Embedded Systems Are Not Just C, Linux and ARM: When Will HR Notice Senior Techn

[okmn]

How can we wake up employers and HR teams to a very simple reality in embedded systems?

Embedded systems are not built only with C/C++, Linux, FreeRTOS and single-board computers, and not only with ARM/STM32.

Microchip still has a very strong 8/16/32-bit ecosystem: PIC18F, dsPIC33CK/CH and many others – with peripherals such as NCO, CLC, DSM, SMT, CWG, HRPWM, APPLL, etc. These peripherals offer capabilities that many engineers today associate only with FPGAs; in fact, a lot of low-end FPGAs (especially those without analog features) do not even provide comparable functionality.

Over the last month, I have reviewed more than 300 job postings. Almost all of them repeat the same pattern:
"C/C++, Linux, ARM/STM32, RTOS, and must have Engineering degree."

But here is the uncomfortable truth:
Not everyone who holds an engineering diploma can actually deliver high-end embedded hardware/firmware design in real projects.
At the same time, many highly experienced technicians and technical specialists with 30+ years of hands-on work (like myself) do design, build and ship complete embedded products – from schematic and PCB to firmware and test.

So my question is:
When will the industry – especially company owners and HR – seriously recognize that senior technicians with proven track records in embedded design can be just as valuable as, or sometimes even more effective than, a "fresh" engineer on paper?

Is this shift in mindset possible, or will we keep filtering out real experience just because the CV does not say "BSc in Engineering"?

[RGV250]

Hi,
I have found the more degree's you get the less common sense you have.
When I was at work the people with hands on experience were way better than the people with all the certificates.

For example, I believe Les is self taught but could teach the programmers at Microchip a thing or two.

Bob

[trastikata]

When that none-sense called a "Degree" or "Diploma" will stop 

Originally an "engineer" was someone who built and operated "engines" or devised clever contraptions, evolving from military contexts to civil and other fields, reflecting a focus on practical application and problem-solving, not someone who graduated from overly-paid university.

Bottom line - Degree's and Diploma's don't get the job finished, but common sense and constant self improvement through self education.

Looking for the "big money" - start stealing from and exploiting people.
Looking for cause job with no responsibility - go to administration, preferably government
Looking for miserable, underpaid and unappreciated but fulfilling and rewarding job - go to the real sector and do something real, tangible, with your own hands and brains.

I think nowadays the mindset is, especially in Europe and to much lesser extend in America, that Diploma's and Degree's get the job done, and I can see the results from this thinking every day.

[Stephen Moss]

So my question is:
When will the industry – especially company owners and HR – seriously recognize that senior technicians with proven track records in embedded design can be just as valuable as, or sometimes even more effective than, a "fresh" engineer on paper?

Is this shift in mindset possible, or will we keep filtering out real experience just because the CV does not say "BSc in Engineering"?

Unfortunately, I don't think things will change in that regard, or if it does it will be a long hard slog to change it like Sisyphus pushing that damn rock.

I have not bothered even looking at other jobs for about 20 years because they usually ask for a degree and so I know my application will go in the bin as I don't have one, particularly as many such jobs are being advertised by employment agencies who will filter out my application because of that, whereas if it went to someone technical they may overlook the lack of degree and appreciate lots of real world experience.

Personally, I would rather have someone without a degree but a lot of practical real world experience than someone with degree and no experience as I don't think a degree worth the paper it is written on these days. When I started working at the university back in 1990 third year undergrads really knew their stuff, now I would not trust them to be able to make an LED light up. For example I had a pair last year who effectively had their BA in the bag as it was an end of year project and I had to help them get a LASER diode and Photodiode working when all that was need was Ohms law to calculate the correct resistor.
To quote one of my favourite lines from a movie 'you can teach monkeys to do better than that'.

[Abdullah]

I also used to think that a degree is everything. I don't have any degree either, but now I understand that if a person learns on their own and works hard, the knowledge they gain doesn't require any degree. In fact, such a person is usually better at practical work than a degree-holder, because their hands-on experience is greater than theoretical book knowledge.

[joesaliba]

I do not have a degree either, but was eager to learn. Fortunately, once I moved to a company working on lifts.

the manager asked me what degree / certificates I have. I replied: - I have none, but willing to get some certificates if you need them.

He smiled and replied, "No need, you can begin tomorrow as I have seen a lot of people with certificates and they do not even know how to read a circuit diagram, so is your turn  now to prove I am right".

That job opened some good opportunities to me.

[charliecoutas]

No degree either but a strong curiosity to find out what makes things work. You can tell if somebody has that quality. I was lucky enough to spend some time with our Les a few months ago and boy, he also has a strong passion for finding out about the world.

I think that old thing about being able to take a horse to water but not make it drink is so true.

Charlie

[CPR]

How can we wake up employers and HR teams to a very simple reality in embedded systems?

Embedded systems are not built only with C/C++, Linux, FreeRTOS and single-board computers, and not only with ARM/STM32

You've answered your own question. C/C++ is the universal and industry accepted standard. And. That's just how it is.

I have a degree in Electronics. Quite useless - without the real work experience. So I absolutely agree with you all.

A "Basic" compiler such as Positron, will <unfortunately> always be viewed as niche, as it's "Basic" - something seen as just a hobbyists tool. Moreover it's the creation of just one man. Les. Which isn't encouraging. Despite how wonderful and efficient it is. So Positron I think, will alas, be forever stuck in that hobbyists niche and never become mainstream despite how excellent it really is. And. That's a real shame. Convincing Industry to adopt it? is going to be next to impossible. I'll continue to support Les. I'm retired. I've grown to love the compiler and use it for many of my PIC projects. But for Industry? No.

[shantanu@india]

To me embedded electronics is 90% hardware & 10% coding. What is the use of writing code if you cannot independently design and prototype a PCB after delving deeply into the architecture of the microcontroller/microprocessor? Nowadays fresh engineering graduates(electronics/instrumentation/mechatronics/electrical)know how to code in C in the Arduino framework.But most of them cannot read the colour code of a resistor , nor can they use a soldering iron!!When I interview youngsters the first thing I do is to throw a random resistance & ask its value. To me coding is secondary.

It is true that C/BASIC/LINUX etc. etc. doesn't really matter when the finished product is good to go!But unfortunately majority of the job interviews focus on C-coding skills. That is because C is the de-facto standard in embedded computing. This is not going to change in the near future.

[Stephen Moss]

You've answered your own question. C/C++ is the universal and industry accepted standard. And. That's just how it is.

I may be wrong but I think that was because early BASIC languages were interpreted, not compliable which meant the resulting code could only be run in computer that has the applicable software (i.e., Microsoft BASIC) installed and interpreted code is slower to execute.

The reason that is unlikely to change is that if you employ 10 programmers and they are all using a version of C, it is easy for one programmer to take over from another or for several to work together coding different parts of a large project. That does not works as well if some are coding in C, others in BASIC, others in Assembler and the reminder in Python.
Oddly the undergraduate computing courses where I work were in C for that industry compatible reason, so the change to Python a few years ago seems a odd choice.

[top204]

Many thanks for your kind words people. :-) I am self taught on computers and digital, because when I took my city & guilds, it was only for repairing TVs and things, so it was analogue only.

In its sensible format, the C language can be really good. However, on certain microcontrollers, it is not very good because it is a stack and pointer based language, and a lot of microcontrollers have not got that mechanism, so the assembler code is bloated. It is also too flexible in its syntax, and some of the combinations are dreadful. They look "really cool", as the nerds would say, because it all fits in a single line. However, the assembler code produced for that dreadfully convoluted line of C code, is so bad, it is sometimes hard to believe it was generated by a C compiler. :-)

I totally agree with you Stephen... The Python language has taken hold, because it is cross platform. However, it "is" an interpreted language, just like BASIC was in the 1980s, and is very, very inefficient, and extremely bloated. But that does not seem to matter these days. But you get these people saying "it is not interpreted, it creates byte code" :-) But that is also what some of the interpreted BASIC's did in the 1980s, to make them a little faster. But they were still slow and cumbersome.

The same was said of interpreted BASIC in the 1980s as well, in that it was a standard language, and is fast enough for what is required. But that moved on after a few years, and it was no longer fast enough. So the exact same thing is happening now, and computers are actually taking backward steps, in that they are no longer becoming more efficient, just larger so they can cater for the bloated, interpreted, cross platform languages. :-)

Just look at Windows, and that dreadful MPLABX, among the many dreadfully bloated and sluggish programs now.

[See_Mos]

Some of the above comments remind me of an 'engineer' we employed.  He had a BSC degree in electronics but when I handed him a 1N4007 and asked him to identify which end was which he could not tell me.  He could talk electronics but had no practical experience so he didn't last long.

Like Les I am self taught with the aid of books and electronic magazines and worked my way up from repairing  radios and record players through colour TV and video recorders before moving into industrial electronics and getting involved in automation and repairing equipment for the North Sea oil rigs.

[Abdullah]

After listening to everything you've said, I've become convinced that language, culture, or even climate changes don't really make a difference. The problems I've seen among ordinary Pakistani people are the same problems you're describing, and they are 100% accurate. Therefore, regardless of the language, the issues are essentially the same everywhere.

[RGV250]

I wonder if the person writing the code for the speed camera farce had a degree rather than common sense

For anyone not in the UK, when the variable speed limit sign went down there was a delay so the camera did not catch you incorrectly as it changed. Trouble is they also left the delay when the speed limit sign went up so the camera still thought the lower limit was still in place.

Bob

[John Lawton]

Possibly the person writing the code was not a motorist?

John