Start free trial
Trending New Genres Authors
EnglishEnglish
EspañolSpanish
简体中文Chinese
繁體中文Chinese (Traditional)
FrançaisFrench
DeutschGerman
日本語Japanese
PortuguêsPortuguese
ItalianoItalian
한국어Korean
РусскийRussian
NederlandsDutch
العربيةArabic
PolskiPolish
हिन्दीHindi
Tiếng ViệtVietnamese
SvenskaSwedish
ΕλληνικάGreek
TürkçeTurkish
ไทยThai
ČeštinaCzech
RomânăRomanian
MagyarHungarian
УкраїнськаUkrainian
IndonesiaIndonesian
DanskDanish
SuomiFinnish
БългарскиBulgarian
עבריתHebrew
NorskNorwegian
HrvatskiCroatian
CatalàCatalan
SlovenčinaSlovak
LietuviųLithuanian
SlovenščinaSlovenian
СрпскиSerbian
EestiEstonian
LatviešuLatvian
فارسیPersian
മലയാളംMalayalam
தமிழ்Tamil
اردوUrdu
Searching...
SoBrief
Big Data

Big Data

A Revolution That Will Transform How We Live, Work, and Think
Big data can predict what matters before experts see it, but prediction is not explanation.
by Viktor Mayer-Schönberger 2013 242 pages
3.69
8k+ ratings
Listen
Immersive
V2.3
Amazon Kindle Audible
Try Full Access for 3 Days
Unlock listening & more!
Continue

Key Takeaways

Treat data scale as a change in what questions become answerable

Scale changes the question. Big data means using information at a scale that can reveal insights or create value impossible to reach with smaller datasets. During the 2009 H1N1 outbreak, Google searches offered a faster picture of flu activity than delayed official reports. Oren Etzioni’s Farecast service analyzed ticket-price histories to advise when to buy, without needing to understand the airlines’ pricing rules. By 2012, its predictions were right about three-quarters of the time and saved travelers an average of $50 per ticket.

More than faster computers. The book’s central claim is that this shift comes from new ways of using data, not simply from better machines. When quantity changes, new patterns and possibilities can emerge.

Analysis

The strongest version of this argument is about expanding the range of observable questions. A sparse dataset can answer only what researchers thought to ask in advance; a broad one can expose patterns they did not anticipate. Yet volume alone does not guarantee insight. Search records describe people who use a particular platform, not everyone who has the flu. Scale changes what can be seen, while collection choices still shape the view.

Use whole datasets when the answer depends on rare subgroups

Samples blur at the edges. Random sampling remains useful: about 1,100 randomly selected responses can estimate a yes-or-no answer for a very large population within roughly three percentage points in most cases. But that precision does not transfer neatly to small subgroups. A survey of a thousand people may include too few affluent voters in one region to say much about them. The book calls the alternative N=all: using the complete dataset, or as much of it as practical.

Rare signals need context. In eleven years of sumo records, Steven Levitt and a colleague examined more than 64,000 matches. They found suspiciously high win rates in bouts where one wrestler needed a victory to retain rank, a pattern a small sample might have missed.

Analysis

The lesson is not that sampling has become obsolete. Sampling can be cheaper, faster, and more protective of privacy, and a well-designed sample can outperform a biased census. The useful distinction is between questions about an overall average and questions about unusual cases or fine-grained groups. When a rare event matters, a full record can reveal patterns that an average smooths away, provided the records actually cover the relevant population.

Trade pinpoint accuracy for breadth only when errors average out

Messiness can buy reach. Large datasets often contain inconsistent labels, missing values, delayed readings, or outright errors. The authors argue that accepting some imperfections can be worthwhile when more observations reveal a clearer overall pattern. Google’s translation system, for example, drew on an enormous collection of internet text, including flawed and uneven translations. Its scale helped it outperform systems trained on much smaller, carefully curated material.

Precision still has a place. A vineyard with many inexpensive temperature sensors may get a richer picture than one precise sensor can provide, even if some readings are wrong. The tradeoff works only when mistakes do not systematically skew the result. An approximate answer may suit trend detection; a bank balance or a safety-critical calculation requires stricter accuracy.

Analysis

This is a practical distinction between random noise and systematic error. Random mistakes can become less influential as observations accumulate, while a shared bias can grow more convincing with every added record. The decision therefore should not be simply whether a dataset is clean. It should be whether its flaws distort the particular conclusion being drawn. A messy weather signal may help anticipate a trend; the same tolerance would be reckless in a clinical dosage.

Let correlations predict useful signals, but don't mistake them for causes

A proxy can be enough. A correlation is a statistical relationship between two things that tend to change together. It can make one measurable signal a useful proxy for something harder to observe, without explaining why the connection exists. Amazon’s recommendations improved when Greg Linden’s team matched products that customers bought together, rather than trying to match each customer to similar shoppers. Walmart found that Pop-Tarts sales rose before hurricanes and placed them near storm supplies.

Prediction is not explanation. Correlations can guide decisions, such as what to stock or what to inspect, but coincidence and unstable patterns can mislead. Data analysis may identify a useful signal before anyone has a credible causal account. That can be enough for a low-risk forecast, but not necessarily for deciding what intervention will fix the problem.

Analysis

The distinction resembles the difference between a thermometer and a theory of heat: a measure can guide action without explaining the underlying process. That is powerful in time-sensitive settings, where waiting for a complete explanation can be costly. It is also a reason to limit the action to what the evidence supports. A reliable warning signal may justify checking a machine; it does not, by itself, identify the repair.

Turn overlooked traces into data before deciding they are worthless

Datafication reveals hidden structure. Datafication means turning information about something into a quantified form that can be recorded and analyzed. It differs from digitization, which merely converts material into computer-readable form. Matthew Fontaine Maury recognized that discarded ship logs recorded winds, currents, and weather. He organized more than a million observations into ocean charts, helping ships choose faster routes and reducing some voyages by about a third.

New forms invite new uses. Scanning a book creates a digital image; recognizing its words makes the text searchable and analyzable. Likewise, turning location, engine vibration, or a person’s interactions into data can support uses beyond the original purpose. The value begins not only with collecting information, but with representing it in a form that permits comparison and reuse.

Analysis

Maury’s story shows that data innovation is partly a change in attention: material dismissed as clutter can become evidence once someone sees a way to organize it. The same principle applies to institutions whose records were created for accounting, maintenance, or communication rather than research. But quantification is never neutral. Choosing what to count, how to classify it, and what to leave out determines which patterns become visible.

Price data by its future uses, not its first job

Reuse creates option value. The authors use this term for the potential worth of data across possible future uses, beyond the purpose for which it was collected. A ReCaptcha user originally typed distorted words to prove they were human; the same effort also helped decipher text that optical character recognition could not read. Search typos can improve spelling suggestions, while old ticket records can forecast future fares.

Information can work repeatedly. Unlike a physical resource that is consumed, data can often be analyzed again or combined with other information. The digital traces left by ordinary actions are called data exhaust. Their usefulness may fade, however: an old purchase may no longer reflect a customer’s interests. Data holders need to judge which records remain informative and which have become noise.

Analysis

The option-value idea changes the economics of record keeping. A dataset can be more valuable to someone who sees a secondary use than to the organization that collected it. This creates incentives to preserve and share information, but also incentives to gather more than people expect. The crucial question is therefore not just whether data has future value, but who can realize that value and who bears the cost of its reuse.

Build advantage by pairing data access with imagination and analytical skill

Three ingredients create value. The book describes three roles in the big-data economy: holders who control information, specialists who can analyze it, and people with a big-data mindset who see new ways to use it. A retailer might own transaction records, hire an analytics firm to find patterns, and then apply an unexpected insight to its operations. Data intermediaries such as Inrix combine traffic information from many sources to make predictions no single carmaker could produce as well.

Control can become leverage. The authors expect data itself to grow in importance as analytical skills become more widespread. Companies can also change what they sell: Rolls-Royce used engine-monitoring information to build an ongoing service business around aircraft engines, not just their manufacture.

Analysis

The value chain helps explain why a technically gifted analyst may not capture the largest share of the gains. Access, interpretation, and commercial application are separate capabilities, and bargaining power often belongs to the party that controls scarce information. Shared intermediaries can improve predictions by combining rival firms’ records, but the firms must trust the intermediary and agree on who benefits from the pooled resource.

Use data to challenge expertise, not to erase judgment

Analytics can expose inherited assumptions. In baseball, the Oakland Athletics used measures such as on-base percentage to challenge scouts’ subjective judgments about players. Amazon likewise found that automated recommendations sold more books than its editorial team’s selections. The point is not that specialists know nothing; it is that experience can miss patterns visible across many cases. One study cited in the book found productivity up to 6 percent higher at firms that emphasized data-driven decisions.

Expertise still improves the question. New York’s analysts learned from inspectors that fresh exterior brickwork suggested a property owner cared for a building. They found permit records that could represent that observation in their model. Data analysis and field knowledge worked together: experienced workers supplied clues, while the model helped rank a vast list of properties.

Analysis

Data can discipline intuition, especially when people rely on vague impressions that are hard to compare across cases. Yet a model is only as useful as its target and the measures chosen to represent it. Domain experts can identify when a proxy misses important context, while analysts can test whether expert beliefs hold across many observations. The productive arrangement is neither blind deference to experience nor blind deference to a score.

Treat data as evidence, never as an unquestionable verdict

Numbers can be gamed. Robert McNamara, the U.S. defense secretary during the Vietnam War, relied heavily on statistics such as enemy body counts to judge progress. Generals later reported that these figures were often inflated, partly because officers knew how much senior leaders valued them. A metric intended to describe a war began to shape what people reported about it.

A measure is not the thing itself. The book warns against treating data as a source of automatic truth. Google continued asking job candidates for academic scores even after internal analysis found little connection to job performance. Collecting more numbers cannot repair a poorly chosen measure, a distorted reporting process, or an interpretation that exceeds the evidence.

Analysis

This is a classic measurement problem: once a number becomes a target, people may optimize the number rather than the underlying goal. The danger grows when leaders mistake legibility for understanding. A dashboard can make a complex institution look manageable while hiding the incentives that produced its entries. Good practice asks how a metric was generated, what behavior it rewards, and what important outcomes it leaves invisible.

Never punish people for predictions about choices they have not made

Risk is not guilt. Predictive systems can estimate who may face health problems, default on a loan, or be involved in crime. The book argues that such forecasts can help allocate attention, but should not be treated as proof of responsibility. One parole prediction method described in the text claimed at least 75 percent accuracy for homicide involvement. Even on that claim, roughly one person in four would be wrongly predicted.

Future behavior remains a choice. Predictive policing and preemptive screening raise the danger of imposing restrictions before any offense occurs. A person cannot meaningfully disprove a prediction about an act that has not happened. Using probability to guide support or inspection is different from using it to assign blame or punishment.

Analysis

This concern goes beyond model accuracy. Even a well-calibrated forecast can distribute burdens unfairly if some people are watched more closely and therefore generate more recorded incidents. The distinction between prevention and punishment matters: offering voluntary support is not equivalent to restricting liberty. Any intervention should be judged by its consequences for the person classified, not just by whether it lowers an aggregate risk score.

Make data users accountable and algorithms appealable

Shift responsibility to data users. Because many secondary uses cannot be anticipated when information is collected, the authors argue that notice and consent alone cannot protect privacy. They propose data-user accountability: organizations assess the risks of a planned use and take responsibility for preventing harm. A related technique, differential privacy, deliberately makes results approximate so that analysis of a large dataset is less likely to expose a specific person.

Create ways to challenge decisions. The authors propose algorithmists, independent or in-house reviewers who examine the data, methods, and interpretation behind consequential predictions. For sensitive decisions, they call for openness about the basis of a prediction, expert certification, and a way for affected people to disprove it. These protections aim to make complex systems answerable to people.

Analysis

The proposal recognizes that privacy is not a one-time permission slip. It is an ongoing relationship shaped by what organizations do with information. Accountability can support useful analysis without requiring repeated consent for every unforeseen purpose, but it needs enforceable consequences and genuinely independent review. Otherwise, an audit risks becoming a compliance ritual that documents a decision without giving the affected person a meaningful route to contest it.

Treat forecasts as maps for action, not portraits of reality

Prediction can save lives. New York City had roughly 25,000 complaints each year about illegal building conversions, but only about 200 inspectors. Mike Flowers’s team combined records from 19 agencies with five years of fire data to rank which complaints deserved urgent attention. Inspectors had previously issued vacate orders in 13 percent of inspected cases; with the rankings, the figure rose above 70 percent. Field observations, including clues about building upkeep, helped improve the model.

No model sees everything. Forecasts estimate likely outcomes, not settled futures, and the information available is always incomplete. The book’s closing argument is that data should help people act, not replace creativity, intuition, or responsibility. Better predictions can guide attention; they cannot decide by themselves what society should value.

Analysis

The New York example makes the promise concrete: limited staff can be directed toward the cases most likely to be dangerous. It also shows why prediction should remain part of a human process. Inspectors supplied observations the dataset did not initially contain, and officials still had to decide how to use the rankings. A forecast is most valuable when it sharpens judgment without pretending to settle every question.

Analysis

The book’s most durable insight is not that bigger datasets always produce better answers. It is that the costs of collecting, storing, and recombining information have fallen enough to make new forms of inquiry practical. The examples make that shift tangible: Maury converts neglected ship logs into navigational guidance, Amazon replaces human book curation with purchase-pattern recommendations, and New York City directs scarce inspectors toward dangerous buildings. Together, these stories show data as infrastructure, not merely as a technical input.

The argument is strongest when it distinguishes prediction from explanation. A correlation can be valuable without revealing a cause, especially when the goal is triage, forecasting, or identifying a promising question for further study. It becomes much weaker when the authors imply that broad data can routinely displace theory or expert judgment. Data are not a transparent view of reality. Platforms select who becomes visible, measurement systems encode assumptions, and larger searches create more opportunities to find coincidences. A massive dataset can make a narrow or biased observation look precise.

The political analysis is equally important. Prediction changes power because institutions can classify people before they act, often without giving them a comprehensible reason or a way to contest the classification. The proposed safeguards, accountability, human agency, and algorithmic review, address genuine problems, but their effectiveness depends on independence, enforceable remedies, and access to evidence. An auditor with no authority to halt a harmful system offers little protection.

The book is best read as a guide to a new form of evidence, not as a prophecy that data will replace judgment. Its practical standard is modest and powerful: use broad information to notice patterns and improve decisions, while preserving causal investigation where stakes demand it, and humility wherever predictions touch human lives.

Summary in 30 Seconds
Large datasets can reveal patterns and rare cases that small samples miss, though imperfect data helps only when errors do not systematically distort results. Correlations can guide forecasts without explaining causes. Information gains value when it is made reusable and combined with analytical skill and domain knowledge. Predictions can help direct scarce resources, but they are uncertain, not proof of guilt or a substitute for human judgment. Organizations need oversight and ways for people to challenge consequential decisions.
Contains spoilers

Last updated:

Report Issue

Review Summary

3.69 out of 5
Average of 8k+ ratings from Goodreads and Amazon.

Big Data receives mixed reviews, with praise for its accessible overview of the topic and illustrative examples. Critics note redundancy and oversimplification. Readers appreciate insights into data's impact on society, privacy concerns, and future implications. Some find the content outdated or lacking depth. The book is recommended for those new to big data concepts but may disappoint experts. Overall, it's viewed as a thought-provoking introduction to an increasingly important field, albeit with limitations in scope and detail.

Your rating:
4.29
284 ratings
Want to read the full book?

Glossary

Big data

Large-scale analysis

The book defines big data as using information at a scale that makes new insights or forms of value possible, beyond what smaller datasets can support. Its importance lies not just in the quantity of data or computing power, but in new ways of analyzing, reusing, and acting on information.

N=all

Use the full dataset

N=all describes an approach that uses the complete dataset for a phenomenon, or as much of it as practical, rather than relying on a sample. Broader coverage can reveal rare cases and small subgroups that sampling may obscure. It does not mean that every possible fact about the world can be collected.

Datafication

Quantify for analysis

Datafication is the process of turning information about an object, activity, or relationship into a quantified form that can be recorded, compared, and analyzed. It differs from digitization, which converts material into computer-readable form but does not necessarily make its content searchable or analyzable as data.

Option value

Potential value from reuse

The option value of data is the value available through its possible future uses, not only through the purpose for which it was first collected. Search queries, for example, may serve an immediate search and later help improve spelling, predict trends, or support another service.

Data exhaust

Traces left by activity

Data exhaust is the information produced as a byproduct of actions, especially digital interactions. Clicks, search terms, page visits, and corrections can be gathered and analyzed to improve a service or create another one. The traces may be incidental to the user, but valuable to the organization that collects them.

Recombinant data

Combined datasets

Recombinant data refers to combining information from different sources to answer questions that none of the datasets could answer as effectively alone. The book’s example joins Danish cellphone-subscription records, cancer records, and demographic information to study whether cellphone use is associated with cancer.

Reality mining

Infer behavior from phones

Reality mining is the analysis of large quantities of mobile-phone data to make inferences about human behavior. Location and communication patterns can help reveal social activity or identify emerging trends. The method can support public-health and urban research, but also creates risks when personal movements are used without adequate safeguards.

Algorithmists

Auditors of algorithms

Algorithmists are proposed specialists who review the data, analytical methods, and interpretations behind consequential predictions. They may work independently or within organizations, assessing whether a system is sound and responding to complaints from people affected by its decisions. Their role is meant to make complex data systems more accountable and contestable.

Differential privacy

Blur results to protect people

Differential privacy is a technical approach that deliberately makes the results of queries approximate, reducing the chance that an individual can be identified from the output. The goal is to preserve useful patterns in a large dataset while making it harder to connect a specific result to a specific person.

Culturomics

Study culture through text

Culturomics is the quantitative study of language and cultural trends using large collections of digitized, searchable texts. Researchers can track how words and references change over time, compare writing, or study patterns of attention. Its findings depend on which books were preserved and digitized, so the text collection is not a complete record of society.

FAQ

What is "Big Data: A Revolution That Will Transform How We Live, Work, and Think" by Viktor Mayer-Schönberger about?

  • Big Data Revolution Overview: The book explores how the explosion of data and new analytical techniques are transforming society, business, healthcare, and governance.
  • Shift in Data Analysis: It highlights the move from small, exact datasets to vast, messy ones, focusing on correlations rather than causality.
  • Societal and Ethical Implications: The authors discuss both the benefits and risks of big data, including privacy concerns and the need for new governance principles.
  • Real-World Case Studies: Examples like Google Flu Trends and Farecast illustrate how big data is applied in practice.

Why should I read "Big Data: A Revolution That Will Transform How We Live, Work, and Think" by Viktor Mayer-Schönberger?

  • Comprehensive Introduction: The book offers an accessible yet thorough introduction to big data’s impact across sectors, making it suitable for both novices and experts.
  • Mindset Shifts: It explains the fundamental changes in thinking required to leverage big data, such as embracing messiness and prioritizing correlation.
  • Practical and Ethical Guidance: The authors provide frameworks for harnessing big data’s potential while addressing privacy and ethical challenges.
  • Preparation for the Future: Reading it equips you to navigate and succeed in an increasingly data-driven world.

What are the key takeaways from "Big Data: A Revolution That Will Transform How We Live, Work, and Think" by Viktor Mayer-Schönberger?

  • Three Major Shifts: The book identifies using all data (N=all), embracing messiness, and valuing correlation over causality as core changes in data analysis.
  • Data as a New Asset: Data is positioned as a vital economic input, with value increasing through reuse and combination.
  • Risks and Governance: It stresses the need for new privacy frameworks and algorithmic accountability to manage big data’s risks.
  • Changing Expertise: The rise of data skills and the decline of traditional subject-matter supremacy are highlighted as workforce trends.

How does "Big Data" by Viktor Mayer-Schönberger define and explain the concept of datafication?

  • Definition of Datafication: Datafication is the process of turning aspects of life—like behavior, location, and interactions—into quantifiable data for analysis.
  • Difference from Digitization: Unlike digitization, which converts analog to digital, datafication makes intangible phenomena measurable and analyzable.
  • Historical and Modern Context: The book traces datafication from early record-keeping to modern GPS and sensor data, showing its deep roots and current acceleration.
  • Enabling New Insights: Datafication expands what can be analyzed, leading to innovations in health, advertising, and urban management.

What are the main mindset shifts about data analysis described in "Big Data" by Viktor Mayer-Schönberger?

  • From Sampling to N=all: The book advocates analyzing entire datasets instead of small samples, reducing sampling bias and increasing insight granularity.
  • Embracing Messiness: It encourages tolerating imperfect, inconsistent data, as large volumes can compensate for inaccuracies and yield better results.
  • Prioritizing Correlation Over Causality: The focus shifts to finding predictive correlations rather than understanding underlying causes, challenging traditional scientific approaches.
  • Actionable Insights: These shifts enable faster, more practical decision-making in business, health, and governance.

How does "Big Data" by Viktor Mayer-Schönberger explain the use of correlations instead of causality in data analysis?

  • Correlation as a Tool: The book shows that correlations allow for accurate predictions without needing to understand why phenomena occur.
  • Practical Examples: Cases like Amazon’s recommendations and Walmart’s hurricane inventory illustrate how knowing “what” is happening can be sufficient.
  • Limitations of Causality: Establishing causality is often impractical; big data’s abundance makes correlation-based analysis more feasible and sometimes more useful.
  • Bias Correction: The authors argue that humans are prone to see causality where none exists, making correlation-based approaches a valuable corrective.

What is the "option value of data" according to "Big Data" by Viktor Mayer-Schönberger?

  • Definition of Option Value: Data’s value lies not just in its primary use but in its potential for novel, secondary applications.
  • Reuse and Combination: Combining datasets can unlock insights and value that are impossible from isolated data sources.
  • Business Implications: Companies that recognize and exploit the option value of data gain competitive advantages and create new business models.
  • Challenges in Valuation: Traditional accounting often fails to capture data’s true worth, making it an underappreciated asset.

How does "Big Data" by Viktor Mayer-Schönberger categorize big data companies and their roles in the data value chain?

  • Data Holders: These organizations control access to large datasets but may not analyze or innovate with the data themselves (e.g., MasterCard).
  • Data Specialists: Firms with analytics expertise extract insights from data, often serving clients who own the data (e.g., Accenture).
  • Big-Data Mindset Innovators: Individuals or startups who creatively use data, sometimes without initially owning it, to disrupt industries (e.g., FlightCaster, Decide.com).
  • Success Through Combination: The most successful companies combine data access, analytical skills, and innovative thinking.

What are the main risks and challenges of big data discussed in "Big Data" by Viktor Mayer-Schönberger?

  • Privacy Erosion: The scale and depth of personal data collection make traditional privacy protections inadequate, and anonymization is often ineffective.
  • Predictive Punishment: Using analytics to penalize individuals based on predicted behavior raises ethical and legal concerns about justice and free will.
  • Overreliance on Data: The “dictatorship of data” can lead to decisions that ignore context, causality, or human judgment, with historical examples as warnings.
  • Need for New Governance: The book calls for new frameworks to manage these risks responsibly.

How does "Big Data" by Viktor Mayer-Schönberger address privacy and propose new governance solutions?

  • Limitations of Consent: The traditional “notice and consent” model is inadequate for big data, as future uses of data are often unforeseeable.
  • Accountability Shift: The authors propose holding data users accountable for their actions, including formal risk assessments and regulatory oversight.
  • Technical Solutions: Innovations like differential privacy are discussed as ways to protect individual identities while preserving data utility.
  • Algorithm Auditing: The book advocates for a new class of professionals (“algorithmists”) to audit and certify algorithms for fairness and transparency.

How does "Big Data" by Viktor Mayer-Schönberger describe the changing role of expertise and decision-making in the big data era?

  • Decline of Traditional Expertise: Data analysis skills and the ability to interpret correlations are becoming more valuable than deep subject-matter knowledge.
  • Rise of Data-Driven Decisions: Organizations increasingly rely on predictive models and analytics, reducing dependence on intuition or gut feeling.
  • Interdisciplinary Skills Needed: Mathematics, statistics, programming, and data science are foundational, complementing domain knowledge.
  • Examples in Practice: Sports teams, tech companies, and city governments are highlighted as early adopters of data-driven decision-making.

What are the most memorable quotes from "Big Data" by Viktor Mayer-Schönberger and what do they mean?

  • “To measure is to know.” — Lord Kelvin: Emphasizes the importance of quantification for understanding, a theme central to big data’s expansion of measurement.
  • “More trumps better.” — Big Data Principle: Suggests that large, messy datasets often yield better insights than small, clean ones, challenging traditional data quality norms.
  • “Big data is about what, not why.” — Core Insight: Highlights the shift from seeking causal explanations to focusing on predictive correlations.
  • “Data is a platform.” — Tim O’Reilly: Underlines data’s foundational role in enabling new products, services, and business models in the digital economy.

About the Author

Viktor Mayer-Schönberger is a renowned expert in big data and internet governance. As a professor at Oxford University's Internet Institute, he has authored numerous articles and books on digital technology's societal impact. His work "Delete: The Virtue of Forgetting in the Digital Age" explores digital memory's implications. Mayer-Schönberger's expertise is sought after by global corporations and organizations, including Microsoft and the World Economic Forum, where he serves on advisory boards. His research and insights contribute significantly to understanding the evolving digital landscape and its effects on governance, regulation, and society.

Download PDF

To save this Big Data summary for later, download the free PDF. You can print it out, or read offline at your convenience.
Download PDF
File size: 0.44 MB     Pages: 24

Download EPUB

To read this Big Data summary on your e-reader device or app, download the free EPUB. The .epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.
Download EPUB
File size: 1.42 MB     Pages: 29
Want to read the full book?
Follow
Listen
Home
Swipe
Library
Get App
Try Full Access for 3 Days
Listen, bookmark, and more
Compare Features Free Pro
📖 Read Summaries
Read unlimited summaries. Free users get 3 per month
🎧 Listen to Summaries
Listen to unlimited summaries in 40 languages
—
❤️ Unlimited Bookmarks
Free users are limited to 4
—
📜 Unlimited History
Free users are limited to 4
—
📥 Unlimited Downloads
Free users are limited to 1
—
Risk-Free Timeline
Today: Get Instant Access
Listen to full summaries of 26,000+ books. That's 12,000+ hours of audio!
Day 2: Trial Reminder
We'll send you a notification that your trial is ending soon.
Day 3: Your subscription begins
You'll be charged on Nov 24,
cancel anytime before.
Consume 2.8× More Books
2.8× more books Listening Reading
Our users love us
1M+ readers
Trustpilot Rating
TrustPilot
4.6 Excellent
This site is a total game-changer. I've been flying through book summaries like never before. Highly, highly recommend.
— Dave G
Worth my money and time, and really well made. I've never seen this quality of summaries on other websites. Very helpful!
— Em
Highly recommended!! Fantastic service. Perfect for those that want a little more than a teaser but not all the intricate details of a full audio book.
— Greg M
Save 62%
Yearly
$119.88 $44.99/year/yr
$3.75/mo
Monthly
$9.99/mo
Start a 3-Day Free Trial
3 days free, then $44.99/year. Cancel anytime.
Unlock a world of fiction & nonfiction books
26,000+ books for the price of 2 books
Read any book in 10 minutes
Discover new books like Tinder
Request any book if it's not summarized
Read more books than anyone you know
#1 app for book lovers
Lifelike & immersive summaries
30-day money-back guarantee
Download summaries in EPUBs or PDFs
Cancel anytime in a few clicks

Why one summary?

Scanner
Find a barcode to scan

⭐
⭐
⭐
We have a special gift for you
Open
38% off
DISCOUNT FOR YOU
⭐
⭐
⭐
$79.99
$49.99/year
Only $4.16 per month
Continue
2 taps to start, super easy to cancel
Settings
General
Widget
Loading...