472: Field of Boxes

<laugh> Here with Christopher White. And it is just going to be us chatting. It is going to be us chatting about, well, the biggest thing in my life right now, except for work-

CW (00:01:04):

Dog.

EW (00:01:04):

Dog, husband, house, everything else.

CW (00:01:09):

Could be bigger than some of those.

EW (00:01:11):

Okay. So, any questions?

CW (00:01:15):

About what?

EW (00:01:16):

Well, I could either go chapter by chapter, and talk about the book-

CW (00:01:19):

What book? What is happening? <laugh>

EW (00:01:26):

<laugh> Okay, so-

CW (00:01:28):

You have finished- Not only have you finished your book-

EW (00:01:30):

Today is when-

CW (00:01:31):

But your book has been-

EW (00:01:32):

The electronic book came out.

CW (00:01:33):

Somewhat released.

EW (00:01:35):

On Thursday, March 7th, 2024 the book is out on electronic platforms and it will be paper copy in two weeks.

CW (00:01:48):

And for people who have not been listening to the show, which is tuned back in after five or ten years, this is the second edition of your "Making Embedded Systems," compendium of all things embedded systems firmware.

EW (00:02:02):

Yes, it is my book. Yes.

CW (00:02:04):

Okay, got it. And you finished it last week and somehow it is released today.

EW (00:02:10):

That is not quite true. It has been done since December.

CW (00:02:14):

But your last chance to do anything was last week. I just think that timeline is impressive.

EW (00:02:19):

Well, that is why the paper copies are a couple weeks out yet.

CW (00:02:21):

Still. That is fast. I was surprised that-

EW (00:02:24):

The last thing I did was the second or third quality check.

CW (00:02:29):

I know. It is just I am used to things that- Big publishers taking a while like, "Oh, the book is done," and then you will have it in six or seven months.

EW (00:02:37):

No, O'Reilly is more like you will have it in a month. But I was pretty on top of things, so our last quality pass through did not have a lot of changes.

CW (00:02:46):

Okay. This is the second edition of the book. The book originally came out in a year that I cannot remember the number of.

EW (00:02:53):

I think it was 2011.

CW (00:02:54):

2011? Okay.

EW (00:02:56):

I think it was November of 2011.

CW (00:02:57):

Okay.

EW (00:02:57):

So it was kind of 2012.

CW (00:03:00):

So that is 13 years ago.

EW (00:03:06):

Sure. 13 and almost a half.

CW (00:03:09):

So what possessed you to update the book? Nothing has changed in embedded. I can tell you, because I am still using the same stupid protocols. The data sheets-

EW (00:03:18):

It is still all 8051.

CW (00:03:19):

Data sheets are still written by people who hate you specifically. Yeah. So why did you want to update the book?

EW (00:03:30):

Things have changed. There have been some changes.

CW (00:03:32):

<laugh> What?

EW (00:03:32):

The whole industry is far more simplified in terms of it is all ARMs now. Sure, you get-

CW (00:03:46):

Do not email us. <laugh>

EW (00:03:48):

There are a few that are not, but even PICs are ARMs now.

CW (00:03:52):

<laugh> Even- All right. Yes. Well, the RISC-V people would have- And the Espressif people would probably take issue with some of that. But yes, there is a lot of ARM going around.

EW (00:04:03):

And the tools have changed. And the availability of hobby boards for professional development is just amazing.

CW (00:04:15):

Yes. And the kinds of things that are popular to do with embedded systems now rely heavily on Bluetooth and IoT, which-

EW (00:04:28):

So I had a chapter about how to build your own bootloader in the original edition. Now that chapter is expanded, and it is all about how to manage your IoT system. Because really you should not be building your own bootloader. You need that information enough to understand what somebody else has built for you. But really you should be using Nordic's or TI's or whoever has an OTA or a DFU system.

(00:04:54):

Then it talks about what the security things are that you should be looking for in those. And how to build your probably multi-image with signature thing.

(00:05:06):

It is not like it tells you how to do this for TI or Nordic or ST. It tells you the theory of it, of these are the parts you are looking for, these are the things that you should expect. And if you have to build it yourself, these are the pitfalls. But overall, this is the idea of what you are doing.

CW (00:05:27):

Okay. So that is one specific thing you have expanded. Do you want to go chapter by chapter, or topic by topic? How would you like to talk about your book?

EW (00:05:34):

Well, I could also just tell you everything that went wrong.

CW (00:05:36):

Let us start with everything that went wrong.

EW (00:05:38):

Okay. So chapter two, I go over system diagrams. Because I think it is really important to have the idea of being able to take an idea, and turn it into a napkin sketch, and then turn it into a design. I think architecture works really well like that.

(00:05:58):

And I think if you come to a new company and you are faced with a wall of code, the way to get around it is through diagrams. Is through trying to summarize your information as you learn it, into different kinds of diagrams.

(00:06:11):

I have one that is a block diagram, which I think most people understand. And a layered diagram, which is trying to figure out where the APIs go. And then like an org chart diagram that is where code flows. Like if you are walking through it, you go from "Main" to this subsystem to that subsystem and the control flows. That one is about control.

(00:06:35):

But I added a new one, added one to talk about the context of your diagram. This is really important in IoT stuff, because it is not just "Box." It is "Box" in a field of boxes, and then these all go up to a cloud, and then you have an interface.

(00:06:50):

Even in things like toys, a context diagram is a good idea, because you can get an idea of what it is supposed to do. You are looking at it from the outside, instead of the inside.

CW (00:07:02):

It is kind of a user-

(00:07:03):

It is very much user centric.

(00:07:04):

I hate to say user story, but it is user-

EW (00:07:06):

Yeah.

CW (00:07:06):

It is how it sits in the actual world, instead of looking inside the box and seeing what parts are made of.

EW (00:07:14):

So I made a diagram. My diagrams when I do them, they are on a yellow pad with- I pencil them in and then I ink them, and they are very messy. I am not a graphic artist. I will occasionally draw and I will take a long time to draw something. But diagrams themselves tend to be not great.

(00:07:36):

O'Reilly gives me an artist, Kate Dullea. It is in the book, she is credited. Kate was great. But my notes on this first context diagram had a toy and then it had an IoT system, to compare the complexity of what this diagram might look like, this context diagram.

(00:07:59):

The toy, I said, "Make it a little creepy," just because I thought that would be funny. So she sends me back this diagram. It is like a mouse skull on top of a giant two story tall toy, with people running away from it looking terrified. Stick figures looking terrified, which I thought was fantastic. I mean, how do you make stick figures look terrified?

CW (00:08:28):

I thought it was great. Should have kept it.

EW (00:08:32):

<laugh> That was not what you said initially. I think what you said initially is, "I am going to have nightmares."

CW (00:08:37):

Nah. That was great. No, that was somebody else.

EW (00:08:41):

And so they-

CW (00:08:43):

It was very "Donnie Darko."

EW (00:08:45):

I tried to be polite and subtle and say, "Maybe not." And that apparently did not come through, because I was reviewing all of the images. Some of them I had very technical detailed feedback on. And this one was just a, "Wow, that is terrifying," sort of feedback.

(00:09:04):

Eventually the production editor finally said, "Do you really want this in there like that?" I said, "No!"

CW (00:09:11):

Yes!

EW (00:09:14):

She did a tamer one. It is still pretty creepy, because the notes did say make it creepy. But it is not mouse skull on top of stuffed animal, that is giant, terrifyingly. So that is good.

(00:09:28):

At the end of the book- So my cover animal is a great-eared nightjar-

CW (00:09:35):

Same animal as before. Just different drawing.

EW (00:09:37):

Different drawing, now in color. Which by the way, the e-book is in color, but the paper book is not. There is not a lot of color used. A couple of diagrams are a little simpler with color.

(00:09:47):

But the nightjar- In the first edition, they have this thing at the end that talks about what the cover animal is. They had this long section about goatsuckers and chupacabra. I took that and I was like, "No, we are not doing that this time. This is a nightjar. A nightjar is a well understood animal. It is not called a goatsucker by anybody who even is familiar with this creature." Because nightjars are super cool. So I was excited to be able to mark that off.

(00:10:24):

And then they added a whole bunch of stuff about the sounds it makes, which are kind of creepy, but at least it is not a goatsucker anymore. They kept wanting me to say that the animal was a great-eared goatsucker, and I could not do it. It took me two years to figure out that nobody called it that.

(00:10:41):

The great-eared nightjar is fantastic. If you have ever seen "How to train your dragon," it looks just like the black dragon in "How to train your dragon."

CW (00:10:54):

Sort of. Yeah. Weird tufted ears and a big wide mouth. Strange. Strange bird.

EW (00:11:01):

And fast. They are super fast. What is the name? Toothless. Toothless. So let us see. What else went horribly wrong?

CW (00:11:11):

None of these things sound horribly wrong.

EW (00:11:13):

Well, at the time they were kind of stressful. Like the index. The index was 10% the length of the book. And I sent that back and said, "No, I do not think that is right. I think maybe 5% of the book would be good." Because there was a lot of- It was not just redundancy, it was just stuff that did not belong in there.

(00:11:33):

But then even as I had emailed, "I think this should be shorter." I also wanted to add jokes. So hopefully- I have not seen the final index. I got the final PDF this morning, but I did not look at the index, but hopefully in the index there are-

CW (00:11:51):

Some Easter eggs. That is the secret stuff.

EW (00:11:52):

Some Easter- Oh, you do not want me to tell them?

CW (00:11:54):

Yeah, no, you can find stuff.

EW (00:11:56):

Well, there is one that people know about if they have seen the first edition, but now it is in the index. There is a triceratops data sheet that is now in the index. If you want to look up where the triceratops data sheet is, you can-

CW (00:12:08):

Okay. Well none of these things sound horribly wrong.

EW (00:12:13):

Okay.

CW (00:12:14):

You sound like normal editing. <laugh> "Horribly wrong" would be, "Well, I wrote this whole section about ST, and then they went out of business."

EW (00:12:23):

Oh, no, nothing like that.

CW (00:12:26):

"Horribly wrong" would be, "I wrote all this binary math stuff, and it turns out that binary math is not right."

EW (00:12:33):

Well, when they translated my book from the original, I do not know what language-

CW (00:12:39):

English?

EW (00:12:39):

It was written in-

CW (00:12:41):

English? You wrote in English.

EW (00:12:42):

No, no, no.

CW (00:12:42):

<laugh>

EW (00:12:43):

They encode it. I think it was encoded in XML.

CW (00:12:46):

Oh, okay.

EW (00:12:48):

They said, "Do you want to use our HTML editor, or do you want to use Google Docs?" For the first edition they said, "Do you want to use our XML editor, or do you want to use Word?" And like an idiot, I said, "I will use your XML editor. I will use to Word write and then I will put it all in your XML editor."

(00:13:08):

This time I said, "I want to use Google Docs, because Google Docs is super easy." But that translation, all of the squared, everything that had a superscript-

CW (00:13:22):

Oh, okay.

EW (00:13:24):

Went into the line.

CW (00:13:26):

Went into the line?

EW (00:13:27):

So it was no longer a superscript.

CW (00:13:28):

Oh, it was just a normal number?

EW (00:13:28):

It was just a normal number.

CW (00:13:28):

Oh my God.

EW (00:13:33):

So as I was editing, that was one of the things I was finding. Hopefully I got all of those. Let us see. But that was good. There was stuff like that where I was going through and fixing- I fixed all the errata of course. And then, I needed to break up chapter six.

CW (00:13:58):

What was chapter six?

EW (00:14:00):

The kitchen sink. I think it was called "Peripherals." It was a mashup of listing all of the peripherals, and all of the communication methods. So it was like SPI and IMUs, which yes, they go together, but it was a super long chapter. I was really sad about that, because I liked having ten chapters. Ten was just a good number.

CW (00:14:26):

<laugh> Okay.

EW (00:14:27):

But then that led to a few more chapters. Then I had 13 for the longest time, but I did not really want to have 13. I do not know why. I do not know why the numbers matter. They do not matter. Obviously they do not matter. Now it has 14, because we added one about motors.

(00:14:47):

It also has a hundred more pages. Of course, I do not know where the index came in, but I am pretty sure it is a hundred more pages. Not all of that is index. Which means it also has four more interview questions. Well three more interview questions, and one advice on interviewing.

(00:15:07):

Because each chapter, at the end of the chapter, it has an interview question. If you have not read the book, it will not make a lot of sense. But the interview questions are from the perspective of the interviewer. Like, "I have to interview somebody today. What should I ask them, and what should I expect in response?"

(00:15:25):

I had so much fun writing those, because you are not asking textbook questions of, "What are the pros and cons of I2C versus SPI?" You are talking about how does a system boot up from nothing. Then it goes through and it says you are listening for things like C startup and memory and-

CW (00:15:55):

Vector table.

EW (00:15:56):

How you get to "Main," and vector table. It was way more fun to pose questions that way. And I thought it was more interesting to read an answer that was more on your side, instead of adversarial. Like, "Here is what we would look for," instead of "Your answer is wrong." I had fun writing some new ones of those.

(00:16:24):

What else did I tell people? Oh, like I said, it is an e-book. You can get it from eBooks.com or Google Play or Amazon Kindle. There will be a coupon.

CW (00:16:42):

What is on O'Reilly?

EW (00:16:44):

Does O'Reilly actually sell their books anymore?

CW (00:16:46):

Oh, I do not know.

EW (00:16:47):

Oh, there is the O'Reilly Safari System. It is not even that, what it is called anymore. Oh my God, I am totally showing my age. It is the O'Reilly Learning System. They usually have a seven day trial thing. I have coupons for 30 day trials, which I will write down here, "I will put into the show notes," if you want to try a 30 day O'Reilly thing.

(00:17:07):

I just found that the O'Reilly Learning System has a whole bunch of webinars. I knew they had classes, but I was not really sure about their classes. But the webinars are one to eight hours of pretty detailed focused information. Of course, it is more software than hardware, but still, they were interesting.

(00:17:30):

I think I am doing a webinar, probably in April. I will let everybody know when it happens. I sent over a couple of abstracts this morning. Because I do not really know what to talk to the O'Reilly audience about, because it is not really hardware. Do they really want an introduction to embedded systems? I can totally do that, but it is kind of boring.

CW (00:17:52):

I mean your book is there, so it would seem like something embedded systems would make sense. I do not know what the "O'Reilly audience" means.

EW (00:18:01):

They have a large mailing list of people who have attended these in the past.

CW (00:18:09):

Okay. Yes. What else?

EW (00:18:13):

Oh, I said I broke up chapter six. Now I have a communications chapter, and a chapter about peripherals. The peripherals I handle in a different way, and I talk a lot more about flash memory. Not just about using it internally, but about all the things you can do with externally.

(00:18:32):

Before I had a section about LCDs and fonts and how all of those work. I still have that, but now I talk more about, "Well, how do you get the assets into the flash?" And again, it is not "How do you get the assets into this flash?" It is "How do you go about putting assets in flash? What are your options when you are going in manufacturing? Do you do it through the code? When you update the code, do you think about updating the assets?" It is all of these things, as you are building systems.

(00:19:02):

There is more code now, but not really in the book. There is a GitHub repository that goes with the book. Which is supposed to have a lot of code, but turns out has a lot of links to other people's code. Because why bother to write it myself, if somebody has already done a pretty good job.

(00:19:23):

Let us see, what else did chapter six become? The communications. Communications got split into local communications, and non-local communications?

CW (00:19:36):

Sure.

EW (00:19:38):

Cloudish communications? Then that got mushed into IoT stuff. So that moved around. With my Classpert class, I had this nice section about main loops and how they work. I guess that is something that people do not think about much.

(00:19:58):

You have a main loop and you have while(1) and stuff happens over and over again. After that you add some interrupts. The interrupts do stuff. Maybe the interrupts actually do the stuff, and then you come back to the main and it just spins.

(00:20:19):

But then maybe the interrupts start having event handlers. And main starts looking more like a state machine, an event handler, or event loop. Then it goes through each step one by one, until you end up with active objects. When I did this as a talk with Classpert folks, it was a, "Okay, I have a system that needs to do this. How do we handle that?" and building up the means that way.

(00:20:43):

I did not get to do that so much in the book, because you cannot have that two-way dialogue of course. But I did try to build up why you want to do events, instead of handling things in interrupts. Why active objects come up when you are dealing mostly with frameworks like Nordic and other BLE stuff, where somebody is running a big system and your code is actually pretty small in this system.

(00:21:11):

Let us see, what else?

CW (00:21:13):

Was there anything that was outdated that you cut?

EW (00:21:17):

Little pieces. I switched from an NXP processor in an example, to an ST processor, just because that is- NXP is still around, but I did not know whether to call it NXP or Phillips, so I just gave up on that one. Look ST, yay.

(00:21:34):

Let us see. The motor section I am not entirely happy with. I think some of the diagrams are badly sized, but that is my fault.

CW (00:21:50):

No, you are not coming on the show to badmouth your own book. Motor section is great.

EW (00:21:55):

Last time we talked about my book, I think the show was titled "Worst book ever."

CW (00:22:00):

That was because we were making fun of the one bad review you got on Amazon.

EW (00:22:05):

I got more than one, but yes, the worst. Now is the time for me to say, "Please review the book. Review the book, especially if you like it enough to give it five stars." The reviews are super important. I really appreciate them.

(00:22:20):

I do not care if you review this one or the older one, but put it under the second edition version. That would be really nice to have more reviews there.

CW (00:22:30):

<laugh> Okay.

EW (00:22:31):

I am sorry. Was that aside kind of too weird?

CW (00:22:35):

No. <laugh> All right. Unless you want to read the whole book, I do not know what else to say about the book. It is out. Go get it.

EW (00:22:43):

It is out. Go get it. Yay!

CW (00:22:45):

Or pre-order it for April, if you want in paper. And it should be out April-ish.

EW (00:22:49):

It should be out sooner than April. I think Amazon is showing April 7th, but the paper should ship sooner. There is this whole, if you are late to Amazon, they get really mad, so they put in some buffer.

CW (00:23:03):

Okay.

EW (00:23:03):

I did not quite use that buffer, and I am not quite sure how I did not use that buffer. Because I did put a delay in at some point where I was like, "Nope, this is broken. We got to fix this before we go on."

CW (00:23:16):

The buffer they tell you, and the buffer they tell themselves are probably quite different.

EW (00:23:20):

Yes. Probably. <laugh> I think that is what happened.

CW (00:23:21):

There is a much bigger buffer that is real.

EW (00:23:27):

Yes. I should thank some of the people who helped. Chris Svec at iRobot helped quite a bit with the writing and technical review. Rene Novor, also technical review and writing. He is fantastic at catching code errors, which I am terrible at catching.

(00:23:51):

Phillip Johnston of Embedded Artistry and John Catsoulis were all my technical editors. They did a fantastic job. Any errors that are left are my fault, not theirs, of course. Except for Svec. They may all be Svec's fault. We know.

CW (00:24:10):

I do not think so.

EW (00:24:11):

I asked them what they wanted their titles to be for thanking and sending them stuff. Chris Svec and I almost agree that his title could be, "Not the Embedded co-host." <laugh>

CW (00:24:28):

<laugh>

EW (00:24:28):

But no, he gets some "Director of robots and software engineering" or something that sounds like he is about to take over the world.

CW (00:24:36):

Did you enjoy writing it? Or was it a chore?

EW (00:24:41):

At this point-

CW (00:24:42):

Did it take as long to write the second edition, as the first edition?

EW (00:24:44):

Yeah, it did. It really did. At this point, after the first edition, I was like "Never, ever, ever, ever again." And I had some of that in February as we were what felt like rushing and stressful to me, but probably was not, because they built in buffers that I was not aware of.

(00:25:05):

But overall, it is kind of cool paging through it. It is sure a lot. I did not think I knew that much. Every time I have to turn to a page to explain to somebody- A couple of times I have had people ask me questions. I am like, "Well, I am not going to bother to generate new words for this. I will just turn to this page, and not read it to them, but show them the diagram and explain whatever." That has been really cool.

(00:25:42):

It is always a good feeling, but it is also like, "Okay, I do kind of know what I am talking about here." You are smiling at me like "You are crazy."

CW (00:25:54):

No. <laugh>

EW (00:25:55):

I have seen you give that smile to the dog.

CW (00:25:57):

It is just weird to have imposter syndrome after you have written a book. <laugh> Maybe not weird, but just seems like you have been living it for so long, that you feel like you actually do know what you are talking about.

(00:26:09):

But it was a little different writing it this time, because last time I think you had a full-time job too, right? So you had a little bit more freedom to-

EW (00:26:21):

I was working for the custom chip Cortex-M0 credit card thingamabob.

CW (00:26:29):

Oh, right. No, you were not full time.

EW (00:26:30):

So that was not entirely full time.

CW (00:26:31):

Okay.

EW (00:26:31):

I did not have a podcast though.

CW (00:26:35):

Right.

EW (00:26:40):

I think it was about the same amount of time. Just because it was going through and checking all of the old stuff, and writing the new. And trying to go through some of my other explanation things? Like going through all the Classpert- I would have mini lectures that were strewn throughout the course, and those were all new stuff. Translating those into the book was good. I did not get to do all of them.

(00:27:12):

I did not want to do a whole chapter about internationalization, because it is so boring. But if you start internationalization and you do not know where to start, it is so hard to collect all the information. And I do have a blog post about it, but last time I checked its images were all messed up. I should fix that, but I have not yet. I have been busy writing a book. Aah.

CW (00:27:39):

All right. Well that is the book.

EW (00:27:41):

That is the book. Please take a look. If you hate it, please wait about three months before telling me that. Just because I need a little good time before we get to the bads. If you like it, I do not know. I get weirdly shy if people tell me they like it.

CW (00:28:01):

Just review it.

EW (00:28:02):

Yes, just review it. Review it to other people. Okay, so I thought that was going to take up the whole show.

CW (00:28:10):

That is 28 minutes.

EW (00:28:12):

Okay, wait a minute. I know we have two sponsors right now, and I need to make announcements for both.

CW (00:28:18):

You do?

EW (00:28:20):

Yes.

CW (00:28:20):

Okay.

EW (00:28:20):

For Nordic, who I think is not sponsoring this show, but sponsored the last one and will sponsor the next one. We still have the giveaway. Technically I should be announcing a winner, but I did not choose a winner. So instead, I am going to encourage you all to enter.

(00:28:40):

The current question is, "What would you like to change about the Power Profiler Kit II?What would be different? What could be better?" Send me an email with that, [show at embedded.fm]. Those will all go into a hopper.

(00:28:59):

Then I will share the interesting tidbits with Nordic, but also choose three winners over these three months. Which probably at the rate it is going, is going to be three winners at the end of the three months. Enter early in case I draw early.

CW (00:29:17):

Okay.

EW (00:29:20):

For Memfault, they are doing something a little different. They had been doing the webinar things, and so they are doing a different-

CW (00:29:30):

The panel discussion.

EW (00:29:31):

The panel discussion. Now they are going to have- Actually, why do I not just read this?

CW (00:29:40):

Yeah.

EW (00:29:46):

Memfault is hosting its first launch week of the year. On Tuesday, March 12th, which is very soon, the Memfault CEO François Baldassari will showcase how to evaluate the health and performance of embedded devices with Memfault's observability platform.

(00:30:08):

Join this webinar to discover how simple it is to monitor three necessary device measures: stability, battery, and connectivity. There will be a registration link in the show notes, and I think that is going to be pretty cool. We really do appreciate Memfault for sponsoring this show, and several others over the last few months.

(00:30:18):

Using their observability platform, it will help you to figure out how to deal with your systems if it gets larger, how to debug it, how to make sure it is healthy. It can be IoT or distributed systems, or just a whole bunch of small systems connected together. That is what Memfault is there for.

CW (00:30:37):

Cool.

EW (00:30:38):

How are you doing?

CW (00:30:40):

I am all right.

EW (00:30:44):

<laugh> You got a piano not too long ago.

CW (00:30:48):

Okay. I do not know what that has to do with anything, but continue.

EW (00:30:52):

And one of our show friends asked about learning piano as a beginner.

CW (00:31:01):

Okay.

EW (00:31:02):

What advice do you have for somebody learning piano as a beginner?

CW (00:31:06):

Get a teacher.

EW (00:31:08):

Okay.

CW (00:31:09):

I mean, that is not necessarily obvious. There are a lot of resources out there, right, on YouTube and books now, that were not when I was learning or starting to learn back in the eighties. If you wanted to learn an instrument, you had to get a teacher, pretty much.

EW (00:31:26):

Do you still have to? You said, "Get a teacher."

CW (00:31:29):

It is helpful. It is really helpful, because <sigh> there is a lot of stuff- Unless you are really rigorous about recording yourself, and zeroing in on your hands and things. When you are learning technique, on any instrument, not just piano, it is real hard to critique yourself if you do not know what is correct.

(00:31:48):

The best you can do when you are first starting, and even when you are advanced, sometimes there is stuff where somebody needs to point out, "Okay, here is how you should be doing this technique to be more efficient, or put less stress on your wrists or hands or whatever."

(00:32:04):

There is stuff you can watch on YouTube. It is like, "Here is the proper technique and here is what your hand is doing." But it is real hard, if you do not know anything, to figure out how you are supposed to hold your hand. You can get part of the way there. But it is helpful to have a teacher.

(00:32:21):

And it is helpful to have a teacher to motivate you, because I think, at least for me, I am at the point where I can self-motivate some, but it took a long time to get there. When I was a kid learning- I do not know how old this person was, is, but when I was a-

EW (00:32:38):

Late twenties.

CW (00:32:39):

Yeah. When was a kid learning, there was no way I was going to practice without some mean lady coming yell at me every week. And even now, I think I should have-

EW (00:32:48):

Hoping his teacher is not listening.

CW (00:32:52):

<laugh> It is a metaphorical mean lady.

EW (00:32:53):

<laugh>

CW (00:32:53):

Even now I think I should get instructed periodically for all the instruments I play, because there is that aspect of it. It is like, "Okay, I need to practice." I can go practice for an hour and get nothing done, because I am ADHD about it, and just flitting around between things and not being serious.

(00:33:10):

Whereas if I had a teacher say, "This week you are going to do these three things, for at least ten minutes a day," and then have somebody to say, "Did you do that?"

EW (00:33:20):

But you have had drum teachers who-

CW (00:33:22):

I have had bad teachers. Yeah, that is true.

EW (00:33:24):

Oh, okay. But they did not have a plan like that.

CW (00:33:27):

But I was already somewhat intermediate advanced at that point, which is kind of paradoxical. If I had been a beginner it would have been fine. But sometimes once you get to a particular level, the teachers do not know how to teach you.

EW (00:33:39):

Or they treat you more as colleagues, with suggestions instead of-

CW (00:33:43):

Yeah. Which they should not have, because way better than me. But anyway. So teacher is good. If you do not have a piano, there are a million options out there. I would not- It depends on how serious this person is.

EW (00:33:56):

Let us say serious, just because that is the most interesting.

CW (00:33:58):

If you are somewhat serious, then you want a piano keyboard. The keyboard mechanism. Ignore the sounds for now. You want a keyboard mechanism that as closely approximates how the key mechanism of an actual piano feels, as you can afford. That is where I would put the money, is on better key action.

EW (00:34:20):

This is because with real piano, when you push it down there are-

CW (00:34:25):

There are linkages and then levers and then a hammer, that goes and strikes the strings.

EW (00:34:31):

So you have a force to it. And if you push hard, you get a different sound than if you push softly.

CW (00:34:38):

Right. But you can get that effect with cheapo synthesizer keys, where they just have a pressure sensor.

EW (00:34:43):

But it is the other effect. It is the feedback, that is-

CW (00:34:46):

It is the weight of the key, the travel distance. There are actually things where as you press a key, as you go through- It is called- shoot, what is it called? Not "Lift off." I cannot remember the term.

(00:35:03):

But on a real piano, there is a moment where the hammer starts to move in a certain way, and there is a little hitch, and then past that, it eases up. That is not super important to simulate necessarily, but a lot of the electronic keyboards do simulate that, so it feels right.

(00:35:20):

Yeah. There is a whole- Your fingers- There is a whole technique thing with the way actual piano key actions work, that you do not get with a synthesizer action keyboard. Which is basically a switch or a pressure sensor, and a spring.

EW (00:35:39):

And that matters.

CW (00:35:41):

It matters if-

EW (00:35:41):

That matters if you are going to play piano someday.

CW (00:35:42):

It matters if you want to play piano. If you want to learn keyboards to do some synth stuff or organ music, then it does not matter so much. So that is a distinction. If you want to learn piano, you get something that feels like a piano.

(00:35:58):

There are a whole range of things. You can get just a controller that has weighted key action, or what sometimes they call "Grand-" A lot of them are called "Grand feel actions." There are a whole bunch of different kinds.

(00:36:10):

Roland has them, Kawai has them, Yamaha has them, Casio has them. They all have different names and numbers. And they get more expensive, the more closely they approximate a key action.

(00:36:21):

Some of them feel pretty good, but they are just plastic mechanisms that are simulating the feel, without actually simulating the whole mechanism. Some go all the way up to where it is a piano keyboard mechanism, and then the hammer goes and strikes a sensor. So they go all the way. This is a piano keyboard, and the hammer strikes a sensor. That is like the one I have. So there is a range there.

(00:36:48):

For the sound part, you can get some that have good sounds in them. Most of them will have adequate sounds in them. You can get some that have no sounds in them.

(00:36:55):

They are just midi keyboards that you plug into either a tablet or a Raspberry Pi or a laptop or whatever. Then you would have to buy some software that is a sampling library for pianos, which sounds really good. So you can do it that way too. And then you can always keep upgrading the sounds as things get better.

(00:37:15):

The other slight thing that is nice to have, is you do want to have the pedals. At least the damper pedal, which is where people- You hold it down and the notes keep going, the sustain pedal. You at least want to have one pedal for that, because the damper pedal is super important for most piano music.

(00:37:32):

You can buy those externally and just have it on the floor with a cable to them. Some of the other models have a whole cabinet thing, and that is part of the cabinet on the floor.

(00:37:46):

Then you want a good seat. You do not want to be sitting in a chair that is at the wrong height, or a weird bench that is uncomfortable. So you want to make sure you get something that is at the right height, and is not uncomfortable to sit on.

EW (00:38:00):

Because like tapping on a keyboard, it is pretty easy to be holding your arm, hand, wrist, fingers wrong. And since you are doing it a whole bunch, and you are doing it wrong-

CW (00:38:11):

Piano is a repetitive stress injury magnet, if you do it wrong. Especially if you practice a lot.

EW (00:38:18):

It is funny to hear you say teacher, because I know you do not have one. I know you have been in classes and stuff lately. But you do not have any dedicated teachers for any of the instruments. And yet I totally see where you are coming from.

(00:38:30):

I think about learning yoga. A physical activity kind of needs a physical teacher. You can learn yoga from videos, but until you have somebody actually correct you and say, "This is what it is supposed to feel like," it is hard to really learn what you are doing.

CW (00:38:49):

Yeah. And sometimes somebody needs to reach across, and grab your hand and twist it. Or hold your elbow up and say, "This is what this should be." Or show you their hand while they are doing something, and you can look around. You cannot do that on YouTube.

EW (00:39:02):

Yeah, YouTube you cannot see behind.

CW (00:39:04):

Or they can hold it in a particular way and say, "Okay, see this." There are some really good piano technique videos out there. But again, it is a different experience, and you have to- It depends on how you learn.

(00:39:18):

It is true all instruments. All instruments involve technique at some level. Sometimes it gets really esoteric, and some of that stuff you may not get from a teacher, and you will have to learn for yourself once you have got fundamentals.

EW (00:39:33):

You signed up for an online course system for piano?

CW (00:39:36):

Yeah.

EW (00:39:36):

Do you like it?

CW (00:39:37):

I do. I wish I had more time for it, since it is another thing I have to add to my day. Well- There are lessons. I can take their video lessons. Some of them very good. I am learning technique things that I did not know, and that I wish my teacher had told me.

(00:39:56):

But they are not like- The thing with videos, it is often like, "Give me the thing that within five minutes, that I did not know before will just transform my playing."

EW (00:40:04):

<laugh>

CW (00:40:04):

It is never like that.

EW (00:40:06):

YouTube has plenty of videos that say they will do that for you.

CW (00:40:09):

But a lot of these course things are like, "Well, here is how to get faster at say, scales or some fundamental technique." And they will tell me something that is like, "Oh, I had no idea that was how we were supposed to do it." That is actually where the wall is in speed. If you do not do this, you cannot go any faster.

(00:40:23):

But that requires extra practice. Like, "Okay. Now I know the secret." But the secret does not make you faster. The secret puts you on-

EW (00:40:34):

Enables.

CW (00:40:34):

The secret puts you on a new ground floor of practice, to actually implement the secret. But teachers, they have things like that. They have secrets. I learned some from the instructor that did the class, the piano class I took, the group class.

(00:40:50):

They mentioned there was something about scales and fingering. It was like, "Oh, with this kind of scale, your fourth finger always goes on B flat." Or? Yeah, I think it was B flat, or something like that. I was like, "Oh, if I had known that, I would have had to memorize a whole bunch of things. It would have been much easier." But anyway, there was a lot of stuff like that.

EW (00:41:10):

Do you want to recommend to the class system you are taking?

CW (00:41:13):

No.

EW (00:41:13):

Okay.

CW (00:41:13):

No, it is for advanced, at least intermediate to advanced, players. So there is not a lot of beginner stuff. It is called "tonebase." They have several instruments. It is all classical for now. I think there is a little bit of jazz. But they have piano, cello, flute and guitar. Classical.

(00:41:29):

They are big, big site with lots of videos. You can take a quiz, an online quiz, and they will try to guide you through a course thing, with various videos and various techniques and things.

(00:41:44):

It is helpful for me, because I am coming back to piano after not having played it seriously for 30 years. <laugh> So I am trying to take it a little more seriously, and go back and like, "Okay, how should I be practicing? What are some techniques that I kind of know how to do, but maybe I am not doing appropriately." And things like that.

(00:42:05):

But yeah, there are a lot of options for pianos out there. Anywhere from probably- It is going to be a few hundred dollars to many thousands.

EW (00:42:17):

Many, many, many thousands. If you want to get really fancy.

CW (00:42:21):

But for a beginner, you should be able to do something that would serve you for years, for between five hundred and a thousand, I would think. And you want the full 88 keys too. So there is a space cost involved, with even this most compact keyboard.

EW (00:42:42):

If you were somebody who wanted to learn to play music as an adult, would you suggest piano or guitar or drums for their anger?

CW (00:42:52):

Drums is very specific. You have to want to play certain kinds of music and stuff. Drums, unless you buy eDrums, which tend to be more expensive than a basic set of acoustic drums. It go either way, but they are very loud.

EW (00:43:07):

They are.

CW (00:43:09):

Even eDrums, they are-

EW (00:43:12):

Oh, they are so much quieter.

CW (00:43:13):

They are so much quieter. But if you are in an apartment-

EW (00:43:16):

Yes, we did have that guy, when we lived in an apartment, who had the eDrums. It does not help when all you are hearing is thwapping. And the thwapping is not even rhythmic, because they are playing different drums. But you can only hear all the thwaps.

CW (00:43:29):

Yeah, that is what mine sound like, but you do not care. Yeah, they are annoying. I would say eDrums are annoying, and real drums are catastrophic.

EW (00:43:40):

<laugh>

CW (00:43:44):

Especially for the first couple of years where you are not good, which is going to be the case. Hearing somebody play loud drums, arrhythmically is really hard to take. So I apologize to all of our neighbors from-

EW (00:44:00):

<laugh>

CW (00:44:02):

2001 or so, to 2005 or so. Yeah, real drums are super loud, and you have to even manage your own hearing and stuff with them. I would not recommend starting with drums.

(00:44:13):

Guitar. Guitar is very frustrating to start with.

EW (00:44:20):

I heard it was only the F sharp chord that was hard. <laugh>

CW (00:44:24):

EW (00:44:24):

Like I know.

CW (00:44:26):

The F barre chord that everybody harps on, even though you should not bother playing the whole thing. Guitar hurts.

EW (00:44:33):

Your fingers, until you get calluses.

CW (00:44:35):

There is a whole thing with your fingers. There are lots of strange positions you have to bend your fingers into.

EW (00:44:43):

What about repetitive stress injuries?

CW (00:44:44):

To push them into these metal strings, that push back on you. On the one hand, while you are first learning guitar, you are limited in how much you can practice. Because after about a half an hour for a while, you are just going to be like, "I cannot do this. This hurts too much." So that takes a while.

(00:44:59):

Like all things learning, there is the learning curve, where you make a lot of progress early on. But guitar, I feel like you do make that progress, but also it hurts.

(00:45:11):

And there is a bit more to think about. Because you not only have to think about your fretting hand, which holds the notes down. But you have to think about your strumming or picking hand, which is doing rhythmic stuff and playing those notes.

(00:45:25):

And there is a lot of technique stuff that you have to get down before you sound like anything. So you have to be able to fret cleanly without buzzing, and that is going to be harder for different chords. You have to be able to pick in time. You have to be able to pick the right strings without bumping the other strings. You have to sometimes mute stuff that you do not want to ring.

(00:45:46):

There is a lot of stuff with guitar that before you sound okay, you have to get a feel for. You can get to strumming chords and stuff pretty easily, but anything beyond that is harder.

(00:45:58):

Piano. Piano, you can sit down and you can push one key and it is going to make the sound. You can push three keys and it is going to make a chord. And there is a lot of-

EW (00:46:15):

That is my level of playing.

CW (00:46:16):

There is a lot of technique stuff with piano. But the technique stuff comes in feel, like bringing out different notes when you are playing multiple notes at once. Rhythmic sense. All kind of feel stuff.

(00:46:31):

And then techniques, so you are not bumping other notes. And you have got some fluidity and stuff-

EW (00:46:36):

And speed.

CW (00:46:36):

And rhythmic and speed. But the learning curve of piano is much more satisfying, I think. Because there is not that, "Okay, I have to learn this, but also I cannot work the instrument until I learn how to do some things."

(00:46:56):

Guitar is hard. Something like violin or flute or any of the other stringed instruments with bows and stuff, those are in a completely different league, where you are going to sound like a murderer of cats for months before you can- There is a hierarchy <laugh> of pain.

EW (00:47:15):

Oh, definitely.

CW (00:47:15):

But yeah, guitar is harder.

EW (00:47:19):

But the notation is easier.

CW (00:47:21):

Guitar? No! Guitar is much worse. Guitar is much worse. Well, there is-

EW (00:47:27):

There are just those little boxes.

CW (00:47:32):

<sigh> There are three kinds of notation for guitar. There is chord box notation, where it puts little dots on the string locations and you can put your fingering. That is how you can learn some chords.

(00:47:40):

There is tablature, which kind of puts the strings on there, and then puts Xs where the notes, as they go along.

EW (00:47:50):

It is like "Guitar Hero."

CW (00:47:51):

That is okay. It is a bit of a crutch, because you cannot notate a lot of things in there. It is not as easy to see rhythm, it is not as easy to see dynamic markings and stuff. A lot of people use tab. It is not a crutch. I use it all the time. I used the wrong word. But it is limited.

(00:48:11):

You cannot communicate with other musicians through tablature. If I am in a group and I want to show the piano player a part-

EW (00:48:19):

Oh. Oh.

CW (00:48:20):

They are going to look at me like, "What the hell are you giving me this for? I do not even know what that means."

(00:48:23):

So if you want to be an intermediate or advanced player, you should be able to somewhat read music on guitar. And that is trickier because guitar, the way the strings are laid out, there are multiple places for notes. You have to make some choices when you are reading. Anyway, it is more complicated. Piano is one note, one key.

EW (00:48:46):

What you are saying is I should just stick to the ocarina.

CW (00:48:49):

Sure. Do you have one?

EW (00:48:50):

No.

CW (00:48:51):

Okay. I do not know why I have gone on a rant in this episode of Embedded, about pianos.

EW (00:48:57):

Welcome to Embedded, the show for people who enjoy listening to random things. Okay, let us see. The last thing I have on my list actually went into the newsletter last week. I should not have wasted it.

CW (00:49:11):

Well, not everybody is on the newsletter.

EW (00:49:14):

What? Why not?

CW (00:49:15):

Well, just because they hate us, I assume.

EW (00:49:17):

Okay. So we have a newsletter. I think it is under "Subscribe," if you go to the website. It comes out every week. On the weeks where we do a show, it is usually what we said on the show. On the weeks where we do not do a show, it is whatever we think of putting up there. Plus every week there are a random assortment of links.

CW (00:49:36):

A random assortment of links.

EW (00:49:38):

Sometimes they are really, really good. Sometimes they are-

CW (00:49:41):

Really, really random.

EW (00:49:42):

Really, really random. Indeed. Okay.

CW (00:49:43):

<laugh>

EW (00:49:46):

So last week, Ben Henke of Pixelblaze asked me about origami. I do not know if he wanted a short answer, <laugh> or if he realized that asking about origami was kind of like, "Elecia, please talk for 45 minutes."

CW (00:50:05):

Like asking you about pianos?

EW (00:50:09):

<laugh> Ben asked if I knew of any origami applications for flex PCBs. He was not sure if he was interested in pure origami, but related folding structures or self arranging structures or kirigami, which is a form of paper art. He was thinking about different ways a folded structural flex PCB could be interesting.

(00:50:31):

Then linked to this video called "This piece of paper could revolutionize human waste," which is not what you think. It is the paper that DigiKey sends, the Geami paper, that has a whole bunch of holes in it.

(00:50:52):

I responded that I spend an unholy amount of time thinking about this. Not that I have ever done it. Just that I think that there are a lot of interesting things you can do with flex PCBs and how they are folded. Of course, flex PCBs do not fold like paper. If you did, that would break them, probably, or maybe just break the circuit.

(00:51:14):

But whenever you are in the spot of things that cannot really be folded folded, you can create hinges instead. And paper artists as a whole talk more about hinges than folds. And in flex PCB, hinges are- Let us say you take a flat PCB, and then you cut- You are going to have to describe what my hands are doing. <laugh>

CW (00:51:47):

<laugh> Okay.

EW (00:51:49):

Let us see. It is a rectangle and you have it a long ways horizontal.

CW (00:51:55):

Okay, I am holding a PCB.

EW (00:51:58):

And then you cut down the center, but only in the center you leave two strips along the edges.

CW (00:52:06):

Okay, so you have got a four inch PCB, and you might cut the middle two inches.

EW (00:52:12):

You might cut a four inch PCB.

CW (00:52:14):

A line.

EW (00:52:15):

You might cut say a four by two inch PCB.

CW (00:52:19):

Yeah.

EW (00:52:21):

And you cut one and a half inches vertically and then a quarter inch horizontally. So now you have-

CW (00:52:29):

A cross?

EW (00:52:30):

A rectangle that is horizontally big, and then you cut a hole that is vertically big.

CW (00:52:36):

Sure.

EW (00:52:38):

This would be much easier with a whiteboard.

CW (00:52:40):

So now I have a little rectangle in the rectangle.

EW (00:52:41):

Right.

CW (00:52:41):

Okay. Go on.

EW (00:52:44):

Your PCB will fold around that much easier, than it would fold elsewhere. Because you have removed a bunch of pieces, a bunch of paper, and now it can fold back and forth along that.

(00:52:53):

Okay, so this hinge, if you did something like circular pleats, which are really cool- It is when you make a paper- Back to paper. You make a circle of paper, and then you cut out the center, so that you have a donut of paper. Then you make concentric rings inside the paper.

(00:53:13):

And then the first ring you fold up, and the second ring you fold down. Third up. Down, up, down. Okay. So you do this, and in the end you do not end up with a donut that is like a fan. You end up with this saddle shape that starts doing crazy things, the tighter you make the folds.

CW (00:53:33):

Okay.

EW (00:53:33):

I call them "chaos noodles," and I just love them. They are well known in origami. This is not me inventing things. But I think you could do that with a saddle, where you would make arcs instead of folds, and you could still keep the, and that would create this hinging effect, but also the unstable waviness.

(00:53:56):

So I have been thinking about that and how that might be fun for either antennas, which probably has no particular use, or actuators like Carl Bugeja has been doing. Which would it undulate? Or would it do something else? Maybe it would just go into a fixed pattern and then flap back and forth. Although that might be a cute flower effect. I do not know.

(00:54:25):

Anyway, there is also a bunch of origami to do with antennas in 3Ds. And there is so much stuff in space. Anything where you need something to get smaller and bigger.

CW (00:54:42):

Ferris wheels, for instance.

EW (00:54:44):

Yeah, we saw that Ferris wheel, that one is all collapsed down into a box, and then you had a cherry picker?

CW (00:54:53):

Yeah, something like-

EW (00:54:53):

To crane.

CW (00:54:54):

A forklift of some kind.

EW (00:54:56):

And it lifted it up and it was like-

CW (00:54:58):

And then it unfolded itself into-

EW (00:54:59):

It just unspoiled. Yeah, it was cool. So yeah, origami. Stents was the other thing. You think about putting things in the body.

(00:55:10):

There is a whole area, a whole giant field of origami surrounded Miura-ori. Miura is the guy. Then there are a bunch of sub things of Yoshizawa that is a form of Miura. These are all things that collapse, and how they get smaller and bigger.

(00:55:35):

Going back to hinges, you can make the hinges so that they are active. And then that spreads out, like the solar panels on many of the space stuffs, space satellites. Satellites.

CW (00:55:49):

Probes, I believe.

EW (00:55:50):

Probes? Yeah. Those are the ones that spin out. Again, we are talking about hinges. But in origami, the hinges are folds. In flex, it is probably cutouts.

(00:56:04):

Stents have similar things, where the hinges can be hard or soft. If they are hard, they are big. And if they are soft, they are small and collapsed. So you can have them go big or small. It is not exactly soft robotics, but there is definitely some crossover.

(00:56:21):

Anyway, we are not going to put all the links into the show notes, because in the show notes we go into the newsletter, and then we have some sort of weird inception problem. But I will put up some of the notes for the origami. I did add notes about all of these terms that I have been using.

CW (00:56:41):

Okay. Well, I cannot think of anything else random to talk about, that has nothing to do with embedded systems.

EW (00:56:46):

I wrote a book about embedded systems. I do not have to talk about it anymore. I put down all of my knowledge there.

CW (00:56:52):

<laugh> That is right. Well, the show is now random topic that we like.

EW (00:56:57):

Anyway, back to embedded systems. I will be giving a talk at the Embedded Online Conference. It is going to be from one of the new chapters in the book. The talk is called "Causing Chaos and Hard Faults."

(00:57:11):

The chapter in the book is, I believe, "Getting into Trouble," in which I talk about hard faults and debugging and memory issues. Just all of the things that you get into trouble, and how do you start debugging it.

(00:57:28):

For the talk, it will be let us get into trouble on purpose for fun, and then we will figure out how to get out of it. I am looking forward to that. The current plan for that involves puppets. I am sure that Christopher will be talking me out of that. But in the meantime-

CW (00:57:45):

When is this happening again?

EW (00:57:47):

End of April, beginning of May.

CW (00:57:48):

End of April. Okay. All right. <laugh> Better start thinking about that.

EW (00:57:53):

I have, that is how I got to the puppets.

CW (00:57:56):

If there is going to be puppets, you need to write a script.

EW (00:57:59):

Well, yes, no. I know which parts the puppets would say.

CW (00:58:02):

You have to obtain the puppets. You have to kind of block out what the puppets are going to be doing. I thought you were doing something where there was going to be an evil you, and a good you.

EW (00:58:10):

I figured a puppet would be a lot easier, than trying to play myself in two different roles.

CW (00:58:14):

I am not sure. We will find out.

EW (00:58:17):

Well, the only good puppet I have right now is that frog, which I like a lot. But the only good puppet action I know how to do it-

CW (00:58:24):

Does this mean we can buy puppets and write them off?

EW (00:58:27):

Well, the frog, the only thing I know how to do it, is to make it look like it is barfing up a cat hairball. Which is not what I need for this.

CW (00:58:33):

That is why you need a puppetry teacher. They can teach you fundamental techniques.

EW (00:58:36):

I need a puppetry teacher, and a bunch of puppets. And that is it for Embedded.

CW (00:58:42):

Good. Yes. Because I have to edit this. And I also have to do actual work. And I have to put it up.

EW (00:58:48):

Right. Okay. He wants me to stop talking. So thank you for listening. Thank you to the Embedded Patreon Slack for their support through writing a book, and for their random questions that lead me to long diatribes. This week it was the Origami, and then there was an internationalizations one, and who knows what is coming up next. Did I say thank you for listening?

CW (00:59:14):

Yeah.

EW (00:59:14):

Oh, then thank you to Memfault for sponsoring the show. And if you would like to contact us, [show at embedded.fm]. And let us see. I guess I am only reading a little bit of "Winnie the Pooh" today, but it is a good time because it is somebody's birthday there too. Happy birthday to my book. Happy birthday. Oh, okay.

(00:59:37):

[Winnie the Pooh excerpt]