Evolution has its downsides… but maybe it’s inevitable?

In a recent Reuters article, the idea that Darwin’s Theory of Evolution might apply more widely than just to biological systems as we understand them was examined.

It’s an idea expressed in a paper in the Proceedings of the National Academy of Sciences (PNAS). On the roles of function and selection in evolving systems, Wong, Cleland, Arend Jr, and Hazen assert:

We suggest that all evolving systems — including but not limited to life — are composed of diverse components that can combine into configurational states that are then selected for or against based on function. We then identify the fundamental sources of selection — static persistence, dynamic persistence, and novelty generation — and propose a time-asymmetric law that states that the functional information of a system will increase over time when subjected to selection for function(s).

On the roles of function and selection in evolving systems

On the surface, this isn’t entirely surprising. There does seem to be a common tendency for things to endure and evolve in response to circumstances, exploring concepts of ‘fitness’ and ‘optimality’ under selective pressure. In networking, this brings to mind a couple of areas of thought.

For instance, the industry can probably expect more and more refined work in Internet protocol design, because the basic underlying network structures (IPv4 or IPv6, with TCP and UDP) are pretty much given as-is.

At best, expect to see only small functional changes to them as preserving backward compatibility limits the range of variation. This has been seen in the TCP congestion control protocol designs emerging over the last 30 years where more and more refined approaches to rate estimation and sliding window protocols are being applied ‘competitively’ on the Internet.

BBR has aspects that make it out-compete the other protocols for available bandwidth, but this can break down. So, there are niches for the older Cubic and Reno algorithms to persist alongside BBR-based protocols.

Or consider the refinement of IPv6. It initially promised an amazing range of Extension Header behaviours. However, practical challenges in intervening network systems have pushed IPv6 toward a more constrained definition of what it is to be a modern IP protocol with a large address space.  

In certain aspects, IPv6 is now more akin to TCP and UDP with Bigger Addresses (TUBA — an early design concept it originated from) than its initial launch as IPng.

Randy Bush recently reminded me of a quote from Jon Postel that has a less upbeat but alas most realistic take on the future of IETF standards development, which may well fit this model of selective pressures as a general trend:

It’s perfectly appropriate to be upset.  I thought of it in a slightly different way–like a space that we were exploring and, in the early days, we figured out this consistent path through the space: IP, TCP, and so on.  What’s been happening over the last few years is that the IETF is filling the rest of the space with every alternative approach, not necessarily any better.  Every possible alternative is now being written down.  And it’s not useful.

Jon Postel on the IETF, late 1990s

Jon’s critique may feel right, but the reality is that most of these alternative visions of protocols in the IETF won’t survive the rigours of the real world Internet; selection pressure is going to prune this back to a more consistent and evolving set of developments.

Rate this article

---

The views expressed by the authors of this blog are their own and do not necessarily reflect the views of APNIC. Please note a Code of Conduct applies to this blog.