The probability distribution of the age of references in engineering papers

It is argued that, from a statistical analysis of the references cited in the engineering literature, the age of the references cited by the engineering research community can be modeled according to the simple lognormal probability density function (pdf). A more natural estimate of the age of a paper is not the number of years elapsed since its publication but the logarithm of this number. Hence, the temporal distance between two references is better estimated by the ratio of their ages rather than by their difference. This pdf gives a time frame for gauging the period that is necessary to separate research with lasting value from research yielding more ephemeral results. It provides a temporal framework for describing the relevant information that authors extract from the literature. A well-structured research archive (or a collection of books about a specialty in a library) should contain papers or books (the choice of which is, of course, subjective) whose age is distributed proportionately to the lognormal pdf.<
>