. The software industry does not have a good track record of delivering systems on time and budget. In part this is due to weaknesses in software sizing and project effort estimating methods. This paper discusses how some of these weaknesses have arisen historically, resulting in differing views on some basic underlying concepts such as the meaning of software size, types of software size and their relationships, the distinction between software ‘size-drivers’ and project ‘effort-drivers’, the meaning of the weights assigned to the components of software functional size measures, and the meanings of Functional User Requirements and of Non-Functional Requirements. Several conclusions are drawn, particularly that the current definitions of the latter two concepts are misleading in how they deal with quality requirements. Some recommendations are made on the need for ‘coherent size/effort ecosystems’ to help improve project effort estimation, and to revise some existing concept definitions.
