to collaborate or not to collaborate

A few quotes from a fascinating new book, Collaborative Computational Technologies for Biomedical Research (Wiley, 2011):

Even within an institution — which should be legally, strategically, and financially incented for alignment, and for maximizing the opportunities for internal collaboration — barriers still exist. The subunits of the institution: its departments, its divisions, its components produce collaboration “walls” of varying substantiality. Organizational lore and personal relationships add another layer of “not-invented here” (NIH) culture, and allegiances to local agendas, even to the point of disadvantaging the larger institutional unit. In fact, if we wish to pursue the elimination of collaboration barriers we have to realize that many barriers are not institutional at all. Choices to collaborate or not collaborate are sometimes based not just on current affiliations but on past affiliations, degrees obtained, reputations, and even a less than rational bias as to just who our collaboration partners should be.

systems vs system

‘Systems biology’ is an annoyingly waguely-defined area. Its recent history illustrates the self-perpetuating nature of organised science really well. According to Wikipedia,

systems biology refers to a cluster of peripherally overlapping concepts rather than a single well-delineated field. However the term has widespread currency and popularity as of 2007, with chairs and institutes of systems biology proliferating worldwide.

And then:

As of summer 2006, due to a shortage of people in systems biology, several doctoral training centres in systems biology have been established in many parts of the world.

Isn’t it brilliant? We don’t know what it is exactly but we have a shortage of experts in it already. Why, we don’t even know how to spell it. For instance, this publication from ESF uses both ‘system biology’ and ‘systems biology’. For once, I can’t blame the ‘systems biologists’ (whoever they are) for the confusion: there is a long tradition of using plural, rather than singular, in variety of ‘systems sciences’. J. P. Van Gigch wrote in System Design Modeling and Metamodeling (p. 32):

The reader will note that, in this text, we use the term “system” in singular when it applies to only one theory, one paradigm, one approach, one theory of design as in system theory, system paradigm, system approach, or system design, respectively. By contrast, we still use “systems” in plural, when the term “system” applies to more than one system, as in the expression “hard systems domains”, “soft systems domains”, or in “various systems assumptions”. This notation agrees with that recommended by the so-called father of this discipline, L. von Bertalanffy.

Convinced? I am not, really. The construction (plural qualifier followed by a singular noun) seems ungrammatical. We say “star formation” and “pest control”, even though it is understood that we mean more than one star or pest. What is so special about systems?

Why do we even have to mention the systems? What is a system anyway? In thermodynamics, a system is the region of the universe under study. Thus, by definition, thermodynamics studies systems. I like that definition because it is perfectly applicable to every natural science. Therefore, every natural science studies systems.

Ross Ashby, English pioneer of cybernetics, gave another definition:

A system is a set of variables sufficiently isolated to stay constant long enough for us to discuss it.