The Information Tsunami: Big Data

Chaos vs. Meaning

More data does not automatically mean better decisions. The hard part is turning information into understanding.

Revised October 1, 2026

Collecting data is easy to mistake for understanding something. A business can know when you opened an email, which product you hovered over, and how long you hesitated before abandoning your shopping cart. It may still have no idea why you left.

That gap is the useful starting point for Big Data. The interesting question is not how much information we can accumulate, but what we can reasonably learn from it.

Chaos vs. Meaning

Big Data refers to more than a very large spreadsheet. The NIST definition connects the scale, speed, diversity, and variability of data with the need for systems that can store, process, and analyze it efficiently as those demands grow.

A sales database might organize transactions into familiar rows and columns. Other sources arrive as customer messages, audio recordings, images, video, or streams of readings from sensors. Bringing them together involves more than finding somewhere to put them.

The information may use different units, cover different periods, or describe the same person under several identifiers. Some records may be missing. Others may be wrong. A larger collection can give us more opportunities to discover a useful pattern, and more opportunities to mistake a mess for one.

The value lies in the relevant, coherent information we can derive from the data, and in whether that information helps us make a better decision. A warehouse full of numbers is still a warehouse. It does not become an explanation because the rent is impressive.

Put the Question First

Imagine a company trying to understand why deliveries keep arriving late. It could collect every available detail about its customers, vehicles, suppliers, and employees. Or it could begin with a question: where does the delay enter the process?

That question suggests what to examine: order times, warehouse processing, supplier arrivals, route conditions, and delivery records. It also suggests what to compare. Are delays concentrated in one region? Do they begin before the truck leaves? Is the promised delivery window realistic?

The example is hypothetical, but the distinction matters. Data collection should serve an investigation. Otherwise, the dashboard becomes an expensive way to watch the problem continue.

The Cloud Is Infrastructure, Not Insight

Cloud services can provide storage and computing capacity for this work. They are one infrastructure option; Big Data can also be processed on an organization’s own systems or through a combination of approaches.

Moving information to the cloud does not clean it, explain it, or decide which question deserves an answer. Nor does it make existing databases obsolete. Structured records remain useful, often precisely because somebody has already done the unglamorous work of organizing them.

The choice of infrastructure and the quality of the analysis are related decisions, but they are not the same decision.

Useful Applications, Ordinary Limits

The appeal becomes clearer when we describe a task rather than promise a revolution:

• Customer experience: compare transaction records with support messages to investigate where a service is failing.

• Supply chains: examine orders, stock levels, and transport records to identify recurring bottlenecks.

• Energy use: compare meter readings with occupancy and operating schedules to investigate unusual consumption.

• Fraud detection: flag unusual transaction patterns for further review.

These are possible uses, not guarantees. A suspicious payment may be legitimate. An apparent improvement may reflect a change in who was measured. Two trends moving together do not, by themselves, establish that one caused the other.

Where data describes people, there is another question to ask before celebrating its usefulness: should we be collecting and using it this way? Access to information does not settle questions of privacy, permission, or fairness.

More Is a Starting Point

Big Data can expand what we are able to examine. It cannot relieve us of deciding what matters, checking the quality of the evidence, or recognizing the limits of an interpretation.

A small, reliable dataset that answers a clear question can be more useful than an enormous collection assembled without a purpose. The ambition should be to turn the information tsunami into something we can navigate, rather than congratulate ourselves on getting wet.

What does Cloud Computing mean?



Revised in October 2026.

When we hear the terms the cloud or cloud computing, we tend to think vaguely of the Internet. That is not entirely wrong, but the two are not the same thing.

The Internet had existed for decades before people began routinely talking about “the cloud.” What changed was not the network itself, but increasingly what we were asking the network to do for us.

For years, using software usually meant installing it on a particular computer. You bought a program, loaded it onto the machine, and stored the files you created locally. The computer contained both the application and, in most cases, the work produced with it.

Cloud computing began loosening that relationship.

Applications, storage, processing power, and other computing resources could increasingly be provided remotely. Instead of asking What software is installed on this computer?, users could simply open a browser, sign in, and access a service running somewhere else.

The change can seem almost trivial because it happened gradually. Email moved into browsers. Documents could be stored online rather than on a hard drive. Music libraries became accessible from multiple devices. Businesses began using software without installing and maintaining it on every employee’s computer.

What disappeared was not software, but the requirement that the software, the data, and the user all occupy the same machine.

SaaS, PaaS and IaaS

Cloud services are commonly divided into three broad categories.

• Software as a Service (SaaS) is the form most ordinary users encounter. The provider runs the application, while the user accesses it through the Internet. Web-based email, collaborative office software, and many business applications follow this model.

• Platform as a Service (PaaS) provides developers with an environment in which they can build and run applications without managing all of the underlying infrastructure themselves.

• Infrastructure as a Service (IaaS) goes one level deeper. Computing power, servers, networking, and storage can be rented as services rather than purchased and maintained as physical equipment.

The categories differ, but the underlying idea is similar: some part of computing that once had to exist locally can now be provided remotely and on demand.

That is why the cloud is more than another name for the Internet.

The Internet is the network.

The cloud is what became possible when storage, software, and computing power no longer had to live on the machine in front of you.

Millennials and Multiculturalism

Revised in October 2026.

It may seem obvious, but it is worth paying attention to a simple dynamic: majorities shape trends, and trends often contain the seeds of their own perpetuation.

There is much discussion about the multiculturalism characteristic of the global era, and globalization itself provides an obvious explanation. Technology allows people separated by thousands of miles to communicate almost instantly and at negligible cost. Music, food, fashion, language, and ideas cross borders faster than ever before.

But that explanation can obscure another force operating much closer to home.

In the United States, one of the world’s major producers and exporters of popular culture, Millennials were the most racially and ethnically diverse adult generation in the country’s history when this article was originally written. In 2014, 43 percent of Millennial adults were nonwhite. At the same time, Census projections anticipated that the United States as a whole would become a country in which no racial or ethnic group constituted a majority around 2043.

Those figures naturally attracted the attention of brands and marketers. Diversity could be measured demographically: how many Hispanic consumers, how many Asian consumers, how many members of one group or another?

But demographic arithmetic captures only part of what happens.

Culture does not spread in a one-to-one relationship with ancestry. One additional Thai-American consumer does not simply mean one additional consumer of Thai food. People introduce one another to foods, music, customs, aesthetics, language, and ideas. Preferences migrate across demographic boundaries.

That creates a multiplier effect.

A generation that grows up with greater diversity does not merely contain more cultural backgrounds. Its members are also more likely to encounter those backgrounds through classmates, coworkers, friends, partners, neighborhoods, media, and increasingly through digital networks.

Culture therefore becomes harder to map neatly onto race or ethnicity.

Marketing strategist Eddie Yoon made a related argument in the Harvard Business Review: companies often make the mistake of treating culture as if it followed demographic categories automatically, when people's cultural affinities are also expressed through what they choose to consume and participate in.

Someone does not need to belong to the ethnic group that originated a cuisine to love its food, belong to the community that created a musical form to make it part of their identity, or share the background of a cultural tradition to participate in it.

This is where demographic change acquires consequences that simple population percentages cannot capture.

Greater diversity changes not only who is in the market, but also the network through which tastes move. People influence other people, who influence still others. What begins as demographic change can therefore become cultural change at a considerably larger scale.

Companies that continue to imagine consumers primarily as the predictable product of demographic categories risk misunderstanding that process.

The multicultural market is not simply a collection of increasingly diverse groups standing beside one another: it is what happens when those groups continually influence one another.

The Market in Global Times

Revised in October 2026.

The “David and Goliath Economy”

In the new global and digital paradigm, consumers and small businesses have acquired a degree of market power that would have been difficult to imagine only a few decades ago.

A single review, recommendation, complaint, or viral post can now help propel a small business toward success or seriously damage an established one. Platforms that connect consumers directly with businesses have dramatically reduced the cost of making preferences visible. Information that once moved slowly through advertising campaigns, market research, or word of mouth can now spread almost instantly.

The result is something resembling a David and Goliath economy.

Digital markets do not eliminate the advantages of size, but they can reduce some of them. A small company no longer necessarily needs a large advertising budget, extensive physical infrastructure, or an established distribution network to reach customers. In some markets, being small can even become an advantage: fewer sunk costs, faster decisions, greater willingness to experiment, and the ability to respond quickly when demand changes.

The rise of companies such as Uber and Airbnb offered early examples. Their initial strength did not come from owning enormous fleets of cars or portfolios of hotels. It came from using technology to connect existing supply with demand more efficiently.

That distinction matters.

Traditional companies often carry substantial investments in buildings, equipment, personnel, distribution systems, and established business practices. Those assets create enormous strength, but they can also make change expensive. A younger competitor may have fewer resources but considerably more freedom to redesign the way a service is delivered.

There is also a regulatory asymmetry. Established industries usually operate inside legal systems created around established business models. A new company may introduce something that does not fit neatly into existing categories at all.

Is Airbnb a hotel company if it owns no hotels? Is Uber a transportation company if it initially owns no fleet?

For a time, that ambiguity can itself become a competitive advantage. Regulation eventually catches up, as it should, but technological change often moves faster than the legal categories designed to govern it.

Work Becomes Part of the Same Transformation

The same forces are changing not only how companies compete, but also how people participate in the market.

Technology makes it increasingly possible for work to be separated from a particular office, employer, or even country. Independent professionals can sell services directly to clients thousands of miles away. Small companies can assemble teams without maintaining large physical offices. Specialized knowledge that once required access to a particular geographic market can increasingly be offered globally.

This creates opportunity, but it also transfers responsibilities.

A traditional company provides more than a paycheck. It may provide equipment, training, benefits, administrative support, job security, and contributions to social-insurance systems. As work becomes more fragmented among contractors, freelancers, platforms, and small businesses, some of those costs and risks move from institutions to individuals.

So the David and Goliath economy is not simply a story about small companies defeating large ones.

It is a broader redistribution of economic power.

Consumers gain more ability to reward and punish businesses. Entrepreneurs gain cheaper access to markets. Workers gain greater freedom to sell their skills beyond traditional organizational boundaries. Established companies lose some of the protection once provided by scale, geography, and control over distribution.

But none of these advantages is permanent.

Successful Davids eventually grow into larger players, while Goliaths learn to adapt, acquire smaller competitors, or imitate the innovations that once threatened them. Regulators catch up as well. And sometimes the very digital platforms that began by lowering barriers to entry become powerful gatekeepers themselves.

Perhaps that is the most important feature of the new market: power has not disappeared or simply moved from large companies to small ones. It has become more mobile.

And in an economy where information, capital, labor, and consumers can move faster than ever before, the lasting advantage may belong not to the biggest participant, but to the one most capable of adapting when that power moves again.

The Generations of the Global Era

Picture by Matthew G. via Flirck under Creative Commons

Revised in October 2026.

Rethinking Generations

Generations have traditionally been understood in largely regional or national terms.

The Spanish Generation of ’27, for example, described a particular literary and cultural movement. The American Baby Boom generation emerged from a specific demographic and historical experience following World War II. Such categories made sense because many of the events that shaped a generation were primarily local or national.

But globalization and digital technology raise an interesting question:

Does it still make sense to think about generations primarily in regional terms?

People growing up in different countries increasingly encounter the same technologies, platforms, entertainment, products, cultural references, and forms of communication. A teenager in Madrid, Buenos Aires, Seoul, or Chicago may inhabit very different political and economic realities while simultaneously participating in parts of the same digital culture.

That does not erase geography. Language, income, education, politics, religion, and local institutions still shape people profoundly. But geography may no longer monopolize the formation of generational experience.

At any rate, the notion of a “generation” has always been somewhat slippery. People born within the same twenty-year span do not necessarily share the same experiences, values, or historical consciousness, and the boundaries between one generation and the next are largely constructed after the fact. Globalization makes the category even less tidy. If people of the same age can now be shaped simultaneously by local events and by technologies, platforms, and cultural references shared across borders, then a generation may be less a clearly defined group than an overlapping set of experiences distributed unevenly across the world.

The Generations of the Global Era

The emergence of the Millennial label offered an early example.

Millennials were frequently described through their relationship with digital technology: widespread use of smartphones, social networks, online recommendations, and a growing reliance on peer opinion rather than traditional advertising.

The usefulness of those generalizations can certainly be debated. No generation is homogeneous, and someone born in 1990 in a wealthy American suburb did not experience technology in the same way as someone of the same age in a rural community with limited internet access.

Still, the fact that the category could plausibly be discussed across national borders was itself significant.

For perhaps the first time, technological change was creating experiences sufficiently widespread that researchers, marketers, and the media could describe aspects of a generation without beginning exclusively with nationality.

The common denominator was no longer necessarily a war, a political movement, or a demographic event.

It could be a technology.

Democratization and Connection

Mobile technology accelerated that change.

A phone gradually became much more than a device for calling someone. It became a camera, map, newspaper, encyclopedia, bank, workplace, storefront, library, and connection to social networks.

Cloud computing made personal and professional files accessible from almost anywhere. Digital payments began reducing the importance of cash. Social media gave individuals and small businesses access to audiences that once required significant advertising budgets.

None of this eliminated inequality. Access to technology, digital skills, capital, and reliable infrastructure remains uneven. But technology dramatically lowered the cost of participating in activities that once required far greater resources.

That change affects more than markets. It also affects how generations experience the world.

A generation can now be shaped simultaneously by local history and by global technological systems. Its members may vote in different elections, speak different languages, and live under radically different institutions while using the same platforms, reacting to the same memes, consuming the same entertainment, and adapting to the same technological disruptions.

Perhaps generations have not become truly global, but they are no longer entirely local either.

And that may require us to rethink what we mean when we call a group of people a generation.

The Social NetPlots

Revised in October 2026.

Click "I detest it"

Social networks and review platforms are proving to be fertile ground for a new kind of complaint and protest.

What makes them unusual is not simply their reach. It is their capacity to turn the very tools designed to evaluate a business into instruments for challenging it. Ratings, reviews, photographs, recommendations and even the description of what kind of establishment a place is can suddenly become part of the protest itself.

A particularly creative example emerged in a small Texas town in 2014.

Collin Dewberry and Kelley Williams stopped for breakfast at Big Earl’s Bait House and Country Store in Pittsburg, Texas. According to the couple, after they had eaten and paid their bill, a waitress told them they would not be welcome back and used an anti-gay slur. The restaurant owner disputed that their sexual orientation itself was the reason, saying their behavior had violated the establishment’s rules, although news reports also quoted him acknowledging that his daughter, the waitress, had used the slur.

The sign at the entrance offered some indication of the restaurant’s philosophy: men were expected to “act like men,” women to “act like ladies,” and the establishment reserved the right to refuse service.

Then the Internet got involved.

Instead of limiting their response to angry denunciations, Yelp users began rewriting Big Earl’s identity. They posted satirical reviews describing it as an extraordinarily welcoming gay bar. Google reviews and Facebook joined the spectacle, and the joke spread well beyond Pittsburg.

The cleverness of the protest was precisely in its inversion.

Rather than merely saying we dislike this business, users took the public identity the restaurant was trying to project and temporarily substituted an entirely different one. A consumer-review page had temporarily become a piece of collaborative satire.

The protest also began moving offline. Dallas LGBT activists announced plans to travel to Big Earl’s in person, turning physical presence into another response to the restaurant’s attempt to define who belonged there.

Yelp eventually removed many of the reviews because they were not based on actual customer experiences. From Yelp’s perspective, that makes sense: a review system stops functioning as a review system if ratings become votes on whatever controversy happens to surround a business.

But that tension is precisely what makes the phenomenon interesting.

Once people are given a public mechanism for assigning reputation, they do not necessarily restrict themselves to discussing the quality of the pancakes.

And businesses discovered the same possibility.

Only a few months later, California’s Botto Bistro deliberately asked customers to give it one-star Yelp reviews as a protest against the platform itself. The restaurant accumulated hundreds of intentionally terrible reviews, reportedly increased its business in the process, and turned the rating system into an advertising campaign against the rating system.

In both cases, the interesting development is the same: a platform created for one purpose acquires another because its users discover that its mechanisms can be repurposed.

A star rating can become a protest sign, a review can become satire, and a business listing can become a public stage.

Social media did not invent consumer dissatisfaction or political protest. What it changed was the cost and speed with which people could coordinate a response, and the number of existing systems they could appropriate to make that response visible.

Sometimes the most interesting innovation on a social network is not a feature introduced by the company.

It is something the users discover they can do with it.

Advertising in the Age of Ephemerality

Picture by Esther Vargas under Creative Commons License

Revised in October 2026.

You may have noticed a gradual trend toward brevity among some of the most popular online platforms, such as Facebook, Instagram, and, more recently, Snapchat, all characterized by an increasingly casual, brief, and sometimes fleeting style of communication.

Brands are being challenged to promote their products through applications where content may disappear within seconds. It is not an easy task for advertisers to come up with a message that is striking, interesting, and “sticky” enough to capture attention amid a waterfall of tweets, status updates, photographs, and conversations that are often fragmented and lacking in continuity.

Snapchat makes the challenge particularly interesting. Its conversations can be spontaneous, random, and short-lived, making it difficult even for automated systems designed to detect keywords to determine which advertisements might be relevant. Yet advertisers are beginning to include these platforms in their campaigns and, in some cases, to discover unexpected benefits.

Instagram, for example, introduced its first paid mobile advertisements in late 2013. Less than a month later, a report released by the company described the experiment as a success. Three of the four sponsors included in the report said they had achieved the brand perception they were seeking while reaching audiences of between seven and nine million users without flooding Instagram with advertising. In just nine days, Levi’s reached nearly seven million users in the United States, while Ben & Jerry’s achieved similar results, reaching close to ten million people during a campaign of roughly the same duration.

Although the information exchanged on these platforms is highly volatile, fleeting, and often chaotic, the possibilities for advertisers are considerable. Imagine the kind of information available through social media: a teenager may announce a new relationship, post photographs from a vacation, mention a nightclub they found unbearably boring, or casually discuss a product they like or dislike. Individually, these details may seem trivial. Together, they begin to reveal interests, preferences, habits, and intentions.

For advertisers trying to define the profile of their consumers, that information can be extraordinarily valuable.

The content may be ephemeral. Its value to advertisers may not be.