Why Does Software Cost So Much? And Other Puzzles of the Information Age (1995), a DeMarco solo collection published by dorset-house-publishing, gathers essays on software economics, management, and organizational dynamics written between the original peopleware (1987) and the organizational-dynamics-era works of the late 1990s and 2000s. It is the clearest window into DeMarco's thinking during the interval between his two most significant books.
The essay collection as intellectual bridge
The collection spans topics that would be developed more fully in slack and waltzing-with-bears: the economics of software quality, the cost of defects, the measurement of software productivity, the relationship between process and people in determining project outcomes. The title essay addresses one of the most persistent puzzles of the software industry — why software development remains so expensive and so prone to failure despite decades of methodological investment — and gives a version of the peopleware-thesis answer: because the problem is organizational and sociological, not technical and methodological.
The essays also address the coding-war-games data and its implications more directly than is possible within the narrative flow of Peopleware itself. Several essays develop the measurement arguments of controlling-software-projects in light of the Coding War Games findings.
Software economics arguments
The collection's economic analysis is characteristically empirical. DeMarco examines the actual cost structures of software projects, the distribution of effort across phases, the disproportionate cost of late defect discovery, and the economics of testing and quality investment. The argument for early testing investment (developed further in an-early-start-to-testing) appears here in economic terms: the cost of fixing a defect grows exponentially with time, making front-loaded quality investment economically rational even when it appears to slow early progress.
The relationship between this economic analysis and the slack-concept is implicit in the collection but becomes explicit in the later work: organizations that cut quality investment to reduce apparent cost are borrowing against a future that compounds interest rapidly. The pursuit of efficiency in the short term generates inefficiency — in the form of defect correction, rework, and schedule overrun — in the medium term.
Position in the DeMarco bibliography
Why Does Software Cost So Much? is a transitional work — less foundational than peopleware or structured-analysis-and-system-specification, less developed than slack — but valuable for understanding how DeMarco's thinking evolved between the two major phases of his career. The essays show him working through the organizational implications of the Coding War Games data and the limitations of the structured methods tradition he had helped establish, developing the arguments that would reach their fullest expression in Slack.
The collection also demonstrates DeMarco's range as a writer. The essay form suits his voice — direct, empirical, occasionally sardonic — better than the expository book structure, and the collection's best essays rank among his most readable work.