A response to @abraxas3d@defcon.social.
abraxas3d is a friend of mine for several years now. She founded and runs Open Research Institute which as far as I can tell is the only professionally-minded amateur group doing development today.
From her post.
The mission at ORI @OpenResearchIns is educational, personal, and professional advancement through open source digital radio work. The vast majority of this work is done on the amateur radio bands, simply because it’s the best choice for experimental work. We use FPGAs, processors, ADCs, DACs, and a pile of other circuits to get work done.
When it comes to basic electronics competence, we are seeing (in general) a recession or decline. People that used to go into RF hardware in college, for instance, are much more likely to choose computer science. This is primarily because of the perception that it results in more income. There are fewer people, and fewer young people, that have any idea on how their bits are handled electrically. Some are totally unaware of registers, except as a target for program values. In the best possible case, this is totally ok. You shouldn’t need to know details of the transistor model your processor has in order to write a while(1) loop. Specialization means you have to start taking others at their word when you are given the “go flight” signal.
The decline in RF/hardware engineering is mirrored by the influx into computer science, coding, boot camping, etc.
Is this a problem? Well, the market doesn’t currently think so. Markets currently reward pure computer programming solutions at very high rates. But is it a problem? Yeah, it actually might be.
We now have a lot of people with coding skills that don’t understand the circuits that their code runs on, and/or cannot do basic electronics. We also have a shortage of people that can produce designs that support a very demanding computational market.
And that is a problem. I think we will continue to have a lopsided and self-limited approach to computing if we put all our energy and attention in the computer instruction category, and none in the instruction set and hardware category.
What will the eventual re-balancing look like?
I’m replying here because I couldn’t get it to fit in the micro-blog format on Mastodon.
I’m one of those people that went for software. I even had an interest in hardware from the start.
I didn’t switch because of pay - my mind was made up long before I realized it and committed.
Ultimately, I got into software because it was a much faster and therefore more fun creation loop. Making mistakes didn’t result in permanent consequences.
I didn’t have to wait for parts from China - the ones I could afford
at 16, anyway - or spend hours debugging cheap connectors just to get
to hello world a blinking LED, a pattern on a display, or a beep on
a speaker.
In software, the barrier to entry was lower than every other field, the education was better in all regards, the cost of additional learning on my own was free.
It’s only now that I have been slowly dipping my toes more into hardware.
I agree this is a problem. I don’t know how to fix it.
Familiarity
I got here through software - because only on a software salary can you start to dabble with hardware.
Of course, if I want to build something, the first tool I tend to reach for is the one I am most familiar with.
So yeah, I’ll blink an LED with a uc pin. It’s one microcontroller, Michael, what can it cost, ten dollars?
Openness
Open source is a huge component in what makes software good.
Open hardware is harder to get into than closed hardware, even just in terms of availability. That’s assuming there’s even an open competitor - in many cases, there simply is no equivalent.
If the hardware is closed, you can’t do as much with it, you can’t really learn from it in quite the same way you could otherwise. You certainly can’t remix it.
In comparison, open software is much easier to get into than closed software.
For one thing, it’s free. The marginal cost to copy software is zero dollars.
For another, it’s self describing. If you want to know how something works, everything you need is in the code.
I’m not referring to comments. I mean that if you look at a PCB, even PCB design files, you’re looking at ‘compiled’ output unless the creator specifically put notes all over it.
Hardware is vastly more opaque because it is closer to physics.
There is a lot more required knowledge just to start, because all of your abstractions leak so much more due to this proximity to the real world.
That’s only one problem with learning this stuff.
Education
Education around electronics doesn’t get you to something practically useful anywhere near as quickly as software, but worse, hardware education has had no attention paid it in decades.
Formal hardware education, at least undergrad and below, is plagued by cost-cutting and the lab equipment is never fully functional.
My high school electronics lab (yes, I was one of the lucky few!) had I didn’t even know breadboards could fail like that until the logic design course at college. Many classmates never got their hardware working, and at least some of those I know for a fact were not their fault.
On top of this, even stalwarts like UMass Lowell that taught C are feeling the pressure to move to yet higher level languages, partially from the pressure that learning how computers work (or at least how they pretend to work) is too hard, partly from very reasonable demands that students be taught the things that will make their multi hundred thousand dollar investment give a return.
Even learning ‘pointers’ is considered complicated among most CS101 classes.
Nevermind. I can’t start on this, I’ll rant.
Okay, no, I have to. Education is completely broken, and money was never
the problem past a certain point which we passed a long time ago.
The problem is severe incentive misalignment which leads to absolutely fucking no one caring enough to even try to fix it anymo-
Cost
Hardware is a game of the comparatively rich well off. It is only now, with
a software engineers salary, that I can really start to dabble.
And of course, this point in most peoples careers is where they need to save money more than they need to learn new disciplines, whether to start a family, deal with health problems, buy a house (glhf), …
Cost is not just dollars, but time. Same thing, for most of us.
Related, I once read racing quadcopters described as a “rich man’s hobby”. I thought that was a nuts claim at the time, but now I see it.
I am not rich - but I do make more than most families in this country - and I still consider quadcopters expensive, especially if you want to ‘get good’.
The problem with cost and a high barrier to entry with long incubation times for deep understanding like this field has, is you rapidly end up with strong segregative effects. In the same way that the only way to get into motorsports as a kid now is if it runs in the family, understanding how electronics or indeed even software works will - or, as I believe, already does - depend strongly on your family history.
This will ultimately lea-
Risk
Our society has become extremely intolerant of risk. One of those things currently seen as risky - I don’t pretend to understand the thought processes involved - is that of letting children solder. Maybe because it can get hot, I don’t know.
That doesn’t even include LiPo fires, the level of concern I’m referring to is disproportionate to the costs involved whereas you are almost certainly not as concerned about lithium battery fires as you should be.
What I mean is that there are parental and institutional fears around giving young people access to things that can get slightly warm or are slightly sharp, and a societal unwillingness to have separate rules or pathways to eligibility.
I’ve known plenty of adults that could burn down a house with a wet stick, or would be perfectly safe with lighting their cigarette with a flame from a gas can. The problem is that observing the difference, and especially licensing the difference, is another way to spell liability, and no one wants to be liable.
So everyone has the same rules, and the rules are the strictest possible.
That’s the driving factor here. Fear.
This was not terribly relevant to my own path, but I have personally seen it be a controlling factor among parents, and I’m sure you’re aware of it in your own life. It’s impossible to escape.
It’s now a part of the culture. Young people are more risk averse than ever, and who can blame them? We are surrounded by cameras and the associated permanent social and institutional consequences.
This problem is far more widespread and damaging than I can cover here. Just considering the way dat-
Creativity
It sums to worse than its parts.
If the end result is what drives you forward, then what’s the fastest, cheapest, simplest way to making the thing you want?
How long does it take learning a new discipline to get to a minimally useful thing which gives you a sense of accomplishment?
How much of a pain in the ass are the official ways to learn it? How good are the resources to learn it on your own?
How long does it take to get to where you can start to have taste and style, judgement of your own?
How much money does it cost, and how much money can you afford to burn?
Any one of these, when compared to software, seems strong enough to be explanatory. Worse I think it’s all of them combined.
I don’t know how you fix that and rebuild electronics expertise without severe intervention at all layers of the stack.
Maybe we could ban all electronic use before 21 unless, at minimum, assemble it from a kit and program it yourself.
I don’t think that would work, because I don’t think it’s extreme enough ( and I find the idea morally repugnant besides).
Actually I’ve long had some thoughts about this in view of the broader internet and how it’s changed over time. Eternal Sep-
Fun
Learning is fun.
Making is fun.
Learning to make is fun, and making to learn is the most fun.
I think that’s key to the solution, at least for the people that are interested.
This is a golden age. Maybe if you could find a makerspace that has people that like to do more than like to talk about doi-
Learning
So much of learning is doing, and making mistakes. When mistakes cost you time and money, and you have entire generations that are severely fearful of mistakes, then the doing suffers, and you start to run out of doers.
They’re not having fun learning. Ask them. Learning is not a fun thing for most people, because it’s imposed from outside. Most people have never learned by their own will.
The birth rate is dropping like a rock, the economy is worse than ever and we pretend to ourselves that it isn’t that bad because as soon as we stop pretending it will get worse, and the beatings will continue. What morale?
There is no fix. There is no widespread single interv-
The solution
Fuck it. I don’t know.
I think the only thing you can do is help the people around you. To show how it’s fun, to make it easier for others to follow in your footsteps, to share your hardware and knowledge as best you can, to calmly and professionally demonstrate and celebrate excellence in all its forms.
The problem is these valued people - i.e. people with those values, and
people that we value highly - are few and far between, and it takes a
long time to make them. Moreover, we don’t reward them.
Actually, we tend to punish them.
Oops.
The reward for doing work is more work. So if you work against assholes, trying to do the right thing, you become swamped with more of the same type of work. You’ll have a reputation as a troublemaker, hard to get along with - admittedly mostly among assholes that you don’t want to work with and their friends (and their friends …).
Quick: how much does it matter that the reputation is undeserved, and the people that actually work with you know better, when it costs you time, energy, and funding to fight in the first place?
About the only thing you can really do in the long term is try to carve out your own chunk of shielded sanity, isolated from the world; to work with people you like on problems that are fun to solve and happen to pay the bills, even if only enough.
Most people never get there.
And it’s easier to find in software.
Good hunting.
Cool stuff
https://www.openresearch.institute/ - Aforementioned isle of sanity, albeit all-volunteer.
https://eater.net/ - dude that talks about electronics at a much lower level than normal, and explains every bit of it. Sells kits. Solid.
https://nandgame.com/ - in case you missed a class in logic design.
https://artofelectronics.net/ - one of the better books to bootstrap with.