I read an essay called “Software Talmud #00: Anatomy of Automation” about software development that felt like a hall of mirrors.
It is a pile of quotations that argue against making piles. I found that amusing, to say the least.
It invokes the Talmud, which it defines as the wonderful codification of thousands of teachers, a pile of beauty, telling you that the best component of all is the one that was never added. Ok, I’m laughing. This is a man arguing for subtraction who keeps adding. A hammer that says fewer hammers. Alan Watts the “philosophical entertainer” used to give a “stop talking” lecture to audiences who were expected to listen to only him talking, and talking, and talking, rather than be in dialogue.
In the middle is an example of a thing built and then discarded after finding something else already solved the problem. Yet the rest of the piece expands, rather than reduces, the reader’s commitment to it. Haha, good one.
All that being said, it lands on a very somber note. It wraps up like a memorial from Efron Amber Licht to his grandfather and great-uncle, who changed their names so decades later in 1960s America they could write about engineering. These sage engineers indeed solved problems that we seem to forget, in the same way our grandparents paid for comforts they wanted us to enjoy today.
The whole piece is arguing that you can’t trust subtraction until you’ve done the expensive work of deciding what the system is even for, and his own family is the case his writing never resolves, a cut he calls a good one and grieves anyway.
My suspicion is that the discrimination his grandparents suffered, and the freedom the simple name change bought, sit at the kind of distance he can’t see from where he sits. It’s like learning how to dye your hair blond in Germany in 2026 if you expect kids to treat you as a human instead of a threat.
Here’s the problem: He mourns from behind the very protection that he’s mourning. He inherited the freedom and reaches back for the nostalgia of a thing that, kept, would have cost him the freedom he now enjoys.
His elders chose to survive so that he could mourn what they gave up for him. They did exactly what their book preaches. First decide what the system is for, like staying alive and employable in a country with restricted covenants and quotas, then start the cuts to fit in. They got the purpose right. He says he cannot fault the decision, and grieves it anyway. The grief is the tell. It comes from a distance the decision itself created.
And I get a fixation with reducing work before you speed it up with tools. That’s a simple formula, perhaps too simple, missing context. Everyone learns measure twice, cut once, right? Ooops, that’s actually a lesson to increase the work before you automate. Well, surely we’ve all heard waste not, want not. Damnit, that’s thrift inside a job already accepted, not a reason to skip the job. Fine, look before you leap. Nope, that’s caution about how you act, not whether to do the act. A stitch in time saves nine? That’s just measure twice in a different coat, with more effort now to prevent the work you know is coming. Hmmm. I’m still not finding myself on the same page as this essay. Every proverb I reach for slows down and works harder, sharpens the axe and steadies the hand. Not one of them is a reduction, asks whether the work was ever a good idea in the first place. Well, maybe he means what Einstein said, as simple as possible, but not simpler.
Thank you so much for taking the time to read my work. I think you have largely misunderstood it. Maybe I didn’t do a good job explaining it.
“This is a man speaking who says nobody should be speaking.”
I have no idea how you reached this conclusion.
—
“In the middle is an example of a thing built and then discarded after finding something else already solved the problem. Yet the rest of the piece expands, rather than reduces, the reader’s commitment to it.”
That is, in fact, the point. This is me following my own advice. There are no easy solutions.
As H.L. Menken said: “For every complex problem there is an answer that is clear, simple and wrong.” There are no shortcuts to the right-sized solution to an answer; there are at best heuristics. You have to do the work yourself to try and evaluate them.
The development of software is about the constant evaluation of tradeoffs, uncertain solutions to changing and possibly unknown goals. You have to actually do the work, even if the end result leaves you where you started. Choosing to do nothing is not the same thing as not making a choice. There’s a tendency in all creative work to try and keep everything you’ve made, which leads to unavoidable bloat. A good engineer has the discipline and intellectual integrity to throw out unnecessary components even if those components are good in and of themselves and they worked extremely hard on them.
—-
“My suspicion is that the discrimination his grandparents suffered, and the freedom the simple name change bought, sit at the kind of distance he can’t see from where he sits… the grandson who second-guesses that decision raises the question whether he ever faced the kind of system they did.”
To be blunt, I think this is a failure of reading comprehension.
1. I brought it up BECAUSE I see that tension. Why else would I have brought it up?
2. I don’t second-guess that decision. I say, quote, “I cannot fault Grandpa’s logic or his decision”. But I do mourn the consequences of so many similar practical decisions. They may have been the right ones, but they have a cost.
—
“Maybe he means what Einstein said, as simple as possible, but not simpler.”
Absolutely. (I considered the quote but I figured it was overused.)
@Efron, thanks for coming here to say it directly. That’s great and I respect it.
You’re right with “second-guesses”, my phrase, not yours, and the post already quoted you calling the cut a good one. I’ve fixed that paragraph. Grief was the right word all along, and that was yours.
The rest is our differences in reading, rather than incomprehension. You argue for subtraction. I found the essay hard to subtract from. That could be my failure as a reader or your essay doing the very thing it warns against. Probably both. The example you built and threw away was the best part, and I said so, even if it read as a joke.
What stays with me is your grandfather. He decided what the system was for and cut to fit. You honored that decision and mourned it in the same breath. That’s what I wanted most to write about.
On Einstein, we agree. Overused, still right. Thanks for the essay.