How digital products got from the network to your pocket—and what matters now.

Digital products did not begin as apps. They grew from a simple idea: let different computers exchange information, then make that exchange useful to more people.

That history matters because each technical leap solved one problem and exposed another. The internet made networks interoperate. The web made information easier to publish and navigate. Phones made services more personal and present. Today, designers can build with powerful cloud and AI tools, yet affordability, reliability, language and trust still determine whether a product works for someone.

First came the network of networks

Early computer networks were separate systems. The important shift was to make them communicate across their differences. In January 1983, ARPANET moved to TCP/IP, a shared set of protocols that let independent networks exchange data. The Internet Society describes this move as a key step in the internet’s growth from a research effort into a broader network of networks.

For product builders, the lasting lesson is interoperability. A useful system should work with the world around it: existing devices, other services and the unpredictable path between a person and a server. That principle still shows up in open formats, accessible links, API design and graceful failure when a connection drops.

Then the web made information navigable

In 1989, Tim Berners-Lee proposed the World Wide Web while working at CERN, where researchers needed a better way to share information across computers. CERN made the web software available in the public domain on 30 April 1993. A readable page and a link lowered the effort needed to publish, find and follow information.

The web’s breakthrough was not only technical. It gave people a common way to move between documents without needing to know how each computer was organised. That changed the design question from “How do I reach this machine?” to “What information or task should this page help me understand?”

The pocket computer changed the setting

As capable phones became everyday tools, digital products moved from a place someone visited to something they might use while travelling, working, waiting or talking with another person. Small screens encouraged focused tasks. Cameras, location and notifications added new possibilities, but also new privacy and attention costs.

This is where context became part of interface design. A product used on a phone may need to open quickly, make its state obvious, save a person’s work and explain what will happen when service is interrupted. “Mobile first” is only useful if it also means “designed for the conditions of mobile use.”

The status quo: connection is common, but not equal

Connectivity has expanded, but access alone does not tell us whether people can use a service well. The International Telecommunication Union estimated that about 6 billion people—74% of the world’s population—were online in 2025, while 2.2 billion remained offline. Its regional analysis found that in Africa in 2024, internet use reached 57% of the urban population and 23% of the rural population. Those are regional figures; they do not describe every country or community.

The gap is a reminder that network coverage is one part of the experience. Data cost, device capability, electricity, digital skills, language, safety and confidence can all shape whether a service is practical. A product may be technically available and still be out of reach at the moment someone needs it.

At the same time, cloud platforms and generative AI have made it faster to prototype and deliver software. Faster production does not decide what deserves to exist. The useful work remains choosing a real need, respecting people’s data, checking what a system says and making sure the product behaves well when conditions are less than ideal.

Five questions to bring the history into your next build

1. What task is worth making easier? Describe a real moment and the current workaround before listing features.
2. What happens on a weak connection? Show loading, retry and saved-work states; do not leave people guessing.
3. What is the lightest useful experience? Keep the core journey clear on a small screen and avoid requiring more data or permissions than the task needs.
4. What must people be able to trust? Explain data use, identity, moderation and recovery in language people can find.
5. How will you learn whether it helped? Choose a user outcome and a proportionate way to collect evidence, then improve from what you learn.

The history is still being written. The next generation of digital products will be shaped by new tools, but the durable measure is older and simpler: can a person use the product to do something that matters, in the context they actually have?

Corrections: If you spot an error or a source that needs clarification, email [email protected]. We will date substantive corrections on this page.