Talk:Patterns
Contents
Different Definitions of Patterns
Lynn Skinner said "What you are looking for is not patterns in terms of image, but patterns in terms of life cycle, as in produce-use-recycle-reuse?!"
The concept of a pattern as a visual construct is related to Christopher Alexander's usage. both are recurrent and easily recognizable combination of elements. In the "Pattern Language" case, a pattern relates to how a set of problems is solved. Christopher Alexander talks about "City country fingers" as a way to integrate the urban and rural environments. At a more detailed level, he talks about "Light on two side of every room" to capture what he sees as a key element that determines the success and failure of a room. If you believe that this pattern is essential, it drives a particular way of laying out rooms within a building.
'But' Section of Pattern Template
Including positive examples in the 'But' section seems to water down the intent. Would it make more sense to include these in the 'Therefore' section, and leave 'But' for more cautionary consequences and counter-examples?
Discussion with Tom McKeag on Pattern Structure
--Tom McKeag (Dec. 26/2006) I hope this isn’t re-tilling old ground but I can’t help but step back to the broader question of organization of the study before I am able to respond to some of the contributions made already to “the list”. Indeed, my visit to the wiki site and wondering how to jump in and contribute in a meaningful way has prompted these thoughts:
- Hierarchy: the current list is in a matrix format but has no hierarchy of categories of patterns. I would propose the following general subheadings: Energy, Information, Structure, Materials, Motion. These elements seem to represent our human notion of what constitutes physical reality, including “Time”. As you have already discussed, there will be, inevitably, overlap, redundancy and sometimes a bewildering array of interrelationships. I believe that these issues will remain forever insurmountable until these patterns are organized into smaller groups, however imperfectly. In sailor’s argot an undifferentiated list would be called a “hurrah’s nest: everything on top and nothing at hand” and this is how I feel about struggling to contribute to our current list. In order to progress I believe we need to agree on some kind of our own pattern for inquiry and hope you will consider this one. The conservation economy pattern map is an effective one, I believe, because of its generally dendritic and perceivable organization.
- Scale: I also believe that the scale column should be elevated to more of a structural part of this organizational scheme; the second axis in our matrix, if you will. It is a common organizing tool in biology education and for good reason. The fact that physical behavior of materials is markedly different at different scales is, in my opinion, further reason for this approach. General patterns that don’t apply at certain scales should be identified as such and our analysis will be the better for it. I would propose that this scale axis be incremented in the standard levels of biological organization (atom, molecule, organelle, cell, tissue, organ, organism, population, community, ecosystem, biosphere), rather than the general categories now in the spreadsheet.
--Norbert (Dec. 27/2006) Tom, thanks for the feedback! I can certainly add your list of categories to the Patterns Template. Julian Vincent had a similar list in his paper: Information, Energy, Time, Space, Structure, Substance. I think Energy is a bit more general than Motion, and I would concur with adding Time to the list. As far as scale is concerned, I tried to use more neutral terms that would be accessible to a non-biologist. On the other hand, I suspect your list includes information beyond just size. Can we abstract what makes each level of biological organization different (other than size)? Although I am familiar with most of the terms, I tend to get confused between things like population and community.
--Tom McKeag (Dec. 27/2006) I prefer Julian’s list and believe it should be used as part of our organizing structure. I look forward to reading his paper.
As to scale, yes, you are right that the inference is that it is about more than size. One of life’s organizing principles that we have taught (from Hoagland, Exploring the Way Life Works) is the concept of “emergence”. As you know, life is more than the sum of its parts, indeed it is the product of these parts and superior levels of organization exhibit completely new properties impossible to predict by an analysis of the individual components at the lower scale. I would love to know more about these differences at different scales.
As I understand it: population is a collection of the same species; community is a collection of different species relating to each other in a particular place in time; ecosystem is a collection of the biota and the non-living resources that support the living in a particular place in time.
Fil's thoughts on patterns in nature and intent
--Fil Salustri 13:21, 6 March 2007 (EST): Norbert and I had a short email discussion some time ago, that I think is relevant here. Feel free to delete if it ain't.
Let me start with some loose definitions. Caveat: these come partly from my engineering background.
- A process is what we do to achieve a goal.
- Processes are divided into tasks.
- Each task follows a method of some sort.
- Put another way, a method is an abstract description of a way to solve problems. Methods are instanced in tasks, which are then assembled into processes.
Now, since patterns are generative, they have to describe how to solve the problem. So they can be thought of as methods. Instancing a pattern is instancing a method. Patterns are strung together to make up abstract process descriptions. But there's more than one way to wire up patterns together. This leads to a network structure of the patterns. This network is a pattern language.
A pattern has an implied goal - or intent - to solve a particular kind of problem. Otherwise, why bother?
I'm not laying this down as law here, but if we assume this is all true, then a pattern in nature is a description of a method that nature uses to solve problems.
I'm really uneasy with this. I don't believe nature has intent and I don't believe nature solves problems. I believe nature is just a dynamic system that's constantly changing in response to a relatively simple set of rules.
So where does that leave patterns in nature? Have I just painted myself into a corner? I don't think so.
I think we have intent to solve problems. So we can use patterns to capture those methods. The nature part comes from the principles used to guide those patterns. In other words, our patterns in nature should be design methods that embody certain principles that we find in nature and that seem to characterize what we mean by sustainability.
This also leads naturally to pattern languages being specific to particular domains (e.g. building houses).
Whether some kind of pattern meta-language can be written to capture the underlying principles of sustainability seems a bit too Godelian for me. But I'm just an engineer. :-)
--Fil Salustri 11:27, 8 March 2007 (EST): Thoughts on nature, evolution, and the Prisoner's Dilemma
Norbert mentioned the Prisoner's Dilemma (PD) in an email, where he wrote "The Prisoner's Dilemma is one of one that comes to mind, where a purely self-centred approach minimises both player's gains."
Now, natural instincts (e.g. survival, etc) generally indicate that natural critters are self-centred. So why does nature seem to work, per the PD, to minimise organisms gains? I can think of 2 responses.
- Minimising (without negating) gains puts a natural damper on things, so that growth doesn't happen too fast, which could be more harmful in the long run.
- Critters are self-centred because they're in a constant battle with their environment for survival. Balance again: organism vs environment.
This might work for nature, but we can do 2 things nature can't: we can predict and plan. Damping rapid growth helps prevent unforeseeable spikes in circumstance/environment. But we can (in principle, at least) predict how things can go, and then plan for different eventualities. So we can plan for rapid growth if we want to and not necessarily end up mucking things up as badly as nature might if its dampers came off.
Again, tho, this comes back to adaptability. It's one thing to plan, but plans are never perfect nor are they based on 100% reliable information. So we have to plan for alternatives and constantly revise those plans as new information becomes available.
I think this may be a root method of sustainability: to constantly plan & revise. It might even be a pattern (!)
