2014/05/27 B3D Webinar: Unifying Principle
Randy Olson talked about the value of a 'unifying principle' in his ZQ09 interview along with Brian Palermo. A good unifying principle has three criteria:
- makes sense of and provides insights into the current set of observations
- creates opportunities for further exploration, research and practice
- is defensible
Evolution satisfies all of these criteria. Note that evolution does not explain all of biology - we are not talking about some sort of 'grand unifying theory'. It would be hubris to suggest that we can come up with a unifying principle for the breadth of depth of B3D. If we can come up with one that applies to an area of B3D, it could change how we think about B3D, focus how we communicate and encourage us to explore new opportunities.
A B3D webinar will be held on May 27th at 09:00 PDT, 10:00 MDT, 11:00 CDT, 12:00 EDT, 17:00 BST, 18:00 CEST and 19:00 EEST. The first part of the webinar will explore the concept of a unifying principle, the second part will be a brainstorming session on identifying a suitable B3D area and developing ideas that could become a unifying principle. Registration at https://business-hangouts.com/register.php?m=Nzk4OXwxMjY2MQ== is required.
Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License
Sustainable Innovation Network 2010-2012
Homework for next webinar on unifying principles?
We had a good discussion yesterday. My thanks to all the panelists (Ray, Janice, Shoshanah, Taryn, Catalina), commentators (Phillip), and of course Norbert.
We came up with a short list of (potential) unifying principles such as system, function, process (this is not an exaustive list). I think it would be useful to come up with examples of each of these unifying principles - it would help make our discussion more concrete and tangible. So may I request each panelist for the next webinar to come up with an example of each unifying principles he/she thinks is important? Thanks.
I think it would be great if this led to a collaborative article in ZQ.
Best,
Ashok
Sorry I missed it. I still
Sorry I missed it.
I still have to watch the video, but I fully endorse Ashok's idea of turning this into a ZQ article.
re: Homework for next webinar on unifying principles?
One approach to Ashok's homework assignment would be to look at how an idea for a unifying principle leads to novel insights and/or opportunities for further development of B3D. Vincent's energy/materials vs. information/structure comes out of his analysis of how humans and nature solve problems. There are lots of opportunities for figuring out how nature manages on a 'lean energy' diet. His 'make' vs. 'being' is one approach. Vincent made an interesting comment - I am following up with him to see if he has explored this further.
The idea of focusing at the level of relationships/systems lends itself to re-analyzing existing B3D examples in terms of yield and impact. Opportunities might include focusing on the evolving human relationship with our support systems. What opportunities exist for operationalizing relationships/systems? Examples might include Ray Lucchesi's Living Systems Design and Ernst-Jan Mul's Connection-Based Tools for Bio-Inspired Design. I am sure there are all sorts of trans-disciplinary connections we could explore.
Followup on Vincent's quote.
That quote by Julian stirred some ideas.
He wrote: "An organism that is embedded in a larger system tends to only concern itself with those parts of the system that directly affect it." That's not quite right. In a systems view, what's outside the organism doesn't matter; the only thing that matters is what crosses the boundary between the organism and the rest of the system. For instance, a plant doesn't care that the light it receives comes from the sun; it would react identically to any other light, so long as the light was the of the same type as that from the sun. So it's about the boundary and what crosses it, not what's on the other side of the boundary.
That is to say: Function happens at system interfaces.
He wrote: "...an organism that is outside of the system needs to understand how the system works." This made me think of something I'd never thought before (though it may be old news to others, in which case I apologize).
If indeed function happens at interfaces, then perhaps a new way to approach BID is to think purely in systems. We don't want to mimic a gecko's feet, for instance; we want to model all the functions (i.e., the interfaces) that occur between the gecko's feet and the thing the gecko is climbing. We then abstract those models to purely physical/chemical descriptions, then use those as an encyclopaedia of sorts of natural solutions.
Caveat lector: I've only just thought of this, so it's quite raw and may reflect nothing more than my ignorance of BID.
re: Followup on Vincent's quote.
Fil, I think Vincent is referring to the degree of connectedness. Stuart Kauffman has done numerous simulations of the behaviour of systems based on the number of nodes and the connectness of the nodes. Too few connections and the system 'freezes' (is unable to adapt to change). Too many nodes and the system becomes chaotic. I recall Kauffman came up with an optimum connectedness of a shade more than 2 for systems with a large number of nodes and was able to find real-world examples that matched his simulation results. I am not sure if anyone has done any exhaustive evaluation of real-world connectedness although there is a lot of anecdotal evidence.
re: Function happens at system interfaces
Could functions that have no physical/chemical description arise from the interaction? That might open up some interesting possibilities.
Well, depends on what you
Well, depends on what you want to model.
As my background is in engineering, I look at it like this:
Behaviour is the response of a structure to an input. Behaviour is a (logical) function that maps (structure + input) to output. The inputs and outputs are the things that cross the system's boundary. Systems are encapsulated by their boundaries, so all any other system "sees" are the outputs from the system in question. Note that so far, nothing has been said about the system environment, except to assume that one exists from which inputs come and to which outputs go.
Function is the role that a behaviour plays in a system of systems including the one in question. Function doesn't happen without behaviour. So while function can be described in whatever way you want, because it's rooted in behaviour, it can be tracked back to the stuff that crosses the boundary of the system in question.
Here's an example. My son Gianluca loves the colour teal, and loves Volkswagen Golfs. There's a guy up the street who has painted his Golf a flat teal. (I kid you not.) Every time we drive past the guy's house, Gianluca smiles. So one might say that one function of the Golf is that it makes Gianluca happy. That's not a physical/chemical thing. However, when you want to understand how that happens, you have to look at the interactions between the Golf and Gianluca, and those are described by physical/chemical descriptions.
Sorry; that was a bit of a rant.
I have got to read that book!
I have got to read that book! :-)
My first question would be What is a "connection" (per Kauffman). It sounds a lot like a system interface, but I doubt it's exactly the same thing. Don't feel obliged to answer. It will motivate me to crack the spine (virtually) on that book.
re: I have got to read that book!
Fil, At Home in the Universe describes his work using simulations and possible implications while The Origins of Order (written two years earlier) has more 'real-world' examples. I don't think Kauffman actually defines 'connection'. He starts talking about autocatalytic sets modeled by network diagrams where the nodes are chemicals and the lines are reactions. His simulations involve virtual light bulbs (two states) which accept inputs from other light bulbs (number of inputs relates to connectedness) and process the inputs using Boolean functions which determines their target state. Kauffman is primarily interested in system-level attributes, particularly relating to adaptability and stability. I recall he talks a lot of the actual length of state cycles compared to the total number (very large as the number of virtual light bulbs increases) and phase transitions where small modifications lead to large changes in the system attributes. Although the models were very simple, they often lead to complex behavior and allowed Kauffman to explore ideas such as co-evolution and the emergence of order.
Thanks for the
Thanks for the pointers.
Seems to me he's studying/describing systems for which fairly complete information is available and quantitative in nature. Scaling that up would run into some tractibility problems. I prefer the approach Meadows took in Thinking in Systems, which is more macroscopic and can involve a variety of simplifications. You can start with a simplified but still relevant model, then gradually add more elements.
In any case, I've added both his books to my reading list.
...Good thing I've got a sabbatical coming, or I'd never get to any of 'em.
Notes from May 27th Unifying Principles Webinar
This webinar (Google+ post) discussed the concept of a 'Unifying Principle' or narrative (as proposed by Randy Olson in an interview for ZQ) and included a brainstorming session on ideas that could lead to unifying principles for B3D. A copy of the charts are attached.
B3D Webinar: Unifying Principle
00:55 Concept of a 'unifying principle' (Olson/Dobzhansky)
02:10 Criteria and examples from evolution (The Origin of Species)
02:31 ... Insightful
04:01 ... Creates Opportunities
04:39 ... Accessible
06:14 ... Defensible
07:21 ... Asks Fundamental Questions
09:09 Not a 'Grand Theory of Everything'
08:40 Ideas for a 'Unifying Principle of B3D' (out of sequence)
09:41 ... Nature's Storehouse of Innovation (Biomimicry 3.8)
10:19 ... Energy/Matter vs. Information/Structure (Julian Vincent)
11:41 ... 'Making' vs. 'Being' (Julian Vincent)
12:47 Discussion
48:50 Action Plans
Discussion: (times are offsets into webinar)
14:59 Ashok: shift from organisms to systems, the organisms' interaction with other organisms and their context, different way of doing bio-inspired design; Norbert: potentially more 'points of intersection' between biology and technology as we move from organisms to systems (dealing with more fundamental aspects that may be less dependent on similar contexts)
18:30 Shoshanah: scale/hierarchical levels appear to be relevant to successful emulation, relationships important but still unified by the concept of 'function' which can be manifested at different levels; Janice: inter-dependent relationships at the systems level that incorporate functionality seen at the organism level
22:38 Ray: shift from looking at the organism ('thing') to the context, working in the context to understand the 'thing', "making' <-> function/problem solving while 'being' <-> understand as living system/appreciate its essence/evolving relationships; Janice: dangers in focusing on function without appreciating the relationship; Norbert: focus on the evolving human relationship with the systems that support us, better likelihood of avoiding issues with incompatible contexts, more likely to be able to initiate change
27:24 Ashok: challenge of operationalizing a shift from organism -> system or function -> relationship, unclear where to draw the system boundaries, limited literature; Taryn: applicability of 'planetary boundaries' at a larger scale; Philip: interdependence between organisms and inanimate surroundings; 33:18/Shoshanah: often confuse community (grouping of species/organisms) with ecosystem (communities interacting with the abiotic environment), abiotic environment applies across contexts and acts to 'level the playing field'
32:22 Taryn: evolution as unifying principle for B3D, sustainability in the sense of human evolution, focus on evolutionary process; Shoshanah: evolution is fundamental but does not characterize the uniqueness of B3D, evolution not always effective/efficient (we need to be selective); Taryn: Life's Principles, evolution as mutation/replication of genes/combination of genes, need to be taken together to make sense
38:24 Ashok: tendency to focus on the product (of evolution) rather than the process (of evolution); Shoshanah: evolution is slow/wasteful/random making it hard to emulation, focus on retrofitting/working with what already exists, consider other biological processes, processes very relevant when looking at application of B3D in different industrial sectors (form -> interaction)
41:12 Ashok: designing products really involves designing a human behavior/interaction, tend to focus on product/process design rather than the ultimate interaction; Taryn: products of evolution may not be sustainability but the process of things evolving has been sustainable at a certain level; Shoshanah: outcome of natural selection more interesting that those of evolution; Taryn: process of trial and error guiding good/appropriate design; Ashok: all design is dynamic/iterative/evolutionary; Shoshanah: changing environment can shift detrimental mutation to an adaptive one, dynamic relationship between the product and the environment
48:50 Action plans: schedule another brainstorming session for June 17, potential for a paper/ZQ article, test ideas against criteria, progress from the idea to developing to viable unifying principles,
52:33 Ashok: homework assignment - one concrete example that instantiates the ideas (for example, how knowledge of systems has been transferred to human domain)