statistical principles that reporters can most readily apply, is good sense." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"Nature is complex, and almost all methods of observation and experiment are imperfect." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"[…] nonparametric methods […] are methods of examining data that do not rely on a numerical distribution. As a result, they don’t allow a few very large or very small or very wild numbers to run away with the analysis." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"Regression toward the mean is the tendency of all values in every field of science – physical, biological, social, and economic – to move toward the average. […] The regression effect is common to all repeated measurements. Regression is part of an even more basic phenomenon: variation, or variability. Virtually everything that is measured varies from measurement to measurement. When repeated, every experiment has at least slightly different results." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"Statistically, power means the probability of finding something if it’s there.[…] statisticians think of power as a function of both sample size and the accuracy of measurement, because that too affects the probability of finding something." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"The big problems with statistics, say its best practitioners, have little to do with computations and formulas. They have to do with judgment - how to design a study, how to conduct it, then how to analyze and interpret the results. Journalists reporting on statistics have many chances to do harm by shaky reporting, and so are also called on to make sophisticated judgments. How, then, can we tell which studies seem credible, which we should report?" (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"The first thing that you should understand about science is that it is almost always uncertain. The scientific process allows science to move ahead without waiting for an elusive 'proof positive'. […] How can science afford to act on less than certainty? Because science is a continuing story - always retesting ideas. One scientific finding leads scientists to conduct more research, which may support and expand on the original finding." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

"Where many known, measurable factors are involved, statisticians can use mathematical techniques to account for all the variables and try to find which are the truly important predictors. The terms for this include multiple regression, multivariate analysis, and discriminant analysis, and factor, cluster, path, and two-stage least-squares analyses." (Victor Cohn & Lewis Cope, "News & Numbers: A writer’s guide to statistics" 3rd Ed, 2012)

🖍️Antoine Cornuéjols - Collected Quotes

"Hence, has machine learning uncovered truths that escaped the notice of philosophy, psychology, and biology? On one hand, it can be argued that machine learning has at least provided grounds for some of the claims of philosophy regarding the nature of knowledge and its acquisition. Against pure empiricism, induction requires prior knowledge, if only in the form of a constrained hypothesis space. In addition, there is a kind of conservation law at play in induction. The more a priori knowledge there is, the easier learning is and the fewer data are needed, and vice versa. The statistical study of machine learning allows quantifying this trade-off." (Antoine Cornuéjol, "The Necessity of Order in Machine Learning: Is Order in Order?", 2007)

"In effect, machine learning research has already brought us several interesting concepts. Most prominently, it has stressed the benefit of distinguishing between the properties of the hypothesis space - its richness and the valuation scheme associated with it - and the characteristics of the actual search procedure in this space, guided by the training data. This in turn suggests two important factors related to sequencing effects, namely forgetting and the nonoptimality of the search procedure. Both are key parameters than need to be thoroughly understood if one is to master sequencing effects." (Antoine Cornuéjol, "The Necessity of Order in Machine Learning: Is Order in Order?", 2007)

"On the other hand, the algorithms produced in machine learning during the last few decades seem quite remote from what can be expected to account for natural cognition. For one thing, there is virtually no notion of knowledge organization in these methods. Learning is supposed to arise on a blank slate, albeit a constrained one, and its output is not supposed to be used for subsequent learning episodes. Neither is there any hierarchy in the 'knowledge' produced. Learning is not conceived as an ongoing activity but rather as a one-shot process more akin to data analysis than to a gradual discovery development or even to an adaptive process. " (Antoine Cornuéjol, "The Necessity of Order in Machine Learning: Is Order in Order?", 2007)

"[...] the theory that establishes a link between the empirical fit of the candidate hypothesis with respect to the data and its expected value on unseen events becomes essentially inoperative if the data are not supposed to be independent of each other. This requirement is obviously at odds with most natural learning settings, where either the learner is actively searching for data or where learning occurs under the guidance of a teacher who is carefully choosing the data and their order of presentation." (Antoine Cornuéjol, "The Necessity of Order in Machine Learning: Is Order in Order?", 2007)

"There are many control parameters to a learning system. The question is to identify, at a sufficiently high level, the ones that can play a key role in sequencing effects. Because learning can be seen as the search for an optimal hypothesis in a given space under an inductive criteria defined over the training set, three means to control learning readily appear. The first one corresponds to a change of the hypothesis space. The second consists in modifying the optimization landscape. This can be done by changing either the training set (for instance, by a forgetting mechanism) or the inductive criteria. Finally, one can also fiddle with the exploration process. For instance, in the case of a gradient search, slowing down the search process can prevent the system from having time to find the local optimum, which, in turn, can introduce sequencing effects." (Antoine Cornuéjol, "The Necessity of Order in Machine Learning: Is Order in Order?", 2007)

"While it has been always considered that a piece of information could at worst be useless, it should now be acknowledged that it can have a negative impact. There is simply no theory of information at the moment offering a framework ready to account for this in general." (Antoine Cornuéjol, "The Necessity of Order in Machine Learning: Is Order in Order?", 2007)

🖍️Nate Silver - Collected Quotes

"A forecaster should almost never ignore data, especially when she is studying rare events […]. Ignoring data is often a tip-off that the forecaster is overconfident, or is overfitting her model - that she is interested in showing off rather than trying to be accurate."  (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Complex systems seem to have this property, with large periods of apparent stasis marked by sudden and catastrophic failures. These processes may not literally be random, but they are so irreducibly complex (right down to the last grain of sand) that it just won’t be possible to predict them beyond a certain level. […] And yet complex processes produce order and beauty when you zoom out and look at them from enough distance." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Data-driven predictions can succeed - and they can fail. It is when we deny our role in the process that the odds of failure rise. Before we demand more of our data, we need to demand more of ourselves." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Distinguishing the signal from the noise requires both scientific knowledge and self-knowledge." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Finding patterns is easy in any kind of data-rich environment; that's what mediocre gamblers do. The key is in determining whether the patterns represent signal or noise." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"The instinctual shortcut that we take when we have 'too much information' is to engage with it selectively, picking out the parts we like and ignoring the remainder, making allies with those who have made the same choices and enemies of the rest." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"The most basic tenet of chaos theory is that a small change in initial conditions - a butterfly flapping its wings in Brazil - can produce a large and unexpected divergence in outcomes - a tornado in Texas. This does not mean that the behavior of the system is random, as the term 'chaos' might seem to imply. Nor is chaos theory some modern recitation of Murphy’s Law ('whatever can go wrong will go wrong'). It just means that certain types of systems are very hard to predict." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"The numbers have no way of speaking for themselves. We speak for them. We imbue them with meaning." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"The signal is the truth. The noise is what distracts us from the truth." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"The systems are dynamic, meaning that the behavior of the system at one point in time influences its behavior in the future; And they are nonlinear, meaning they abide by exponential rather than additive relationships." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"We forget - or we willfully ignore - that our models are simplifications of the world. We figure that if we make a mistake, it will be at the margin. In complex systems, however, mistakes are not measured in degrees but in whole orders of magnitude." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"We need to stop, and admit it: we have a prediction problem. We love to predict things - and we aren't very good at it." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Whether information comes in a quantitative or qualitative flavor is not as important as how you use it. [...] The key to making a good forecast […] is not in limiting yourself to quantitative information. Rather, it’s having a good process for weighing the information appropriately. […] collect as much information as possible, but then be as rigorous and disciplined as possible when analyzing it. [...] Many times, in fact, it is possible to translate qualitative information into quantitative information." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Statistics is the science of finding relationships and actionable insights from data." (Nate Silver)

🖍️Beau Lotto - Collected Quotes

"Effects without an understanding of the causes behind them, on the other hand, are just bunches of data points floating in the ether, offering nothing useful by themselves. Big Data is information, equivalent to the patterns of light that fall onto the eye. Big Data is like the history of stimuli that our eyes have responded to. And as we discussed earlier, stimuli are themselves meaningless because they could mean anything. The same is true for Big Data, unless something transformative is brought to all those data sets… understanding." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"New information is constantly flowing in, and your brain is constantly integrating it into this statistical distribution that creates your next perception (so in this sense 'reality' is just the product of your brain’s ever-evolving database of consequence). As such, your perception is subject to a statistical phenomenon known in probability theory as kurtosis. Kurtosis in essence means that things tend to become increasingly steep in their distribution [...] that is, skewed in one direction. This applies to ways of seeing everything from current events to ourselves as we lean 'skewedly' toward one interpretation, positive or negative. Things that are highly kurtotic, or skewed, are hard to shift away from. This is another way of saying that seeing differently isn’t just conceptually difficult - it’s statistically difficult." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"Our assumptions are un question ably interconnected. They are nodes with connections (edges) to other nodes. The more foundational the assumption, the more strongly connected it is. What I’m suggesting is that our assumptions and the highly sensitive network of responses, perceptions, behaviors, thoughts, and ideas they create and interact with are a complex system. One of the most basic features of such a network is that when you move or disrupt one thing that is strongly connected, you don’t just affect that one thing, you affect all the other things that are connected to it. Hence small causes can have massive effects (but they don’t have to, and usually don’t actually). In a system of high tension, simple questions targeting basic assumptions have the potential to transform perception in radical  and unpredictable ways." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"Questioning our assumptions is what provokes revolutions, be they tiny or vast, technological or social." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"Understanding reduces the complexity of data by collapsing the dimensionality of information to a lower set of known variables. s revolutions, be they tiny or vast, technological or social." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"The basis of complex systems is actually quite simple (and this is not an attempt to be paradoxical, like an art critic who describes a sculpture as 'big yet small'). What makes a system unpredictable and thus nonlinear (which includes you and your perceptual process, or the process of making collective decisions) is that the components making up the system are interconnected." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"The greatest leaders possess a combination of divergent traits: they are both experts and naïve, creative and efficient, serious and playful, social and reclusive - or at the very least, they surround themselves with this dynamic." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017) 

"The term [Big Data] simply refers to sets of data so immense that they require new methods of mathematical analysis, and numerous servers. Big Data - and, more accurately, the capacity to collect it - has changed the way companies conduct business and governments look at problems, since the belief wildly trumpeted in the media is that this vast repository of information will yield deep insights that were previously out of reach." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"Trust is fundamental to leading others into the dark, since trust enables fear to be 'actionable' as courage rather than actionable as anger. Since the bedrock of trust is faith that all will be OK within uncertainty, leaders’ fundamental role is to ultimately lead themselves. Research has found that successful leaders share three behavioral traits: they lead by example, admit their mistakes, and see positive qualities in others. All three are linked to spaces of play. Leading by example creates a space that is trusted - and without trust, there is no play. Admitting mistakes is to celebrate uncertainty. Seeing qualities in others is to encourage diversity." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"Understanding transcends context, since the different contexts collapse according to their previously unknown similarity, which the principle contains. That is what understanding does. And you actually feel it in your brain when it happens. Your 'cognitive load' decreases, your level of stress and anxiety decrease, and your emotional state improves." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

"What defines a good leader? Enabling other people to step into the unseen. […] as the world becomes increasingly connected and thus unpredictable, the concept of leadership too must change. Rather than lead from the front toward efficiency, offering the answers, a good leader is defined by how he or she leads others into darkness - into uncertainty." (Beau Lotto, "Deviate: The Science of Seeing Differently", 2017)

🖍️Robert Hooke - Collected Quotes

"Accounting figures are a blend of facts and arbitrary procedures that are designed to facilitate the recording and communication of business transactions. Their usefulness in the decision process is sometimes grossly overestimated." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"All of us learn by experience. Except for pure deductive processes, everything we learn is from someone's experience. All experience is a sample from an immense range of possible experience that no one individual can ever take in. It behooves us to know what parts of the information we get from samples can be trusted and what cannot." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Being experimental, however, doesn't necessarily make a scientific study entirely credible. One weakness of experimental work is that it can be out of touch with reality when its controls are so rigid that conclusions are valid only in the experimental situation and don't carryover into the real world." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Correlation analysis is a useful tool for uncovering a tenuous relationship, but it doesn't necessarily provide any real understanding of the relationship, and it certainly doesn't provide any evidence that the relationship is one of cause and effect. People who don't understand correlation tend to credit it with being a more fundamental approach than it is." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Experiments usually are looking for 'signals' of truth, and the search is always ham pered by 'noise' of one kind or another. In judging someone else's experimental results it's important to find out whether they represent a true signal or whether they are just so much noise." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

 "First and foremost an experiment should have a goal, and the goal should be something worth achieving, especially if the experimenter is working on someone else's (for example, the taxpayers') money. 'Worth achieving' implies more than just beneficial; it also should mean that the experiment is the most beneficial thing we can think of doing. Obviously we can't predict accurately the value of an experiment (this may not even be possible after we see how it turns out), but we should feel obliged to make as intelligent a choice as we can. Such a choice is sometimes labeled a 'value judgment'." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"In general a small-scale test or experiment will not detect a small effect, or small differences among various products." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Mistakes arising from retrospective data analysis led to the idea of experimentation, and experience with experimentation led to the idea of controlled experiments and then to the proper design of experiments for efficiency and credibility. When someone is pushing a conclusion at you, it's a good idea to ask where it came from - was there an experiment, and if so, was it controlled and was it relevant?" (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"One important way of developing our powers of discrimination between good and bad statistical studies is to learn about the differences between backward-looking (retrospective or historical) data and data obtained through carefully planned and controlled (forward-looking) experiments." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Only a 0 correlation is uninteresting, and in practice 0 correlations do not occur. When you stuff a bunch of numbers into the correlation formula, the chance of getting exactly 0, even if no correlation is truly present, is about the same as the chance of a tossed coin ending up on edge instead of heads or tails." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Randomization is usually a cheap and harmless way of improving the effectiveness of experimentation with very little extra effort." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Science usually amounts to a lot more than blind trial and error. Good statistics consists of much more than just significance tests; there are more sophisticated tools available for the analysis of results, such as confidence statements, multiple comparisons, and Bayesian analysis, to drop a few names. However, not all scientists are good statisticians, or want to be, and not all people who are called scientists by the media deserve to be so described." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Statistical reasoning is such a fundamental part of experimental science that the study of principles of data analysis has become a vital part of the scientist's education. Furthermore, […] the existence of a lot of data does not necessarily mean that any useful information is there ready to be extracted." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"The idea of statistical significance is valuable because it often keeps us from announcing results that later turn out to be nonresults. A significant result tells us that enough cases were observed to provide reasonable assurance of a real effect. It does not necessarily mean, though, that the effect is big enough to be important." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"Today's scientific investigations are so complicated that even experts in related fields may not understand them well. But there is a logic in the planning of experiments and in the analysis of their results that all intelligent people can grasp, and this logic is a great help in determining when to believe what we hear and read and when to be skeptical. This logic has a great deal to do with statistics, which is why statisticians have a unique interest in the scientific method, and why some knowledge of statistics can so often be brought to bear in distinguishing good arguments from bad ones." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)

"When a real situation involves chance we have to use probability mathematics to understand it quantitatively. Direct mathematical solutions sometimes exist […] but most real systems are too complicated for direct solutions. In these cases the computer, once taught to generate random numbers, can use simulation to get useful answers to otherwise impossible problems." (Robert Hooke, "How to Tell the Liars from the Statisticians", 1983)