Friday Ephemera

Back in 2006, “Zed” Shaw (a famous Ruby-on-Rails programmer) penned an essay called “Rails is a Ghetto” in which he said that if all you know is programming, you’ll be stuck in low-wage trivial coding jobs your whole life. If you know something else and a little bit of programming, you can write your own ticket for the rest of your life because there are tons of fields out there that could desperately benefit from automation and software engineering.
Do you remember Yourdon’s Decline And Fall Of The American Programmer, 1992 followed by his Rise & Resurrection of the American Programmer , 1996? Not to mention his Y2K freak out. I was a fan of his methodology, in the context of the time, but rolled my eyes at the “Decline and Fall” thing after a little rumination about it. Thank God I hadn’t seen that a few years earlier. By the time 1996 rolled around and there was much competition in the software Guru…”evangelists” even…game I was on my way to mentally checking out on what the “experts” had to say. By the time XML was being hyped, I had essentially decided, even though I was on the elite programming team at my company, that keeping up with “the industry” was a fool’s errand. I did my job, did it to best of my ability, but after Dr. Dobbs went web-only I decided if any new tech was all that important, I would hear it from someone else. I still did some advanced stuff. At one of my last efforts I was the one assigned to get their Kafka operation working after a couple guys failed, did some limited Spark/Flink stuff.
That said, “Zed” was himself full of s**t. Unfortunately. I mean, that’s probably how things could/should/ought to be but to be a decent programmer it’s not so important to understand *specific* things other than programming anymore than it is to understand the latest-and-greatest thing. Though I agree with you on PV=nRT or whatever it was. That’s basic high school stuff that people should know*. Far more important is the ability to pick up something you don’t know, have no fear of playing around with it, and in many important respects, not being afraid to ask “dumb” questions. And it’s this latter bit that can get you into trouble for no good reason. I have on occasion asked a dumb…”dumb” question to try to ferret out if perhaps a fundamental assumption may be wrong. Now there’s a jerk way to do this in which you project how f’n smart you are and a proper, yet vulnerable way to do this such that it would be in your best interest to know/understand the personalities of the people on your team first. I found a lot of this got sooooo much easier once my pile of F*** You Money had reached sustainable levels.
I say all this not to bore the non-tech programmers to tears (as I fear we may be doing so) but because the fundamental principles underlying this apply to much more of life in general. Supposedly a liberal, or even “liberal” educations was intended to teach you how to learn by learning many different things. From being around many different people and understanding how much you think you know that you don’t know. Somehow we got to a point fundamentally opposite that, where we tolerate the intellectual preening of those who need to be seen rolling their eyes and stating “Obviously….” when someone asks a question, especially one outside the Narrative of whatever domain. And I blame our colleges and universities, especially those with huge 200-300 student lecture classes where questions are only tolerated in “office hours” sessions. Especially those kinds of office hours where the students are lined up out the door.
*”Stuff people should know”…As, thanks to internet and social media and such, I am in somewhat frequent contact with many of the “smart” kids I knew in high school, and quite a few of the “not-smart” ones as well, I am often quite amused at the things the “smart” kids who got better grades than I in certain subjects seem to know less about those subjects all these years later than one would expect. I’m talking the basic stuff, like how electricity generally works, non-political stuff from basic civics classes, etc. not detail stuff about how to do LaPlace Transforms.
Well, that’s longer and even more boring than I had hoped as I find I get in trouble beyond two paragraphs…or less…

Darleen: That Morticia is the perfect comment.

People do tend to forget the basic high school stuff that they don’t actually use. But someone has to either know that stuff or know to seek out an expert who does know. At the larger companies I worked for in the 70’s/80’s, analysts would work with engineers to develop the specifications and design, and then hand that off to a team of programmers.
Oh, I get that. Guilty as charged as well. The details, yes. But the general ideas? Surely one who took AP history should have some idea/understanding of the Federalist Papers. Not the details as to which specific ones or even a favorite, but that they were there and relevant to the SCOTUS rulings. And yes, in general that is how requirements gathering is done, systems engineers and/or business analysts design the data, data flow, the requirements, etc. I worked on cell phone billing systems for many years. The details of how those things are constructed and/or how the switches work, even in context, I certainly cannot discuss off the top of my head. But the general idea, and especially when being presented by things that make no logical sense, a person should be able to look at/understand these things without having to know the details. The details are what the documentation is for. Though good luck with most of that.
I have written code for billing systems, land line telephone inventory for foreign countries, fingerprint systems, military hardware logistics, rocket launchy thingies, and many other disparate business and military and security systems that in their fundamentals I could not discuss off the top of my head nor did I have any deep, nor sometimes even shallow knowledge about before taking the job. Even the military stuff which I kinda knew on a superficial level as a kid was long gone from detail memory but the general understanding that there are ‘splody things and life-support things and landing-things etc. that obey the general laws of physics. But what I can do in all of that, and this is what I was getting at regarding the “other things a programmer should know” is the ways in which there is fundamental logic in how these systems could/would/should work. It’s not so much important to know everything, in fact it’s bloody impossible, but to understand the philosophical/logical/physical underpinnings of them and how to make them work that is important. Talking about technology or writing a program while people look over your shoulder in an interview, not so much. Talking about how to solve a problem, yes. This gets disparaged, and somewhat rightly so, but I find the way-outside-the-box questions like “how would you stack quarters as tall as the Empire State Building” to be more valid questions than the technical ones. Though more specific “Why are manhole covers round?” to be stupid. AIUI, the latter “correct” accepted answer was itself wrong. Wonder how many companies eliminated people who knew the right answer.

That’s basic high school stuff that people should know
Ah, but see, it isn’t.
The Ideal Gas Law only holds for temperatures and pressures near STP. At extreme temperatures and pressures – like, say, inside a large pressurized industrial gas vessel – the gas molecules are packed close enough to each other that the stereochemistry of the gas molecules starts to matter. Exactly how much you have to modify the Ideal Gas Constant to account for this is a function of the specific gases, their proportion and the pressure and the temperature inside the vessel. Since there’s no convenient equation for this, you have to know how to interpolate properly between data points that have been determined by exhaustive experiment.
Which is why “Zed” Shaw was 100% correct. The biggest problem in the software industry is autodidacts mired deep in the Dunning-Kruger effect who think that a superficial knowledge of their problem domain acquired in a weekend constitutes mastery.

wypipo look at books in a “black space” with predictable results

charming, birdhouse-like structures filled with books that invite neighbors and passers-by to take a book, or donate a book, or both

I like passing on the books I’ve read to friends who I think might enjoy them, and who might have insights that I’ve missed. But it’s specific books to specific friends. I wouldn’t leave books out for strangers: it would be presuming that my box of discards is a public amenity, not only for whatever reading material can be recycled from it, but as a catalyst of conviviality.
But the birdhouse image really gets to the appeal of it, how people imagine it’s going to work. It’s charming, it’s twee, it’s well intentioned, it sets a good example for the children, it adds to the neighborhood scenery by itself and by the activity it generates, it’s an indicator of a good neighborhood if it can stay out without being vandalized, it’s Stuff White People Like.

A library is not so much a marker of wealth and whiteness as it is an affirmation of community and cozy, small-town camaraderie that Inglewood, a mostly Black and Latino city in southwestern Los Angeles County, has plenty of.

How many lies, or hopeful claims, are in that sentence?
Birdhouse libraries come out of white camaraderie and white concepts of coziness and community. We know this because it happened. It’s as white as a white bread and mayonnaise lunch at the skating outing of a Pat Boone fan club.
Are community, coziness, and camaraderie universal concepts that by universal laws lead to book boxes springing up on the curbside? Surely blacks have different concepts of what’s cozy and communal? Block parties with loud music, for example?
She broadcasts the Pat Boone music and pretends to be offended that whites turn up, pretends that Pat Boone is a universal that every culture can adopt as its own.

The response to the library was slow at first — it was the first in the area, and some people mistook it for a birdhouse, or a mailbox. But I was pleased to soon see people stopping by to browse and take home books. Then one morning, glancing out my front window, I saw a young white couple stopped at the library. Instantly, I was flooded with emotions — astonishment, and then resentment, and then astonishment at my resentment.

This would be a great college application essay. The personal initiative taken, the slow but promising takeup, the racial problematization, the “as a Black woman” emotional response, the contextualization of the response within systems of redlining and Black Hair.
I suspect that the initiative didn’t quite snowball, if I that metaphor is acceptable, among her black neighbors, that it’s still just a box of mostly unwanted books and not a self-sustaining communal chain reaction.
She’s looks like she might have a white parent or grandparent, she was married into a Jewish family, she moves in SWPL circles, she has SWPL tastes in birdhouse libraries. I read the article before seeing her picture, and I assumed that she was a with-it white who was living in a black neighborhood and complaining about less with-it whites.
Is she angry because of her own internal conflict between her SWPL tastes and her ancestral black loyalties? Is she angry because her disappointment at the lukewarm takeup among her black neighbors has to be redirected to the SWPL passers by?

but if it were necessary for every single coder to have full knowledge of every single domain that they need to work in, nothing would get done.
Isn’t that the role of a good business analyst? I’ve been programme manager and ultimate business owner of many tech projects and I always got the best results when I gave significant time to both analysts and coders usually together. The investment of my time in the end always saved in development time. I say that as someone with no coding experience. I brought the business expertise and a tight definition of business requirements. I always found that the more the coders knew about what the end users were going to be doing and the limitations they faced in the way they had to do it, the more coders were able to get to a deliverable that worked.

Meanwhile, in Inglewood, California, wypipo look at books in a “black space” with predictable results.
Note this passage from that guest editorial:
“As more Black families moved to the neighborhood, white people moved out in droves. The ground shifted under Uncle Paul’s feet. That white flight forged the chiefly Black and brown South Central of popular imagination and created similar demographics in other city neighborhoods across the country, including Inglewood.”
The author neglects mention that South Central became a hell-hole of crime and disorder and every pathology of lower-class dysfunction. Those white people had good reason to move away. Or does the author somehow think that white people were to blame? That, by moving out, they magically transformed honest black citizens into stupid and vicious criminals?
I have known many people who had to move away from neighborhoods they had assumed they would live in all their lives, due to crime and other dysfunctions caused by newly arrived blacks. This put a dent in their retirement savings, made for longer commutes, and disrupted their networks of friends and neighbors.

if it were necessary for every single coder to have full knowledge of every single domain that they need to work in, nothing would get done.
The vast majority of software serves no useful purpose and has been created by autodidacts seeking a get-rich-quick scheme. The discipline of software engineering is still very much in its infancy despite fifty-some-odd years of knowing full well what works and what doesn’t.
At a wild-assed guess, I’d say about 50% of development time is outright wasted due a lack of any kind of engineering discipline requiring expensive rewrites and bugfixing that could and should have been caught before release. That’s time that could be spent training the software engineers in their problem domains to ensure that the code does what it’s supposed to.