It would be interesting to take the same model but use known inputs for different eras, i.e. 1600s, 1700s, 1800s, and see when it would then predict collapse.
I'm definitely for more guided resources on making Linux apps from idea -> release, which I wish TFA was more of.
I think there's a common misconception that if the developers of ten similar apps would just pool their resources, then one great app would result. Fragmentation certainly can be a problem but I feel like this is some kind of corollary to the mythical man-month.
That the author describes himself as a self-taught handyman may not sound like much to USA readers like myself, but it's my understanding that Canada has a much more widespread formal trade apprentice program than what's available in the states (through unions). I've always wished something like that was developed here, since it doesn't seem like union labor will ever make inroads on the USA residential market.
No it’s not good. The lower the barrier to every the better. More ability to choose the level of quality work you desire. A handyman for bill gates is going to be a complete different class than one for the roach motel but it’s still good if the roach motel can get one.
Mhm. Here in Germany plumber, electrician, etc. are all “protected” jobs. You can’t call yourself a plumber or electrician if you haven’t learned the proper trade. I’d argue it’s good for safety, standards and liability.
Now when it comes to stuff like building a kitchen, you could go with proper carpenters or furniture folks or a handyman. It’s good to have those different categories and know what you would be getting.
Germany takes it too far. Every industry is regulated and licensed. It takes 2 years to call yourself a baker. It’s not a good system.
If you have to regulate industries institute a proficiency test and allow different levels of expertise. A plumber or electrician for home repair is a totally different level of complexity to one in industrial applications but it’s the same certification. The guy doing wiring in my house just does not need that level of education and I don’t really want to pay for it.
If there is a large risk when work is low quality, there is regulation. The more possible danger, the more regulation. This is OK.
But an argument for regulation that is just some hand waving and repeating "standards" is not a legitimate position. It is an attempt to build some little bureaucratic empire which will enrich a group via regulatory capture.
The government wins as they now have more employees and more power and the people already in the trade win. Everyone else loses.
I’m not some free market purist but in this situation we absolutely do not need more government regulation and the risk of low quality work is quite low.
A typical handyman can’t and won’t touch anything that could actually cause harm (gas, plumbing, electrical) and poor quality work will result in a negative reviews and no repeat work for the handyman.
More opportunities for training and upskilling, yes!
But you should not start some unnecessary credentialing/licensing system that restricts the supply.
I think what happens is that the shutdowns are announced more loudly than the launches, which results in people legitimately seeing more shutdowns than launches.
If someone with a non-related degree and profession was interested in a career change to mechatronics/industrial automation and controls, what would people recommend? An engineering degree? Would a one or two year certificate course be enough training to find a good position? Any particular schools that are worth looking at?
I received a degree in electrical and computer eng. In general, CS can get you in the door for industrial automation, particularly with the ability to write simple HTTP servers and a familiarity with networking for embedded products.
As far as "good enough training to find a good position" if you want an "engineer" job title, get a masters degree from an Accreditation Board of Engineering Technologies (ABET) program. You should be able to do this in two years part time study at a local state engineering school. All state engineering schools are approximately equal in this regard.
Controls programming always feels 20 years out of date. 10 years ago I was excited to move projects from ladder to structured text. This year I’m excited to finally use git to manage a PLC code base.
Beckhoff's TwinCAT platform lends itself to git pretty well, there are XML stuff on most files, but it's mostly readable as long as you use STL (Structured Text) which is their recommended language.
Siemens's TIA Portal you can export the source code to make it readable in git as long as you code in SCL (structured control language, which is really structured text), but it's not automatic and a lot of times the binary blob that is the source code and the exported readable code get out of sync cuz of lack of discipline by the programmer.
Rockwell's Studio 5000 can be readable if you use STL (structured text), may be this has changed though, I haven't used it in a long time.
Programming experiences in the industrial controls world is highly dependent on the brand/platform, the IEC 61131-3 describes the standards for industrial controls programming languages, but each brand implements them in a different way so there really is no standard. Writing structured text in one platform does not guarantee the same code will work in a different platform.
If you work at start ups that are in the a manufacturing, processing, or general heavy industry, think SpaceX, Relativity, Rivian, etc. they use structured text (STL) which is like Pascal and usually use Beckhoff's TwinCAT platform which is a modern platform relative to Siemens and Rockwell platforms, but due to their start up nature they usually have a mix of platforms.
I've been working in this field for quite awhile and the systems can be complex and challenging. The software you write controls equipment that can cause massive damage and even death. This leads to very conservative programming languages and frameworks.
The physical dynamics of the processes are the interesting parts for me. Ladder is good for representing Boolean logic but the paradigm has been stretched too far beyond its area of strength.
An engineering degree in electrical, mechanical, robotics, mechatronics, etc. will help get you in the door for the title "Controls Engineer", since you need to know the both hardware and software usually. Having good experience and a portfolio can replace the degree. There are plenty of courses online and in-person that get you into this field pretty quickly, it's relatively easy to learn, but difficult to master due to the diverse applications, platforms, and industries. There are a lot of industrial automation integrators that will take anyone, train them up, and contract them out in exchange for low wages, long work hours, but you get the skills and can then gtfo.
There are also industrial robot programmers who come straight out of high school, get some quick training and go to automotive integrators who contract them out to large automotive companies. This usually requires less engineering knowledge because the industrial robot platforms are relatively easy to learn and what you're doing most of the time is teaching the robot where to go to meet cycle times which is tedious, you use a "pendant" not a laptop usually, but that's changing with all the offline programming software, but reality never matches the simulation.
There reason most people don't like this field is you have to be in the "field", sometimes that's a loud ass, highly dangerous, manufacturing floor, outside in the searing heat or frigid cold, working with people that are.....uh...not the brightest, and work under constant pressure since "controls" is usually the last to get all the specs and the time you get to finish got shortened due to late deliverables from mechanical and electrical. It's not a job you can really do remotely. You can program remotely, but at some point your going to test you code on the machine and you're responsible for not breaking anything or killing anyone.
* no tech work in the U.S. is "easy to get into" at this time. Manufacturing especially is tiers higher than the knowledge taught for any specific degree. Have to go to the right schools that are integrated in that market, or know people.
If you already have any engineering or computer background, you could probably waltz into the lobby of the nearest semiconductor manufacturer and ask for your badge and desk right now.
Finding people who are willing to do proper root cause analysis and not make wild assumptions about everything is pretty much the only important concern for onboarding in most manufacturing*system roles.
You will learn on the job. Go in with an open mind and try to remain humble. Pay will suck at first, but these domains are pretty much limitless in your career potential. I knew of some working in the photolithography area that were virtually classified as a nation state asset. Make yourself that important. You won't get that from college or a certification program. You get that by working with real tools in a real factory and learning about all the strange emergent properties of these systems as they come together.
I worked at BEI Electronics and one key position there was the Industrial Automation role who defined how things got built. Another one to look at is Process Control.
I would like to listen to audiobooks at work, but often there's too much background noise to hear it properly (while not wearing headphones); it makes me wonder if there are audio player apps that will pause and rewind when the mic detects db over a certain threshold.
In my country, no one tips, yet rideshare and food delivery services still exist.
Instead of begging for tips, drivers join a national union and regularly go on strike when a delivery company fails to meet payment expectations. New delivery services typical negotiate a deal with the union first to get drivers, and thus new companies have one less needless marketing expense.
Just don't tip.
It makes no sense for couriers to receive tips based on the price of the package. As far as they should be concerned, a 5$ burger and a 100$ steak are the same thing.
Now that some companies like DoorDash are also branching out and delivering medicines, and it would be preferable if they didn't even know what was inside any package and just charged appropriate by weight, distance, congestion fees for delivery.
