Talk:Universal Diversity Patterns Across the Sciences

To get the most out of our week, it might be good for us to begin our discussions now so that we can hit the ground running on Monday the 23rd. As I looked over the Wiki site, this page seems the best to place this discussion.

Here are some initial questions to help initiate what I hope will be a vigorous pre-meeting discussion:

(1) When looking for universal drivers, how vital is it to concentrate on deceptively simple processes with the fewest limiting assumptions? Is it true that the more complicated the mechanism (or limitations, in terms of MaxEnt), the more likely it will be limited to a subset of disciplines?

(2) What are the potential impacts of turning a continuous PDF into quantized values, particularly in terms of sampling error? Might this process be responsible for some universal patterns? For example, will quantization of abundance generate a bias in sampling error, being greater at smaller abundances? If so, what limitations to pattern will this generate?

(3) Is it optimum for us to focus each day on a process and see how many diversity patterns we can generate from it, or to focus on a universal pattern, and consider the potential universe of fundamental drivers?

(4) What would you suggest as an itinerary?

(5) What specific outcomes would you like to see?

Please feel free to respond to these questions, add in your own, or contribute whatever content you are inspired to upload.

Jeff

---

I'm not sure how this process works, so I will make a comment here and you can tell me if I placed in an appropriate location at the web site. (And also argue against the comment!) I am not sure if the comment is properly saved and accessible to you, so please feedback to me.

I think that by posing the issue as: "... the ultimate mechanisms underlying them would seem to be more general than those typically offered or considered by disciplinary researchers..." we need to be careful not to overlook the possibility that it is the universal irrelevance of mechanism (to first approximation), not a universality of mechanism, that creates putative universality of pattern.

Cheers, John

---

I agree completely with John here, but would argue that perhaps part of our task is to help enlarge the term 'mechanism' to also include mathematical/logical/thermodynamic constraints. For instance, I have no problem in calling the Central Limit Theorem a 'mechanism' as it does generate pattern, albeit in a way different from the typical explanations for pattern provided by most disciplinary explanations.

Jeff

---