There is the budding edge of science (Kuhn’s paradigm shifting science vs normal science) which speculates beyond the empirical evidence. Science practice often overlooks (and sometimes intentionally suppresses) valid data that doesn’t agree with established scientific dogma. The history of science is full of such shenanigans. I don’t personally know, but IMHO almost all of conventional science will eventually be replaced by something else, as Newtonian Science has been replaced by Quantum Science, yet most of the equations of Newtonian Science remain as good approximations for most on-Earth situations.
Kuhn’s work is itself controversial, as Social Epistemology (a discipline where these issues are explored) demonstrates. I was influenced by Steve Fuller’s books, Social Epistemology (1988) and Thomas Kuhn: A philosophical History for Our Times (2001). The former explores the social underbelly of science and the later studies the social influences on Kuhn and analyzes the social responses to Kuhn. Fuller appears discredited today by his studies and defense of Intelligent Design, which he claims is quite distinct from Creationism. If the designer in ID is the autopoietic cosmos itself and not some external god or force, I can cite empirical evidence that might point to this. He is excluded as contributor to Social Epistemology: Essential Readings edited by Goldman and Whitcomb (2011) – which I have not read. The discipline of Social Epistemology has recently merged with the discipline of Information Science, and this is distinct from the discipline of Science and Technology Studies. Today there is an accelerating fragmentation of disciplines with gaps growing between them. And, all this at a time when the belief in empirical evidence is being challenged by major corporations and politicians.
Somewhere else I commented on the limitations of models that exclude humans, so I agree with the criticism of the chemical analogy for a Global Brain. However, it is a common scientific practice to start with analogies and then attempt to modify them for new situations. Within a frame of essential complementarity (which I tentatively support) we might need more than one, logically consistent explanatory system; where only one system can be operational when situational specific.
Paradigm shifting science has many failures and a few outstanding successes. On the other hand, Kuhn’s work has multiple interpretations and remains controversial. Andrew Pickering’s THE MANGLE OF PRACTICE, Time, Agency, and Science (1995), and his later edited The Mangle in Practice: Science, Society, and Becoming (2008) cite detailed studies of how empirical researchers often shift many aspects of their research as they adjust process towards a meaningful conclusion. And I highly recommend his The Cybernetic Brain: Sketches of Another Future (2010) where he champion’s a process ontology over an existential ontology for science. I prefer to use the two ontologies as a complementarity. The upshot of it all is that the nature of science is evolving and if you look beneath the claims for superiority you will find exciting turbulence.
I don’t expect “perfection” in published documents. I have never read a doc where I haven’t noticed somethings “missing”; not so much error as incomplete. Since so much is inter-connected, it is probably impossible to craft/compose a single document (and even a web of documents) that is complete.
I keep remembering a statement from my studies in the philosophy of science:
FACTS ARE THEORY LADEN & THEORIES ARE CULTURE LADEN.
[extracted from an email]