Ask HN: What is a commonly known tech topic that non-techies fail to understand?

Net neutrality...

How data retrieval and storage with DBs works.

How an API works

What tech stack the company uses and why (what benefits are you utilizing)

That technical people are not easily swapped like workers in some other industries.

That technical industries move faster than perhaps any other industry, and in the time you're taking to ponder a single decision your competitors may have built an entire product.

The TV trope of "I can do this in 4 hours" "You have 1" is not actually how things work.

Assume the topic is a difficult one. The manager / exec you are talking to can always interrupt and tell you to skip to the more technical.

If you prepare to explain it to a five year old, and you could already explain it well to someone with domain experience, you can probably explain it to someone that lies in between.

In Windows Development knowledge half life is 3 - 4 years In other words; every 3- 4 years half of my Windows knowledge becomes obsolete

Technical Debt

That the devil is always in the details. How well you can explain your problem to someone is not an indication of how easy it is to solve. The difference between "simple" and "easy".

Bitcoin

Bayesian inference. Someone looking like the suspect was at the murder scene. But there are millions of people looking like the suspect. The murderer was wearing a yellow jacket and the suspect has a yellow jacket in his wardrobe. But there are millions of people owning yellow jackets. If you know how Bayesian inference works, you know that the suspect is very likely the murderer.

A mixed bag here (not all technical but relevant nonetheless),

- No free lunch

You can't invent outcome out of thin air. Laws of physics apply.

I have been often asked to "achieve outcome that requires a lot of effort" but somehow ... somehow find a way to do it without any of the effort.

"Bullet proof our backup systems .... in 2 weeks"

The problem with this is there's always some subset of people who will claim they can do it.

(By diluting the meaning of words such as "bullet proof" low enough to fit the time-frame)

- HTML is not photoshop

So many people think a webpage is just a fancy poster.

- The ability of programmers to solve your problems is directly proportional to how well the business domain and the problem itself is specified.

Software is the outcome of a process/function. If you don't provide your share of the input to it it will not magically come about from the ether. The lack of that information will be obvious in the end product.

If you go to the doctor but are too lazy to talk about which part of your body hurts ... well don't expect miracles.

- Not all problems can be universally and indiscriminately broken down into smaller pieces

Also refer to the mythical man-month. I have seen a lot of people with what I call "lego-world mentality".

They think anything and everything can be broken down into smaller pieces and the output collected and stitched back together and they think that has no effect on anything (quality/timeline/integration).

It's like asking 4 chefs to each cook 1 egg to make one big omelette.

Even things that sometimes look like obvious separate systems/sections only look like that on the surface.

("You do the X part" .. "I do the Y part")

Again, It's not a lego world.

- A sequence of next-best moves does not always lead to the absolute best solution (Local minima/maxima, lazy algorithm to the travelling salesman problem)

Sometimes you have to zoom out. In practice this means progress doesn't look like a consistent straight line over time. Sometimes you pull back to be able to gain more speed and jump further.

Not understanding this results in execs viewing anything that is not strictly and overtly progress as a waste. Further discouraging a class of important work and shaking the carrot on the stick for people to pump out features mindlessly.

- Everything is much more flimsy than what you are lead to believe

Similar to the lego-world mentality. You see execs applying that (imaginary) mindset to certain things such as APIs and cross-system integrations.

They think it's like plumbing. You connect the Zendesk pipe over here to this other pipe and it just all flows nicely. Like a nice clean diagram in your head.

This perception is created because they only consume the surface-level marketing material but don't see the ugly crap under the hood.

As a result they under-estimate how fragile software is.

This happens in a lot of domains. For domains that people are not an expert in they tend to over-estimate how sophisticated/accurate the technology/processes are.