While I really like the article and its analysis, I think there is a more general point here - scale is tricky.
Most programming abstractions and tools (such as Logging / UI design / persistence) across instances are all "easy" when it comes to one machine (or a few machines) in a closed, deterministic environment.
All of these become extremely tricky when you have to deal with many instances across many different possibly indeterministic environments. For UI, it quickly becomes what's the viewport? What's the real estate? How is interaction (touch? mouse? keyboard? numpad?). For persistence, one has to worry about number of simultaneous connections, character encoding and normalization, read / writes per second, synchronisation, etc..
So, also for logging - Are we logging client side or over the network? what's the memory on the client? What else is running on the client? What's the permissions through which we log? Do we care about a system call on this client given any user interaction? etc..
Logging, UI, Persistence, Network protocols, heck even just choosing an intermediate data structure are all tricky .. at scale across devices and indeterministic environments.
Very well written article that allowed a non-biologist but engineer like me to follow. Systems Biology is quickly becoming one of the most interesting fields, IMO.
The observations about organisms catching up and slowing down is consistent with (the regression to the mean)[http://en.wikipedia.org/wiki/Regression_toward_the_mean]. Although it's something I only recently read about in Daniel Kahnemnan's Thinking Fast and Slow, it's applicability is widespread; so much so that I was not surprised seeing it in action here.
One thing that I do not fully agree with (or maybe that I do not fully understand) is the final evolutionary endpoint was measured simply by it's ability to grow in specific lab conditions. While this is a necessary condition for equality of endpoints, it is in no way a sufficient one. It is very possible, and even highly likely that while that goal of fitness is optimized for, evolution changes the final organism's way of reacting to other phenomena. I'm not sure if there is a better way to quantify equality (or some notion of edit distance) though.
From my experience, it's almost always both. There are things about the environment and work nature that are not conducive to the said person - thereby not allowing said person to function optimally.
Most times there are some things under the individual's control. Examples: Being more explicit with the manager about the kind of work that one would like to do, coming in an hour earlier (and leaving an hour earlier) to get some solitude in the mornings in an open floor plan, blocking calendars to disallow meetings at one's most productive time, etc..
But having said that, most relationships quickly become toxic when too many things start to fail. One is never sure if it is just oneself or the environment, if a certain action is just the boss having a bad day or part of a larger corporate thought process. At this stage, one is left with no choice to leave.
The best advice I can give is to take an extended vacation (10+ days) at such a point. Clarity often strikes when the immediacy of the problems are no longer at hand.
Most programming abstractions and tools (such as Logging / UI design / persistence) across instances are all "easy" when it comes to one machine (or a few machines) in a closed, deterministic environment.
All of these become extremely tricky when you have to deal with many instances across many different possibly indeterministic environments. For UI, it quickly becomes what's the viewport? What's the real estate? How is interaction (touch? mouse? keyboard? numpad?). For persistence, one has to worry about number of simultaneous connections, character encoding and normalization, read / writes per second, synchronisation, etc..
So, also for logging - Are we logging client side or over the network? what's the memory on the client? What else is running on the client? What's the permissions through which we log? Do we care about a system call on this client given any user interaction? etc..
Logging, UI, Persistence, Network protocols, heck even just choosing an intermediate data structure are all tricky .. at scale across devices and indeterministic environments.