The stated aim of the author is “to characterize the foundations for conceptual level human-computer interfacing, supporting collaborative envisioning.” The paper itself “grew out of a need [to] understand complex systems,” which “requires creative, productive interaction of multiple disciplines.” The paper covers a lot of ground, tackling many important and interesting issues related to complex systems such as abstraction, structure, relations, hierarchies, the fallacy of reductionism, the role of environment, the role of an observer, perception and conception, viewpoints, metaphors, emergence of thought patterns, emergence of system semantics, collective interactions, and interdisciplinary collaboration. Even category theory is mentioned, in one sentence.

Among other things, Wheeler properly observes the importance of general systems theory “which has been somewhat out of the mainstream since the 50s.” (However, Weinberg’s excellent book [1] is not referenced!) The author also stresses the importance of (and claims the use of) “a software engineering meta-principle, ‘when you find good, informal, implicit thought or work patterns, make them explicit!’” Nevertheless, his first starting point states:

The temptation to define complex systems ... does not give a clearly grounded meaning to the field. Thus we characterize them, weaker but more intuitive, in the spirit of our semantics and with the semantics we use, perfectly adequate.

While we may only agree with the author’s observations about incompleteness (“a complete model of anything is impossible”), it is difficult to agree with his statement that in complex systems, “subsystems are nondecomposable, interactions are a-priori arbitrary.” If this were so, then it would be quite difficult to analyze any (sub)systems. The paper’s presentation culminates in a biological example of envisioning G-actin polymerization.

Many concepts and constructs used or referred to are well known from the work of such philosophers and system thinkers as Bunge and Hayek [2], of such applied mathematicians as Goguen [3], and of analysts and designers using, for example, the reference model of open distributed processing. Unfortunately, none of the above were mentioned in the paper. The importance and role of mathematics in general, and category theory in particular, in the understanding of complex systems (this includes the systematic construction of patterns), and in representing this understanding by means of precise and explicit concept definitions, is emphasized by Hayek in general, and by Goguen [4] and Ehresmann and Vanbremeersch [5] in particular. Again, none of the above were mentioned in the paper, although the concept of “graphic depictions ... having a direct and obvious mapping to ... objects, their parts and their properties,” substantially used by Wheeler, was precisely defined in algebraic semiotics [4]. Precise definitions of this kind, absent in the paper, would certainly help in making explicit the semantics of statements such as:

Natural depictions are naturally isomorphic to a small number of naturally embodied, fundamental image schemas and metaphors: surface, location, path, force, map, movement, container, collections, sequences and structures.

The paper contains quite a few typographical and grammatical errors.