Cataloging a year of metamodeling blogging

Last Saturday, with just minutes to spare in the first calendar week of 2019, I shared a linkdex the ten (primarily) non-philosophical posts of 2018. It was focused on mathematical oncology and fitness landscapes. Now, as the second week runs into its final hour, it is time to start into the more philosophical content.

Here are 18 posts from 2018 on metamodeling.

With a nice number like 18, I feel obliged to divide them into three categories of six articles each. These three categories: (1) abstraction and reductive vs. effective theorie; (2) metamodeling and philosophy of mathematical biology; and the (3) historical context for metamodeling.

You might expect the third category to be an after-though. But it actually includes some of the most read posts of 2018. So do skim the whole list, dear reader.

Next week, I’ll discuss my remaining ten posts of 2018. The posts focused on the interface of science and society.

So first, what is metamodeling?

At is most local, it is the metamodeling-tag on TheEGG. But that isn’t that informative. Etymologically, it is making models of models. This is too restrictive if you limit to formal models of formal models. Thankfully, both others and I long considered mental models as a perfectly reasonable kind of model that theorists try to extract from their experimental counterparts. This would make metamodeling into thinking about models. Potentially too broad, but closer to what I mean.

In the end, I’ll stick to “I’ll know it when I see it”.

Abstraction and Reductive vs. Effective Theories

One of the goals of mathematical biology is to enhance the way that experiment and theory are stitched together. This requires both designing new kinds of experiments and building operationalized — or effective — theories. I’ve been trying to do this with cancer research and EGT for about 5 years — and maybe longer in other fields. But only in 2017 did I realize that the right language is hidden in the distinction between reductive and effective games.

From the get-go, my distinction between reductive and effective games was linked to two different conceptions of fitness. However, in my first draft, I handled this distinction poorly by talking about properties of individuals vs populations. For many, this brought the levels of selections debate to mind, something that I was not trying to engage with.

1. Token vs type fitness and abstraction in evolutionary biology (April 13th)

After reading some philosophy of biology, I realized how to re-frame these two kinds of fitness. In the above post I did that: token vs type fitness. In the next couple of weeks, I’m hoping to finally have the second draft of this work out that incorporated the insights from that blog posts.

2. Abstract is not the opposite of empirical: case of the game assay (June 2nd)

In general, the process for translating token fitness into type fitness can be very complicated. As such, we can view our experimental system as an implementation of our abstract type fitness. In the post above, I use this as a means to think about our game assay as an empirical abstraction.

Once I was in this mindset of abstractions, I used the rest of the year to explore how it relates to others concepts that I’ve discussed before, including heuristic models, in silication vs heuristic approaches to modeling, spatial structure, and reductionism.

3. Heuristic models as inspiration-for and falsifiers-of abstractions (July 14th)
4. Bourbaki vs the Russian method as a lens on heuristic models (November 10th)
5. Effective games from spatial structure (December 7th)
6. Reductionism: to computer science from philosophy (December 29th)

This made up the more specific studies of different approaches to modeling in EGT and the context of my own work.

Metamodeling and Philosophy of Mathematical Biology

The explorations in the above specific content would not have been possible without broader reading and thinking in the philosophy of science. In particular, the philosophy of mathematical biology.

1. A month in papers: mostly philosophy of biology (May 31st)
2. QBIOX: Distinguishing mathematical from verbal models in biology (June 9th)
3. Personal case study on the usefulness of philosophy to biology (June 30th)
4. Methods and morals for mathematical modeling (October 6th)
5. Mathtimidation by analytic solution vs curse of computing by simulation (October 13th)
6. The Noble Eightfold Path to Mathematical Biology (November 3rd)

The last post might as well be attributed to Rob Noble instead of me. It was inspired by his template for doing good work in mathematical biology. All I did was add two steps, some explanatory detail, and named it after Rob. I hope his eightfold path catches on. That would greatly benefit mathbio.

Historical context for metamodeling

Alongside a general philosophical context for modeling, I also find it useful to look at history. This is particularly helpful for disenchanting concepts that seem obvious or natural. For this, it helps to find a time when they were not obvious and were first introduced.

1. Double-entry bookkeeping and Galileo: abstraction vs idealization (June 15th)
2. Algorithmic lens as Alan Turing’s wider impact (June 23rd)
3. John Maynard Smith on reductive vs effective thinking about evolution (July 17th)
4. Darwin as an early algorithmic biologist (August 4th)
5. Hobbes on knowledge & computer simulations of evolution (August 25th)
6. Overcoming folk-physics: the case of projectile motion for Aristotle, John Philoponus, Ibn-Sina & Galileo (September 22nd)

I was surprised by how popular these historic posts turned out to be.

My guess is that there is that too many people view modeling as obvious and modern and don’t think there is much to be gained from a historical perspective. Thus, most people don’t write such perspectives. And when one is provided, people enjoy the story.

But I could be wrong. After all, I am often not an accurate predictor for the popularity of my own posts.

This makes the writing more fun, since it turns the reader response into a surprise. It also results in surreal experience like redditors explaining my own post to me. A nice reminder of how my intended and delivered meaning don’t always align.