“Fish are like trees…” :  the original statement was…

“Managing fisheries is hard: it’s like managing a forest, in which the trees are invisible and keep moving around”

(from an unpublished lecture at Princeton University, ca 1978)

“The real goal of fisheries management is to avoid finding out what the stock-recruitment relationship is. Once you have depleted the stock enough to know, it’s probably too late”

(from an argument with David Cushing, ca 1980)

Shepherd’s Law of Legibilty: The probability that a document will be read is approximately 1/N, where N is the number of pages.

(First enunciated ca 1990)

“Those who are ignorant of the past are condemned to misunderstand the future”

First used in a lecture promoting the study of past climate, ca 2004)
(re-working of “Those who cannot remember the past are condemned to repeat it” by philosopher George Santayana, in The Life of Reason (1905).

“Asymptotically Sensible”        It is important that any mathematical approximation, representation, parameterisation or model should be “asymptotically sensible”. This means that when any  relevant variables are pushed to their extremes, the representation should remain credible, and not lead to infeasible or implausible results (e.g. it should remain positive, or finite, or tend to zero, or whatever asymptotic behaviour is appropriate). This places a strong constraint on the form of such representations. Those based on underlying theoretical considerations are more likely to have this property: polynomial approximations usually lack it !   (used ca 1980 et seq.)

see e.g. https://archive.nafo.int/open/sc/1993/SCR-93-123.PDF