Frustrations

Frustrations

Following other developers, learners and makers are great. It facilitates learning and gives inspiration

But one thing that is often missing from people telling about how it is going, is the failures, frustrations and problems any developer will run into.

For this reason, two of my favorite maker youtube channels are Extractions&Ire (Chemistry) and Code Bullet (machine learning). Because these madlads are brave enough to not just show their process and result, but also their failures, mistakes and errors. And how they overcome them. Not always by learning (Sometimes making a dumb mistake is not really something you can learn from...)

It's good, because it's real.

Code tutorials and guides can give the impression that the normal process of development is "Open IDE, code, fix tiny typo error, compile, success". They don't do it out of malice, but out of a want to be concise. Which is fair.

So I also want to share when things do not go so well. I have programmed Atmel's AVR Chips for quite a while now. But I have done it mostly in microchip studio(former Atmel studio) and a bit in the arduino IDE. A job I am currently applying for, uses visual studio code. Which is fair enough. So to prepare for this specific job, and to acquire this quite good-to-have skill, I want to set that up for myself First things first, since I have not done this before, I cannot know if my code would have a weird error so I want to know everything else is working first. So I write a tiny program which simply have the microcontroller increase a number every 2 seconds and write it to my PC over UART. Takes 2 minutes.... I grab one of my Arduino Nano boards and a USB cable for it. And then... I cannot flash it... Its communication protocol have troubles.

I have seen this before. It is to do with the cables not being correct. If they are USB 2.0, very little magnetic noise can cause trouble. (And you cannot tell if a cable runs USB 2.0 or 3.0 by looking at it... because the universal serial bus is not universal... Insert grump rant here) I then spend an hour finding and trying different USB A to USB B-mini cables. Give up, notes down to buy (and MARK) some USB 3.0 versions for the future. I then grab a Arduino Uni instead, as they use USB B, which is much more resistant to noise... And then spend half an hour trying to find a the cable, as I do not have a lot of them, since... nearly nothing uses them. Finally find it, and yes, the program can now be flashed. So I packed all the cables I tested back in their places, after marking them so I will(hoefully) not have to do this again. Had to take several breaks feeling depressed and grumpy, and all in all, this adventure took 4-5 hours. And now I can START on this... And this is how work sometimes is. And that is ok. It is still... VERY frustrating ...

Sometimes you just accept a flaw and move along

So I am building a robot from scratch. It just needs to be a platform for my other projects. It takes Twist commands (Basically, orders how fast to move in what directions) and return odometry (Basically, where the robot thinks it is), both over wifi via the HTTP protocol. Because the odometry math is a lot easier with stepper motors(motors that moves in tiny discreet steps. Usually several hundreds for a single rotation), these are what I am using: I am using 2 TMC2100 boards to drive them. These have 2 pins that you can set to Ground, supply or let them float, and they are used to configure what amount of micro-stepping you want to do (If a stepper motor have 200 steps for a rotation, micro stepping 4 means it takes 4 smaller steps to do one of those 200 "real" steps, and so takes 800 steps for a full rotation) All great... eeeeexept... for some reason the configuration pins just does... nothing, no matter what I set them to. The boards are just stuck at 16 micro-steps, as if both pins where set to supply. This limits the robots speed to 0.3 meters per second... which is... fine. The amount of work and money it would take to fix this(Likely I would have to buy new boards) is not worth a speed boost that is not needed for the first 2-3 projects I want to test on top of it... so I am just leaving it. "Perfect is the enemy of good" and all that... but it still feels unsatisfying... blaaa

Every Programming expert on the planet: "NEVER throw an exception as normal program flow" Kotlin: "Good practices are for OTHER people!"

Teaching programming

One of the things I want to do is teach. My plan is, now that I (F'''ucking FINALLY!) finished my engineering degree, is to work a few years in the robotics industry, and then spend half my worktime there, and half teaching kids programming and code. Because,

A:

I feel that in the same way that you need to know physics, reading, internet literacy, history and a bunch of other skills to navigate the modern world, you also need programming. Not that everyone should become a software engineer. But everyone should be able to set up a few scripts to automate things on their PC, and know enough about how code works to know how the software that runs the world works. In the same way that you know enough about physics to know how satellites giving you internet works. Could you build that? No, of course not. Nor should you. But you know things can go in orbit. That things can communicate wirelessly over long distances and so on. You know enough about mechanics to use a screwdriver to take simple things apart and clean them, and know that you do NOT know enough to fix the electronics in your dishwasher.

You know it is not magic. Yet... programming is often thought of as that. And it really should not be. I swear to you. It is easy. Not to develop code to run a car, but to have it automatically edit text in a file according to rules you made? You betcha. B:

While learning to program as a job takes years, learning enough to understand the basics, and being able and unafraid to set up a script or throw together a python program to automatically sort files on your work computer really does not take a lot of time or effort. And I will guarantee, most students will benefit a lot from it. C:

Programming is much more a craft, like mechanic or carpenter. But more appealing for kids who are not confident in working with their hands. There is also good reasons for why it used to be thought of as a thing people who did not get along with others did. It is, in essence, defining and solving puzzles. To solve problems, that you also have to define yourself. Many a child have found that programming was a thing they could do, and be good at. A thing that helped them be confident. Maker spaces and learning programs with robotics or game development is filled with youngsters eager to learn. Who have finally find a thing they can be good at! But these programs a rare. And often schools and teachers mismanage how to run them. Not out of malice (There are few professions that is filled with more kindhearted people that deserves our endless appreciation than teachers), but because there are endless well paid and prestigious career opportunities for software developers. And so most schools will not have anyone who can even advice them when they set up these things. There is so much good to do. And I want to do some of it!

My place is never cleaner than when I have coding to do that I really do not want to do

Have you seen a girl boy

I saw a... WONDERFULL poem that really hit me, but I needed to fix it a bit first, before I send it to my boys .Original from here: