Brittle sounds like fragile. This article reminds me of Nassim Taleb's amazing book, Antifragile.
Fragile organizations and systems break from randomness.
Robust ones are able to resist and stay strong.
But Antifragile ones actually benefit from randomness and volatility.
I highly recommend the book -- it's highly relevant to startup companies who, rather than shooting for the moon, should protect their downside (at first) and place small bets on numerous 10x activities.
I would like to ask you what you found so compelling about Antifragile. When I read the book, I found it simply repeated the claim using various examples that were specific to Taleb's life: [My career as an author is antifragile because I could beat up an economist and I would gain notoriety and increase sales.]
However, I was unable to gain any insight into how that applied to typical situations, e.g. investing, career choices, romantic relationships, home purchasing. That is, beyond "do these things in an antifragile way, not a fragile way."
He also follows these things up with canned strategies to make your life antifragile, without making any attempt to demonstrate how or why these strategies are actually so.
Example: he does not recommend the standard investment strategy of purchasing stock funds and bond funds at a ratio that approximates your tolerance for risk. This is, for some reason, fragile. He recommends putting most of your money into the safest investments available on the market, with the rest of your money going into the riskiest moonshots you can find. This is, for some reason, antifragile. "Unimportant" questions like why and how these strategies are classified as fragile/antifragile and what kinds of investments he considers appropriately safe are left as an exercise to the reader.
On the flip side, I can’t recommend against reading the book enough. Your summary is pretty much enough to get the point across, minus endless pages of self-aggrandizement and throwing around preemptive ad hominem attacks toward any group he thinks might disagree with him.
I genuinely can’t understand how anyone could have made it through that book.
I dropped the Black Swan after only a few pages as Taleb managed, within those few pages, to simultaneously mock someone for using a personal anecdote to justify a position only to repeatedly use personal anecdotes over the next few pages to justify his position.
"Brittle" does not necessarily mean "bad". If that one marketing channel is really all you have, and perhaps the entire reason your company exists; then it might be fineto just admit that, make the most of it without overextending yourself or making lots of commitments, and then prepare for starting the next business.
Some permutation of "pair programming" helps with "mission critical" software execution risks. We call this "the bus test" in DC no-nonsense engineering. Project execution must survive anyone getting "hit by a bus."
It does not have to be pair-teams. It can be 3-teams that review each other's code. Crafting shared libraries also helps focus multiple developers around code, docs, calling semantics and help surface implicit performance expectations or penalties early.
I have been the hero developer cranking 10K lines of C per month. But divide and conquer where software development MATTERS isn't very wise. Methods make sense when dozens of jobs, thousands of lives or millions of dollars are at stake every week and month of a project.
We used to call virtuous redundancy "belt and suspenders."
In a certain sense, having a 9 to 5 is an even more extreme version of 'one thing' in that you have one big client that accounts for 100% of your revenue. Of course, there are also efficiencies in having one 'client' that are beneficial to both employee and employer.
On the small scales (i.e., you or me) it can make sense to have your second thing be a large pot of savings, which is a relatively inexpensive way of providing limited redundancy.
In the same way that a data center has full redundancy for failure of an individual machine, but limited redundancy for failure of the city's power grid, all we _really_ need is enough redundancy to give us time to solve the problem.
Large organisations move too slowly for this to be an option, so they need the much more expensive full redundancy.
It's rare in nature when the optimal placement of a 'dial' is at an endpoint (0 or 100). In everyday life, when I see people making such a choice, it's usually because they focused on a single attribute at the expense of all other considerations (e.g. "cheapest" or "best battery life"). While there's value in doing less work, ignoring the other criteria means the choice is not the best.
After almost two decades of professional experience (including a few years of consulting), I'm becoming more convinced that employees often make a similar 'lazy' decision when they choose full time employment. That decision is only optimal for under performers (and only for as long as they can hold on to the gig).
Selling labor in units of time offers no possibility for economic profit. The value of any productivity gains are realized by the employer. This is not the case for a business (or a consultant), where productivity gains yield profit.
EDIT: Consultants sell time too, but are free to adjust their rates as they see fit. Employees have significant barriers to changing their 'price'; infact it often involves a job change. Another challenge for the employee is demonstrating their value created due to the extremely complex structuring of most companies (this makes it hard to objectively justify a raise).
This can be generalized to thinking about what happens when X goes away in your organization or team. If Bill doesn't come in tomorrow what will happen? Planning for those cases and covering your bases for a a more effective organization is what makes them last instead of falling apart at the first rip off the seam.
The author has come up with the theory of diversification. The problem here is that, finding one thing is already hard. You often can’t magically find another thing that’s equally effective and so you burn a lot of money and energy pursuing other strategies. Take a look at what Charlie Munger says about diversification. Forget it, make big huge bets when there is a large positive assymetric risk reward.
That's an interesting analogy but the big difference is that with traditional edit distance every edit is exactly the same -- one character changed to another character. In the business case there's no equivalent "unit" size for changes.
Very true. Edit distance is well formalized, wheres defining a mutation on a business operation is not (though it would be super interesting to think about). So yeah, totally a fuzzy analogy.
What does "brittle" mean that "risky" doesn't? I don't think we need a new word here.
Also, the listed risks are cherry picked to fit the "one thing" pattern. Risks that don't fit that pattern include competition risk, technology risk, and financing risk.
I think of brittleness as a subset of risk. Risk has a very large surface area, and has to be mitigated by a correspondingly large number of techniques. Brittleness seems like a useful subset to address in the context of specific techniques that can avoid it. 'One thing' seems like a useful heuristic in this context.
In my experience "brittleness" has been used a long time to describe computing. They are typically far more efficient than manual methods but also starkly brittle.
The difference between brittle and risk is that brittle connotes unknown/unexpected failures, whereas risk covers the whole gamut but typically known risks.
Fragile organizations and systems break from randomness. Robust ones are able to resist and stay strong. But Antifragile ones actually benefit from randomness and volatility.
I highly recommend the book -- it's highly relevant to startup companies who, rather than shooting for the moon, should protect their downside (at first) and place small bets on numerous 10x activities.