How naval research and development worked in 1900

Research and development in 1900 meant building boilers, guns, machinery, hulls, and sometimes entire ships rather than testing ideas in computer models, so major experiments could involve thousands of people and were difficult to conceal. British government support was scattered among service establishments and officers' private work rather than organized through a modern research agency. Companies preferred incremental projects with a plausible return, whereas a state could absorb a failure that might bankrupt a firm. Private boards could also authorize work more directly, while public projects had to survive Treasury and parliamentary scrutiny. Firms such as Vickers and Armstrong Whitworth could supply almost every part of a navy, army, or railway system from within one organization. Britain possessed several such combines, and their research budgets could rival the entire budgets of small states, but the Royal Navy usually bought or refined work that industry had already advanced.