“The old computing was about what computers could do; the new computing is about what users can do.”
Ben Shneiderman opened his 2002 book Leonardo’s Laptop with this contrast between machine-centered and human-centered computing. Earlier technical narratives often emphasized processor speed, storage, networking, or the ability to automate a difficult operation. Shneiderman did not deny the importance of those achievements. He changed the success criterion. A system matters when people can use it to create, communicate, learn, make decisions, maintain relationships, or solve a problem they recognize. Capability inside the machine becomes a means; the person’s enlarged ability becomes the result that design should evaluate.
This shift is easy to state and difficult to practice. Technology teams can measure latency or benchmark accuracy more readily than confidence, understanding, or long-term benefit. Users are also not a single group with one goal. A feature that gives an expert fine control may burden a novice; automation that saves time for one worker may remove agency from another. Human-centered design therefore requires research across varied contexts, accessible alternatives, and observation of complete tasks. It cannot be reduced to asking whether participants like a screen or adding polish after technical decisions have already constrained the experience.
The quote also creates a demanding standard for AI. A model can generate fluent text or identify patterns, but those demonstrations do not show whether a user can verify the output, recover from error, protect private information, or make a better decision. A tool that produces more content may reduce capability if it increases review burden or undermines learning. Conversely, a modest feature that makes consequences visible and gives people control can be more empowering than a dramatic automation. Evaluation must include the human-system combination, the surrounding workflow, and the distribution of benefits and errors.
Shneiderman’s formulation remains a practical question for every product review: after the feature exists, what can a person reliably do that they could not do before? The answer should be concrete and testable. It may involve speed, but also independence, expression, safety, coordination, or access. The new computing he described is not a particular generation of hardware; it is a design orientation that must be renewed with each technology. Systems earn value when their internal sophistication becomes understandable external capability. The most impressive computer is not necessarily the one that acts alone, but the one that helps people act with greater knowledge and control. Longitudinal research adds another test: whether that capability remains beneficial after novelty fades and users adapt their habits around the tool.
Shneiderman used the sentence at the opening of Leonardo’s Laptop, a book proposing that technology design begin with human needs and activities rather than machine capability.
The contrast did not reject technical progress. It redirected attention toward creativity, communication, learning, and other outcomes that people should be able to achieve with that progress.
A benchmark measures what a system can produce under defined conditions; human-centered evaluation asks whether diverse users can accomplish meaningful tasks in real settings.
Users have different abilities, goals, and power. Design must examine who gains control, who carries new work, and who is exposed to error rather than assuming one universal experience.
AI demonstrations often foreground output quality while overlooking verification, privacy, learning, recovery, and the effort transferred to users who must supervise the system.
Evaluating the complete human-system workflow keeps technological sophistication tied to practical agency: greater ability to understand, decide, create, and act on one’s own terms.
Explore more "Quotes of The Day"
Discover more notable quotes from influential voices across politics, science, business, technology, sports, and culture. Each quote offers insight into how ideas, beliefs, and decisions shape the world around us.
