0:06

Welcome to Embedded. I

0:09

am Elysia White alongside

0:12

Christopher White. It's just us this

0:14

week. And we have many

0:16

things to talk about.

0:17

And we have a new co-host

0:19

I'd like to introduce. Say hello,

0:21

Jojo.

0:23

Jojo is our new little

0:25

dog. Who doesn't

0:27

say much. Who doesn't say much. This is not going

0:29

to be a dog cast. Yes. Like

0:32

the podcats. I promise.

0:34

You posted on

0:37

LinkedIn that you were looking

0:39

for a job. And I have some questions about that. A

0:41

contract.

0:42

I have a job. I

0:44

am fully employed by Logical Elegance Incorporated.

0:47

And in fact, I'm an officer of said corporation.

0:50

I'm not looking for a job. I'm looking for

0:52

clients of Logical Elegance.

0:55

Who, which, to which I might write

0:57

code. At which. For

1:00

which. Got it. Sandwich. Do

1:02

you want a job with CAN? I

1:06

would accept a job with CAN.

1:09

Do you want a job with LAN? I

1:13

will not do a LAN. What

1:15

about HAM? Wait, what?

1:19

As in green eggs and ham. I was going

1:21

for a whole poem here. I heard HAN

1:23

and I thought that was something else I didn't know. It says in the show

1:25

notes, Chris is looking for a job in poem

1:28

form. I got that when we were headed to

1:31

Dr. Seuss. I just didn't know what HAN was.

1:33

I heard HAN. I thought you were talking about a

1:35

Fourier transform window.

1:38

I will not, I

1:41

will not process anything sold,

1:44

sell anything bought or processed, process

1:48

anything bought or sold. I have,

1:50

yeah, I think we're. Okay. Yes, I posted

1:52

on LinkedIn. I have had many responses.

1:55

Thank you

1:56

to those responders. And so you're probably

1:58

not looking anymore? I'm

2:01

in talks with several entities. But

2:05

if more entities would like to talk,

2:07

I can talk with them.

2:09

Perhaps they would like to talk to you.

2:12

I'm pretty booked until

2:13

January. Perhaps they'd like to talk to Jojo.

2:16

Jojo would like any work

2:18

that involves... Hiding. Hiding,

2:20

yes. Which, you

2:23

know, to be fair, I share

2:24

that.

2:25

I think Jojo would be a really good manager as

2:27

long as the answer is

2:30

always nothing. Don't

2:32

move. Yeah. Yeah.

2:35

Yes, so I'm on

2:37

the contract train. I expect to

2:39

be

2:40

having way too much to do in the near

2:42

future.

2:42

That's okay. I guess.

2:45

But, yeah, it's probably time for me to start

2:48

doing things.

2:49

The holidays are coming up, so, you

2:51

know, people won't be expecting

2:52

quite as much. I was going to enter

2:54

this show asking you... telling you that I wasn't ready

2:56

for summer to end. And

2:59

then you were going to tell me that fall is

3:01

about to end, and that I'm an entire season

3:03

behind.

3:05

I think maybe you're approaching two. Of

3:08

course, I don't really understand, because as far as I can

3:10

tell, Halloween was last week. It

3:13

sure feels like it.

3:15

Okay.

3:16

All right. The other thing that needed to

3:18

be talked about was the Nordic giveaway. Right.

3:22

We gave away the

3:25

Power Profiler Kit to me. Wait,

3:28

what?

3:28

Because I

3:29

wanted it. Because

3:31

I control the giveaways. I'm

3:34

pretty sure there's laws against this sort of thing.

3:37

Actually, Raoul won

3:40

the contest, but the contest will be going

3:42

on, and I'm sure I can win next time.

3:45

Okay. So the way the contest

3:47

is going to go on is we

3:49

will have a new drawing every month

3:51

for five more months. Right.

3:55

So including November. And

3:57

everybody who didn't win in the previous

3:59

round...

3:59

We'll roll

4:02

over into the

4:03

next round.

4:04

Yes, and the question is the same. What

4:07

features are you most interested in?

4:10

Although after

4:13

the third month, I think we're switching

4:15

to a different question. Okay, so we'll

4:17

probably clean the slate. Clean the slate after

4:19

two more tries. Okay. We'll

4:21

clean the slate and ask a new question. Which means you'll

4:23

have to re-enter. Right. Okay.

4:26

But until then, you don't have to re-enter. We'll let everybody know. If

4:29

you haven't entered, you feel free to enter for

4:31

this month and next month.

4:32

Right. Showitembedded.fm is

4:35

the best way to enter. Or

4:37

hit the contact link on embedded FM.

4:39

I'm almost ready to say it's the only

4:41

way to enter, but you know.

4:43

I mean, the entries through

4:46

Squarespace comments or YouTube comments.

4:48

Look, YouTube comments is pushing in. Okay,

4:52

but it's a little easier for us if you just hit

4:54

email and then I can put you all into a folder

4:56

and

4:58

then I shout through the house, Christopher, give me a number

5:01

between one and however many

5:03

people have entered. I don't really check the YouTube

5:05

comments all that often.

5:06

I do get emails. So it worked out

5:09

this time, but email is so much easier.

5:12

Yes. Okay. Well, thank you for entering who

5:14

everyone who entered and we will have more.

5:16

And there'll be an ad later in the show.

5:19

We'll do a separate ad. That's

5:21

right right now.

5:25

The show is sponsored by Nordic

5:28

Semiconductor this week. Nordic

5:30

is a market leader in IoT connectivity,

5:33

providing hardware, software, tools

5:36

and services to create the IoT

5:38

devices of the future. They

5:40

specialize in ultra low power

5:42

wireless communication with the wide

5:45

technology portfolios such as Bluetooth

5:47

Low Energy, Low Energy Audio,

5:50

Bluetooth Mesh, Thread, Matter,

5:53

Cellular IoT and Wi-Fi. They have

5:56

thousands of customers worldwide with

5:58

more than a 40% market share. and Bluetooth

6:00

low energy. Nearly 2 million

6:03

systems on a chip produced every

6:05

day. I can see that getting into

6:08

the Nordic developer platform can

6:10

be a little intimidating, but Nordic

6:12

has their Developer Academy, which

6:14

is an online learning platform that equips

6:16

developers to understand

6:19

how to build their IoT products with the Nordic

6:21

solutions. They also have the DevZone,

6:23

which brings together tech support and a community

6:26

of customers to provide assistance,

6:28

troubleshooting, and all kinds

6:30

of information on technical topics. Visit

6:33

nordicsemi.com for more information.

6:36

Visit academy.nordicsemi.com

6:39

and devzone.nordicsemi.com

6:41

for more information about their learning platforms.

6:47

Okay, so now because

6:50

people have given us money,

6:51

we need to find ways to get

6:53

rid of it.

6:55

Oh,

7:04

question your upbringing. Okay,

7:07

sure, sure. Okay, not,

7:10

you know, we can afford

7:13

to pay for JoJo. Now,

7:15

sure. Ways to get rid of it. Ways to get rid

7:17

of the

7:18

money.

7:21

Drop computer equipment on the ground. No,

7:23

no, no, I'm sorry. Okay,

7:26

so

7:27

I saw that somebody made an incredibly

7:30

beautiful,

7:31

colorful, adorable

7:35

ruler out of PCB with

7:38

red and green and white

7:42

and light green and copper.

7:45

Colored silkscreen, right? Is that how they do it?

7:47

Well, I think they only had two colors

7:49

of silkscreen, and they used the board

7:52

and the masking

7:53

as the other color.

7:55

All right.

7:56

Which was, I mean, it was really nice. And

7:58

so I was thinking, well, maybe we should do

7:59

one because we also got a request

8:02

to do a badge after

8:04

the Supercon badges,

8:08

which is weird because we don't go anywhere so

8:10

why would we need anybody to have badges? I don't

8:13

think they're real badges anymore. I

8:15

think once they're like a foot wide with

8:17

a TFT 15 inch display

8:20

and stereo speakers it ceases to be a badge,

8:23

it's more of a

8:24

plate armor.

8:28

And so between these two things

8:30

I have been contemplating

8:35

trying to make an embedded FM board. An

8:38

embedded FM board. I polled

8:41

the Patreon listener Slack group

8:45

and the highest priority,

8:48

and possibly I did influence this one

8:50

with my description, was that it

8:52

be pretty. And then

8:54

after that the next criteria was that it be a radio

8:57

or that it be used for learning embedded

8:59

and

8:59

those were tied as criteria. A radio.

9:03

And I looked up simple FM

9:07

radio circuits and that's pretty straightforward,

9:09

that's like ten components. I

9:13

looked up one of

9:15

the

9:17

STM32 disco

9:19

boards, which is one I use in my classroom,

9:23

and that was a lot of components. In

9:25

part because there's the whole programming subsystem.

9:28

I wondered about that as

9:30

a, at

9:31

least that would be useful, I don't want to really make

9:33

something that isn't

9:34

useful. I mean a radio isn't useful.

9:36

I know, that's the problem.

9:38

I mean we could make a board that has an FM

9:40

radio chip on it and you could

9:42

use that as one of the things to program with.

9:44

I don't know. Yeah,

9:46

I mean, I don't know. Just

9:50

another good board is kind of weird.

9:52

I mean if I could use it for teaching and

9:54

stuff it might be nice. But

9:57

the disco board I was looking at retails for 20 bucks.

10:00

And there's no way we'd be able to make it for

10:02

less than 40 because we wouldn't be making as many. And

10:05

once I made it pretty, it would be bigger.

10:08

It just... Ramify?

10:10

How do you

10:12

use that as a verb? I'm sure that it works

10:14

here.

10:15

The things...

10:17

The complexities ramify throughout? Sure.

10:19

Okay. One

10:22

decision ramifies into further

10:25

downstream decisions. It's a very

10:27

Neil Stevenson word. Yeah. Well,

10:31

did you actually go out and kind of

10:33

price a bomb out? No.

10:35

I used it... If it retails for 20 and

10:38

the bomb in 10KQ probably is five

10:40

then, which means that in

10:48

hundreds, it would probably be four

10:50

to five times that. We could mostly

10:53

break even on it if it was just

10:54

for fun.

10:56

But there's a lot of work involved. You're not

10:59

just going to make a... slap a board together and... Well,

11:01

and I'm not going to be the one to be able to put together

11:04

this board. I mean, even with schematics from

11:07

the disco boards available, if

11:09

I want to make it pretty, I'm going to go talk to somebody

11:11

who's done that a few times.

11:16

I started to be excited about

11:18

this and then I started

11:20

to think about another t-shirt

11:21

design. What about something smaller?

11:24

Like a ruler? I mean, I think the dev board idea

11:26

is more useful, but it's also why?

11:30

Way more work and why. So

11:35

I think we should continue to think about it, but maybe think

11:37

about something that's more show related,

11:39

maybe has some programmability, but it's like

11:42

something that could run circuit Python. That

11:45

would not be as

11:46

challenging, I think.

11:49

It has just a STM32

11:51

with a bunch of stuff on it.

11:54

Circuit Python with some pretty lights, some

11:56

fun stuff, the logo.

11:59

Put the FM radio on it and have it do

12:02

something with it. I don't know. Let's

12:05

put some more thought into it or have people have

12:07

suggestions of what they'd like to see. Could

12:09

it be a fun little artifact from the show with people like that stuff?

12:12

Yeah. Circuit

12:15

Python might be the way to go. Then

12:18

we don't have the programming subsystem, but we

12:20

do have USB. And

12:22

one of the things, I started talking to

12:25

Carrie at Alpenglow. Yeah. And

12:27

she pointed out that if

12:30

you're making these, you're

12:32

going to have to test them. And

12:34

they aren't going to be enough to make a testing

12:37

jig. Yeah. So we're going

12:39

to be programming these ourselves. And

12:41

now the amount of work starts to scale up. Certainly

12:44

not programming a bare pedal disco board

12:47

with anything

12:49

or testing it. So if you want to do circuit Python,

12:52

that's pretty extractable, but. Even

12:54

that, you still have to,

12:56

anyway,

12:57

it's a thought. A ruler

13:00

is still an option. That would be very simple.

13:02

It would still be pretty, but

13:04

kind of useless. But then the

13:07

Patreon folks did say that they wouldn't mind

13:09

another ruler.

13:10

The other option to take

13:12

some of the work away from us is to still

13:14

do a pretty board. Still have something that does

13:17

something, but to ship it as a kit.

13:20

So we'd have to build one to

13:22

make sure the kit works, maybe two, program

13:25

it. But otherwise it's

13:27

a soldering, make sure we

13:29

pick parts that are soldering and stuff, but make

13:31

a soldering kit

13:33

thing,

13:34

educational.

13:35

I mean, that wouldn't work for probably

13:37

your class stuff because you don't want to make students

13:39

actually build their own boards, but.

13:42

Those could be separate. That would be something

13:45

that could be, like you could take that later

13:47

and say, okay, send this to an assembly house for

13:49

class purposes. I

13:52

don't know. That would take some of the labor out

13:54

of it. And if

13:57

it doesn't work, then

13:59

we'd have to enter.

13:59

interact with people and say, okay, well, maybe you need

14:02

a replacement.

14:03

That's not the other thing.

14:04

Um,

14:06

but then there's also the idea of just doing another t-shirt

14:08

design, yeah.

14:11

Uh, which I would not do through teespring

14:13

again because they were not great. Yeah.

14:16

All right. So yeah. Um,

14:20

I don't want it to be like a teensy where

14:22

we are in the board manufacturing.

14:25

Not unless, not unless, uh,

14:27

for some weird reason, we, we,

14:30

we, we were suddenly super popular

14:32

and we could just outsource the whole thing and, and,

14:35

and, you know, live in the Bahamas

14:38

as board board mavens.

14:39

Of course, Sergio

14:42

asked, uh, had mentioned they

14:44

had been a while since they

14:46

last got an embedded FM

14:48

sticker, asked

14:49

if those were still a thing and should

14:52

we use generative AI to design the next

14:54

one? Let me get that first question out of the way.

14:57

They are still a thing. I meant to send some to super

14:59

con for people to give out. And

15:02

then I didn't because I just

15:04

lacked on it. Um, I have a box

15:07

of stickers and

15:08

I should send them out.

15:11

Um,

15:12

I guess send me

15:14

your mailing address

15:18

and I will make sticker Christmas

15:20

cards.

15:22

Maybe that's the way to go. Holiday

15:25

cards, holiday, winter cards,

15:27

new year's cards.

15:28

I mean, for the most part, I only

15:30

have new year's cards. So,

15:32

um, yeah.

15:35

Uh, the second part of your question should be

15:37

used generative AI for this. The

15:40

answer to your question is no. Go get a cup of

15:42

coffee. No, no, I'm done.

15:44

Oh,

15:45

okay. Answer was no. Oh, just that. Um,

15:49

I mean, I like our current design, so I don't know if we need a new

15:51

one.

15:52

Sure. Yeah.

15:53

But if we did the answer

15:55

would still be no.

15:57

Um, yeah,

15:58

this answer would be.

15:59

No. Unless

16:01

you want to

16:03

convince me.

16:04

Oh, so asked how do you make an ethical

16:07

choice of LLM slash

16:09

AI, not the ethics of

16:11

using LLM? Large

16:15

language model. Okay, right.

16:16

Not the ethics of using one. But

16:20

assuming

16:20

you are going to use one, are

16:23

there ethical

16:24

choices or ethical pathways

16:27

you should follow in order

16:29

to choose which one? I'm

16:32

not an expert on this, but I do know

16:35

some things. I

16:38

think, so primarily

16:40

what you're asking about is how

16:43

the... It was two parts, but the first major

16:45

part is what it has been trained on. I

16:49

think that's the major ethical issue just

16:51

for general use that people might

16:53

be concerned about apart

16:56

from some downstream things. And

16:59

that's a bit of a challenge. So

17:01

the big ones have been trained on the wide internet.

17:04

So I think from open AI, chat GPT

17:06

has been trained on whatever

17:08

is publicly available on the internet, whether it's copy written

17:10

or not. So Reddit, public

17:14

available images, stock images, news

17:16

images, people's personal images, Twitter,

17:19

whatever, all that stuff.

17:20

But also Google

17:22

has a

17:23

bazillion books. A bazillion

17:25

books, possibly. Many of which are copyright.

17:27

Possibly. Art work,

17:31

deviant art is a big, big independent

17:33

art site and I think they've all trolled through there.

17:36

So it's difficult.

17:40

So the big ones, I think Bard, his Google's,

17:43

I think that this is similarly trained on whatever

17:45

Google can see. Chat

17:47

GPT is obviously trained on a lot.

17:50

There are some smaller ones and I don't have them

17:52

on the top of my head that are

17:55

supposed to be ethically trained, ethically

17:57

trained, where they have... either

18:01

found things that are obviously public domain

18:03

to train on or gotten people's

18:06

permission for large sets.

18:09

I think Adobe has been trying

18:12

to do that. So I think the stuff that's built into

18:14

Photoshop,

18:16

I think Firefly is what it's called, quote

18:18

me on that. I think that

18:20

has been trained in things

18:22

that are licenseable so

18:25

that there are fewer copyright issues.

18:28

And I know there's a couple of smaller open source

18:32

kind of chat GPT things, large language models

18:34

that are claiming

18:36

to be trained on things that

18:38

are permissible. You're just going to have

18:40

to look and see

18:43

what details you have about the particular model that

18:45

you're looking at using. And it's different

18:48

between chat GPT and say the

18:51

image models, although those are starting to

18:53

cross cross pollinate.

18:55

So chat GPT and Dolly are now kind

18:57

of merged so you can do image stuff back and forth

18:59

with tech stuff and have conversations. So

19:02

that's

19:04

confusing as to what to

19:06

do there. So

19:09

it's not really a great scene. I think as time goes

19:11

on, more things that will be clear

19:13

about their training will

19:15

emerge. And it will be easier

19:18

to make a choice that you can feel better about on

19:21

the upstream side, on the training side. On the

19:23

downstream side, there's

19:26

still the issues about what are you displacing?

19:29

Are you really

19:31

solving a problem or are you playing with a fun toy? There

19:35

was something else about... So

19:40

there's a lot of difficult kind of

19:43

things that go into

19:45

into the choices right now. I

19:47

use some of this stuff. I've

19:49

been using OpenAI as a model called Whisper,

19:53

which is for transcription,

19:56

which has been... I've been experimenting with podcast.

20:00

there yet. Mostly because

20:03

it doesn't recognize different speakers, so it

20:05

just gives you this wall of text of everybody

20:07

speaking. And there are some apps that kind

20:09

of you can drop in all the separate

20:12

audio files and it'll transcribe them separately,

20:14

but it's not quite working right yet. That's

20:17

very useful. But it's separate than the transcripts

20:19

we pay for and are available. And

20:21

those are all clean and

20:24

speakers are identified. What

20:27

do you think about this question?

20:30

I

20:31

mean,

20:33

it's kind of fun because you get

20:35

pretty irritated anytime anybody

20:37

breaks this up. No,

20:40

I can talk reasonably about it. I get irritated

20:42

when

20:43

people

20:47

don't think about the usage

20:50

and are just evangelists for it.

20:53

I wanted to write a joke

20:55

with it. I think I wanted

20:58

to suggest that your

21:00

post to LinkedIn be in SUS

21:03

form or be in

21:05

the form of you were looking at

21:07

dating profile where you were looking

21:10

for a device to date, not

21:13

date, to work on.

21:16

And

21:18

these were jokes that I

21:20

just wanted to think about. I didn't want to fully

21:22

finish.

21:24

But

21:26

you've kind of convinced me that letting it write my

21:28

jokes is probably not that good for me

21:31

or for the world. And

21:34

part of that is because I saw

21:37

from three brown,

21:39

one blue or three blue, one brown. I

21:42

never remember which way that goes, but the YouTube

21:43

people with the math. The

21:49

drilling is really important

21:52

and that mathematicians

21:55

and scientists and physicists through

21:58

history, the people we really like. look

22:00

up to Newton and folks,

22:04

drills

22:05

throughout their lives.

22:08

They did problems. They did problems

22:10

they knew the answer to, and they

22:13

practiced.

22:16

And that practice ends

22:18

up building your intuition. When

22:20

we think about solving problems with

22:24

analytics and intuition, you don't get

22:26

that without all the drills.

22:29

Or if you do, you're super lucky. But

22:32

I need to see a problem a few times before

22:35

I can really ace it off

22:37

the bat. And I have

22:39

to admit, looking at students

22:42

in class, I

22:45

feel like I'm constantly being quizzed on

22:47

things. Oh yeah, yeah. But

22:50

I think that makes me better because I am constantly

22:52

being drilled on embedded

22:55

topics. Why would you do this versus

22:57

that? And why

22:59

did the bit masks work this way? And

23:03

spy versus I squared C, I am never going

23:05

to get those too confused. I know some people will, but

23:08

I will always know when to use it. I

23:10

squared C is the crappy one. That

23:13

is not true, but go ahead.

23:17

They have their purposes.

23:19

No, no, it's fine. I just never get it right the

23:21

first time. Whereas with spy, I almost

23:23

always get it right

23:25

the second time at least. And

23:28

so

23:29

between thinking about

23:32

drills

23:32

for math and physics problems,

23:35

and thinking about having an LLM

23:38

help me to write my jokes, I

23:43

kind of wonder if I do

23:45

fall into the habit of letting it do that if

23:48

I

23:49

get out of practice myself. I

23:53

think that's certainly something I haven't

23:56

necessarily. And it's something I enjoy.

23:59

Except when I, you know, kind of just half-assed

24:02

it like I did with, you should

24:04

do that in a Suse poem.

24:07

That's something that's getting on the edges of something that bothers

24:09

me about the LMs and probably would bother

24:11

me about any AI

24:14

that purports

24:16

to converse and think, or not

24:19

think, but you know, that we can have conversations

24:22

with, that we can treat as a sidekick and

24:25

as knowledge enough to do many, many

24:28

tasks. One

24:31

of the things is what you're saying, like getting out of practice

24:34

of mundane stuff. Now

24:37

plenty of technology gets rid of mundane stuff for

24:39

us. Calculator. Right, right.

24:42

And I'm not going to argue that necessarily being,

24:44

everyone needs to be super fast at

24:46

arithmetic

24:48

in a world with calculators. That's definitely

24:50

a benefit. But it sure helps for estimation. But

24:53

I wonder about communication and writing.

24:55

See, that's the thing. I

24:57

was hoping somebody would write a thank you letter and I'm like, okay,

25:00

chat GPT is the way I would go with

25:02

this.

25:04

But maybe that's because that's not a natural

25:06

skill that I

25:08

am good at.

25:10

Polite

25:11

talk.

25:12

Yeah, yeah. So one of the

25:14

things I said on Mastodon a couple of days ago that

25:18

I've been thinking about, and

25:20

it was triggered by somebody who was doing some writing and

25:22

they threw their writing at chat GPT to

25:24

have it proofread and check

25:26

on some things and stuff. I'll

25:29

just say what I wrote.

25:32

I have trouble articulating a lot of my qualms about large

25:34

language models and where they're headed. But

25:36

one of them is that it's going to be so easy to just

25:38

grab one and have it work with you on writing or

25:41

brainstorming or getting a second opinion on

25:43

something rather than reaching out to someone and having

25:45

a chat.

25:46

Especially for people

25:48

who the latter, reaching out to someone

25:51

takes a bit of activation energy and

25:53

to rude people.

25:55

Some people will probably say that's good, that

25:58

you have this thing to bounce out of. ideas

26:00

off of or check you at a moment's

26:02

notice. You're not wasting anybody else's time.

26:06

Okay, so first of all, I'm

26:08

interested in why you think that would

26:11

be a waste of somebody's time. If

26:13

somebody asked me to proofread something, I

26:16

would

26:17

say yes, pretty much immediately.

26:19

Wouldn't you? I have many things

26:21

for you. Do you understand what I'm saying, though?

26:24

I do. But

26:26

I mean, writing is a human process.

26:29

Communication is a human process. And

26:31

I

26:33

feel like if we're...

26:35

I

26:36

just worry

26:38

about things. I worry that

26:41

reducing opportunities for human

26:43

interaction and improvement of communication

26:46

and going, replacing

26:50

most of that with communicating with something that

26:52

is the average of the internet is

26:56

not a net benefit, because everything

26:59

will be dulled down. It's

27:01

like, you know, if you had Clippy, right,

27:04

and Clippy was real good, and you let it just, you

27:07

know, and people would react negatively that

27:09

because Clippy was bad. But how

27:13

much would you let Microsoft Word write for you?

27:16

Why? It was Microsoft Word, really?

27:18

I mean, for

27:20

cover letters? No. Are things

27:22

that I don't care about?

27:23

Fine. You know, I'm not

27:26

the writing police, but

27:28

certainly boilerplate legal kind of documents, fine,

27:31

whatever. My book? But no, those

27:33

have been templatized forever, right? You could you

27:35

could have gone before chat GPT existed to

27:37

no low press, or wherever

27:40

and found to give me a template for a business letter

27:42

complaining about this. That's fine. But

27:45

for for a piece of

27:47

professional writing, or a piece of fiction,

27:50

or your book, you don't think your

27:52

book would be worsened by running

27:55

it through this thing? Yeah.

27:59

Yes, it would be because I mean

28:03

from the sidebar on how

28:05

we still use modems to

28:08

the detailed factual information

28:11

that has been checked by experts, no,

28:14

you're not going to get that from an LLM. And

28:16

then to take what you were saying,

28:20

now you're out of practice, now people are out of practice of proofreading,

28:24

which means they're going to be worse writers

28:26

because one of the ways to learn to write

28:29

is to read other people's writing and

28:31

find out what's wrong with it because it's

28:33

hard to do that for your own work. And

28:38

you know, people outsourcing other stuff are going

28:40

to get out of practice of the mundane writing but

28:42

the mundane writing is what keeps you sharp. There's

28:44

a lot of emails we write and stuff. I'm getting

28:47

emails I'm sure written by a chat GPT now

28:49

because they just have that

28:51

kind of fragrance.

28:53

Well, but those are spam

28:55

and junk emails. Some

28:57

of them, but some of them I think are on LinkedIn. LinkedIn

29:00

now I just saw while I was there has a button you can

29:02

press to do all of your outreach

29:04

stuff using an LLM. It'll

29:06

write all your... Yes, but those totally feel

29:09

like this person hasn't

29:10

bothered to read my profile. For now. Anyway,

29:13

I don't want to go on too long about this because

29:15

I get off track and

29:19

people know how I feel. I

29:21

mean, having said all this, a

29:24

few people seem to think I'm anti-AI. First

29:28

of all, I hate the term AI because I think it's completely

29:30

white watching all of this.

29:34

I prefer the term ML because

29:37

that covers a lot more things

29:39

that aren't necessarily neural networks. It covers

29:41

types of neural networks that

29:43

do

29:44

very useful things in limited scoped

29:47

ways. Anyway, I do not hate

29:49

AI and ML. I, in fact, have

29:51

one and possibly two clients where

29:54

I'm working on ML stuff. I

29:56

am familiar with it. I know how it works. My

29:59

opinion...

29:59

are probably verging on

30:02

philosophical at this point and not technical.

30:04

Okay, I have some lightning round questions. Are

30:06

you ready?

30:08

That's a little late in the show for that, but okay.

30:10

Simon, finish one gig or start

30:12

a dozen?

30:14

At this rate, I'm starting a dozen.

30:18

I currently have four

30:20

open

30:21

tasks

30:23

between teaching class, finishing the book,

30:26

and two clients. I

30:29

have a lot.

30:30

I'm going from

30:31

zero or maybe 0.25 clients

30:33

to possibly three. Ah,

30:36

Speck would like to know pancakes or waffles?

30:39

Ooh. Um, well,

30:41

wait, wait, which kind of waffles?

30:44

I believe you can choose your own.

30:46

Alright.

30:47

I'm gonna admit something here. I eat

30:49

a lot of toaster waffles. Mm-hmm.

30:52

It doesn't necessarily mean I like them better than pancakes, but they're

30:54

easier to make. Probably

30:57

pancakes. I have a intense

31:00

sense of deja vu. Did we not have

31:02

a very long discussion about waffles and pancakes

31:04

on this very show at some point in the last five years?

31:07

I don't know. Somebody

31:08

find that out. Probably pancakes.

31:11

They're delicious for three bites. And

31:13

then suddenly, for me at least, it feels

31:15

like I've eaten that spray foam

31:18

that you used to seal your house with. And

31:21

then it comes out real small, but then it expands to 18 times

31:23

its size. So I have trouble with pancakes from

31:25

a mass to enjoyment ratio standpoint. And

31:32

your toaster waffles don't expand

31:34

at all because they have negative calories.

31:37

Oh, they cut calories? They don't have very

31:39

many calories. They're like,

31:41

you know, normal. But if

31:44

I... Okay, Belgian waffle.

31:46

Final answer.

31:49

I would say pancakes,

31:52

whether you are talking about the food or

31:54

the style of holding hands. The

31:56

what? Moving on. Excuse

31:59

me?

32:00

The style of holding hands?

32:02

Yes. If you hold hands with somebody and

32:04

you hold hands and your palms are together and your

32:06

fingers, your fingers touch

32:08

your fingers, that's pancake-hand

32:11

style. If you do waffles, you interlace

32:14

them. And that's

32:16

waffle style of holding hands. I've

32:19

never

32:19

in my life heard this. And

32:21

waffle style is okay

32:23

unless your partner's hands are much bigger than

32:25

yours, in which case waffle styles

32:28

can be uncomfortable. But pancake

32:30

is worse if you're sweaty-handed.

32:32

I'm learning so much from

32:34

this podcast.

32:36

I really think, you know, people, we're

32:38

really helping people. Okay,

32:40

one more silly one and then we'll actually

32:42

do some embedded stuff.

32:44

Butter. Salted or not?

32:46

I mean,

32:47

for

32:48

just eating, salted. What

32:50

are you... Straight out of the carton. For

32:54

baking, unsalted... Didn't

32:57

we... I saw

32:59

Paul Hawley would just answer this question. For

33:02

unsalted, if you're baking, you can

33:04

control the amount of salt in the recipe if you use

33:06

unsalted. Otherwise, you don't know how much salt you're adding.

33:10

Okay, as the person who bakes, we

33:12

always use salted because it lasts so much longer.

33:15

A little bit of that, too.

33:16

And then you just taste to figure out how

33:18

much salt you need to add. But

33:21

I know that that's not the correct

33:23

way of baking, but it's so much easier.

33:26

Yeah, well, I'm not going to taste raw cake batter. I

33:28

know, and it's much to your detriment. No,

33:30

it isn't.

33:31

Okay, I promised that. It

33:34

is.

33:36

If you could only use one microcontroller for

33:39

the rest of your life, Pedro would like to know which one and

33:41

why.

33:44

The basic stamp.

33:47

That's not true. No,

33:49

I just thought I'd be contrary. One

33:52

microcontroller for the rest of my life.

33:55

How long am I going to live? Let's

33:59

get the parameters. Let's say at least

34:01

another 10 years. All right.

34:03

I mean, you know,

34:06

nobody ever got

34:08

fired for using an STM32 F4 something.

34:15

Yeah, I have an STM32 F4 and an L4. I

34:20

know those are kind of different,

34:22

but they're not that different.

34:23

I mean, you can do an awful lot with one of those. And

34:27

the ecosystem is there and it's not going to go away. Yeah.

34:31

Well documented and everybody's used it. I mean, that's

34:33

kind of the boring

34:35

mainstream. And

34:38

I've seen some M55s and

34:40

some M85s coming out, but

34:43

I'm not very familiar with the specs on those

34:45

other than they go faster.

34:48

Paul, now I'm thinking, well, I don't know. I

34:52

don't know. The problem

34:54

is I don't have any particular, everybody

34:56

wants me to use something different when I show up

34:59

at a client. So I don't really form

35:01

attachments. I know the ones that I enjoy

35:03

using.

35:04

Yeah, you know, I haven't

35:06

said that. I'm like,

35:08

but you have the Nordic thingy 52 on

35:10

your desk and you're playing with it and having a good time.

35:12

And you don't even remember what chip

35:15

is inside there half the time. And

35:19

then, so, okay. So yeah, I'm going to, I'm going to, I'm going to

35:21

switch it around. I'm going to say the easiest thing

35:23

to use

35:24

for a wide variety of applications. And that's

35:27

probably not an SDM because you

35:29

have to have a lot of ancillary stuff with a lot of those. So

35:33

something Nordic that's got Wi-Fi and

35:35

Bluetooth.

35:36

It would depend on what I wanted to work

35:38

on. I mean, that's the thing.

35:41

I mean, if I have to choose, okay, if I

35:44

have to choose one microcontroller to use for the rest of my life,

35:46

something's gone wrong.

35:48

And therefore I don't think I want to work on embedded

35:50

systems anymore. I'm probably digging a moat.

35:53

Okay. Yes.

35:55

Yeah. And

35:57

I mean, I'm not going to do

35:59

that. I don't know

36:01

that I would use...

36:04

If I think about projects I would do,

36:08

I wouldn't use either one of those

36:10

to do a big robotics project.

36:13

I would actually use something Linux-based.

36:16

If I had to do a small robotics

36:19

project with lots of motor control, I

36:22

would go to the TI Piccolo line.

36:25

If I had to do Bluetooth, I would use Nordic

36:28

unless somebody really made me use TI. Not

36:31

that TI is bad, just that Nordic

36:33

is a lot easier. That includes

36:36

Bluetooth, Thread, Matter. Well,

36:38

you don't have to do... Zigbee,

36:41

whatever. They're difficult.

36:46

Okay, split the difference.

36:48

A bag of 80-51s, final answer. Thank

36:52

you. I'm sure that is your

36:54

final answer. Is

36:57

there a contract or job that you didn't

36:59

take that you wish you had and why?

37:00

Oh, gosh. I kind of

37:02

wish I had

37:04

gotten hired by Fitbit when there

37:06

were only a couple of engineers there,

37:10

before Haiko was the manager.

37:13

Not that that would have been bad, just that

37:16

I could have been hired then when there were only

37:18

a couple of engineers and then I would have

37:20

gotten lots of stock options. Oh, I see. I

37:26

don't think there's any jobs I gave up that

37:28

were financial screw-ups.

37:30

There's

37:32

definitely jobs I took that were financial

37:35

screw-ups. Let

37:38

me think. There were too many jobs

37:40

that I didn't end up taking contracts.

37:44

No, every contract I've turned down

37:47

has been the

37:48

correct thing to do in hindsight.

37:51

I mean, they usually go off and you never

37:53

hear from them again. Right,

37:55

but sometimes I check back or

37:57

hear things from people who's too...

37:59

was

38:03

not a formal group of people, but I knew people

38:05

who also were contracting at those places

38:08

and heard later about things that happened.

38:11

No, I've been pretty lucky or

38:14

good about choosing, so there's no ones that

38:16

got away, I don't think. No.

38:20

Very

38:21

good. Certainly there have been

38:23

things that might have been lucrative that

38:26

people

38:27

had expressed interest

38:29

but never got to the point of me interviewing

38:31

or anything because I just didn't want to do it.

38:34

There's a lot that I nipped in the bud before

38:36

I even got to the stage of interviews

38:38

or anything. There's

38:39

a lot of things you don't want to work

38:40

on because you have the luxury

38:43

of not having to work on them.

38:46

Okay, let's see. Next question was

38:48

from AmbadatEmbedded,

38:52

who asked the

38:54

favorite jack-of-all-trades micro

38:56

or a favorite specific

38:58

task micro? I feel like we've kind of covered that. Yeah.

39:02

There's a follow-up

39:04

question from the same person. Is the Arduino

39:06

Uno now

39:07

obsolete? Oh my God. It

39:10

was obsolete when it came out. It depends

39:12

on your

39:14

definition of obsolete. If you've got one, it

39:16

can do things. There's

39:17

plenty it can do still.

39:20

But even when it came out, it was quite

39:22

expensive for what it could do. Yeah,

39:25

it's a thing. It's a really old product.

39:28

I wouldn't. There's a lot you can get for

39:30

the same or less price now

39:33

that can do a lot more. Do they still

39:35

sell them? Oh yeah, they still sell them.

39:37

Okay. But on the other hand, they

39:39

are the easiest, lowest

39:42

common denominator. Almost

39:44

everyone can get a compiler to run on their system.

39:47

That's the thing about this stuff. That's why I'm

39:49

saying it's hard to define obsolete because

39:52

computers get obsolete because of the

39:55

software stops being supported. So

39:58

a laptop from 2000.

39:59

2005 may not even be able to surf the web

40:02

anymore because it can't handle You

40:05

know doesn't have an upper you can't upgrade upgraded

40:07

to an operating system that can even do a T

40:09

GPS or something, right? That's

40:11

not the case of embedded stuff. Like that's why

40:14

80-51s are still

40:16

still being used in in in

40:18

modern things right or or derivatives

40:21

of 80-51s on FPGAs or

40:23

or things so I don't

40:26

think it's obsolete as long as it

40:28

does what you want it to

40:31

But again, I wouldn't there's other things I would pay

40:33

more for unless you had a really good reason pay There's

40:36

other things I would For

40:38

equivalent price would buy instead of an you

40:40

know

40:42

For hobbyist stuff before

40:44

I'd return it out. So I get Python would

40:46

be much higher I'm a Python or you know

40:50

And that doesn't even dictate

40:52

what border is right? You could have a Raspberry Pi 2040 or You

40:55

know, a

40:56

fruit has a circuit python

41:00

Little itty-bitty chip on

41:02

a ruler that cracks

41:03

me up. I have one on my desk

41:05

Yeah, but if

41:07

it works then you've got them so

41:10

talking about historical I went and

41:13

somebody linked in linked

41:16

to The bite

41:19

magazine back issues. Oh neat And

41:22

I looked at the first one from 1975 and went through a couple

41:24

a couple

41:27

of them Kind of thinking it

41:29

would be fun to do a live.

41:32

I Didn't tweet live

41:34

mastodonting. That doesn't sound right

41:36

now

41:39

But reading

41:41

those in commenting, you know trying to read a

41:43

magazine a day or a week and and

41:46

commenting And it

41:48

was it was kind of a museum I did

41:50

it a little bit

41:50

for paid

41:51

for the patreon slot

41:54

group

41:55

and Just you know Kind

41:58

of making fun above the idea of

42:02

needing to repurpose keyboards

42:05

because they weren't standard yet, and so you'd

42:07

have to change them around. But

42:11

then there were other articles that were

42:13

like, how to write

42:15

your own assembly with the idea of how to understand

42:18

assembly as you write it, like what is the piece?

42:20

That article could be published today and would be just

42:23

as useful.

42:23

That's because

42:25

embedded is still quite- In 1975. Still

42:28

somewhat backwards.

42:28

But now we have new keyboards, so it's okay. Yeah,

42:31

and you don't even have a big clacky one. No.

42:35

That's what all the coffee achiever

42:37

people type on. So you can hear them from 10

42:39

miles away. Type,

42:42

mechanical keyboard people, you're fine, sorry.

42:45

I'm not really mad at you. It's

42:48

just that he had to share an office with me for so

42:51

long that he knows that if anything makes noise,

42:54

it's my opportunity to turn around and glare

42:56

at it. Yeah, I had to type on a pillow. A

43:00

little letters that I sharpied on. Keyboard

43:04

is inside there. Mostly lined

43:06

up. Okay,

43:11

so let's see what else we have next. I

43:14

recall an episode where you talked to you

43:16

about installing an antenna to talk to your

43:19

dad. How is that going? You know what?

43:21

It's going. So we

43:24

always knew it was gonna take us a while

43:26

because we both had to learn Morse code and

43:29

we both had to get radio stuff and learn how that works. So

43:32

I have gotten Morse code to the point where

43:34

I'm bad at it

43:37

at about 10 words per minute, which is better

43:40

than when I started, which was I wasn't able to do anything.

43:43

So I can send okay. Receiving

43:48

is still challenging, so I'm

43:50

working on that. I think my dad has

43:52

gotten somewhat more proficient than

43:54

me. He

43:56

has a radio and he's been testing it. He's

44:01

set up his antenna and he's receiving stuff. I

44:04

have a radio kit. He has the same

44:06

radio I do, but he got it assembled. I got the kit. And

44:10

I started it this week, and I have

44:12

wound a grand total of one toroid.

44:16

How many do you have to wind? I think there's three

44:18

or four, but this is a hard one because it's

44:20

got multiple winds.

44:22

So

44:24

it's

44:25

got 38 winds with two ends,

44:27

and then five, and then five.

44:30

So it's a little teensy. Like

44:32

it's the size of a dime, so it's kind of fiddly

44:34

with all the magnum. I love nodding. Anyway, what?

44:37

You know what teensy means.

44:38

No, no, it's not the teensy part that I'm writing.

44:40

It's the winding and the fives and the 38. Five

44:43

winds. It's

44:45

just winding. You count them. That's

44:47

a lot of wind. Where's the cheese? Anyway,

44:50

so that's going to take me at least a week or two to

44:52

finish, and then yeah,

44:54

I'm going to set it up and

44:56

see if I can hear anything, and we'll

44:58

proceed from there. But anyway, I don't

45:00

know.

45:01

You have the wire for the antenna, but

45:03

we haven't set it up yet.

45:05

No, I have the

45:07

antenna, and I'm going to probably put it in the front

45:09

yard.

45:10

Oh, I thought,

45:11

oh, good. That's better than my idea of getting

45:13

a crossbow and firing it

45:17

into one of the taller trees. You know

45:19

my track record with

45:22

a projectile weapon. I was going to get to fire

45:24

the crossbow. Projectile weapons

45:26

and technology and trees. The

45:29

last time I

45:31

used a projectile weapon in the hopes of doing

45:33

something technology related, I ended up

45:36

in the ER.

45:39

Yeah, so yeah, the front yard,

45:41

because he's east.

45:43

So I need to have it, and if

45:45

I put it back here to the house, I'd have to get it pretty

45:47

high up to clear the house. So

45:50

if I have to do it in the front yard, I don't have to

45:53

clear the house. So it'll be oriented this way.

45:56

It radiates.

45:57

We should put it along the path where the squirrels are.

45:59

Maybe we can talk. to them.

46:01

Yes, I'm certain

46:03

they know Morse code too. We can build a

46:05

squirrel army. So

46:06

yeah, the project is going.

46:09

I'm hoping in the next month or two to try

46:11

it out. The nice thing about, you know, trying to talk

46:13

to someone that

46:14

you have an out of band communication with is we

46:16

can kind of pick a frequency while

46:20

texting to each other or you're on the phone and

46:22

say, okay, did you hear my CQ?

46:27

So we'll know if it's working or not. Just

46:29

kind of

46:31

randomly try at random times and see if anybody's

46:33

listening. So yeah, the hard part

46:35

about Morse, part

46:38

of it more as I've talked about before is

46:41

I thought I got to the point where I could understand the letters, but once

46:43

real words were coming past, it was harder to,

46:47

harder to copy what was being,

46:48

being said and the

46:51

patterns of letters.

46:53

When it's random, they tend not to have

46:56

a lot of the quick letters

46:59

together as often as real

47:01

words with, you know, because

47:03

the EIEIO sounds a lot like

47:05

very fast. It's

47:08

yeah. He, no,

47:10

no Q. Yeah,

47:13

exactly. And common in English,

47:16

common English words have the common letters frequently,

47:19

right? And the common frequent letters are very

47:21

short in Morse code for obvious reasons, but

47:25

if you're not trying to hear them go by real

47:27

quick and get lost. So

47:29

there's that problem. And then there's all these, I

47:32

almost said something that we get ham radio people

47:34

mad at me, but I can't

47:37

hear them because I don't have a radio. There's all these abbreviations

47:39

for stuff. Well, there's that. I

47:47

mean, there's the standard ham abbreviations of the Q

47:49

code.

47:50

Those are all, there's a few of those and most of

47:52

them are fine, but there's all these

47:56

nevermind. Go on. I don't know. I'm just randomly

47:59

now putting letters together.

48:02

There's a mode in the app I'm using to learn,

48:05

which is called QSO Bot.

48:07

It's

48:09

a bot

48:11

where you Morse code at it, and it

48:14

Morse codes back at you in kind of a simulation

48:17

of a normal code conversation. I

48:21

got one back from it, and it's

48:23

just all random letters. I'm like, this

48:25

is supposed to be English. In

48:27

the app, it's a little glossary, it turns out, of

48:30

common Morse

48:32

code abbreviations, of just normal words. They're

48:35

not like specific ham stuff. It

48:38

was like, r for some word, fr

48:41

for from, and

48:43

so all these common words are just shrunk to ...

48:50

I don't know any of them. I'm sitting here copying this

48:52

going, what is this thing saying to me? It's

48:54

like, oh, how's the weather out there? It

48:57

was like, h, wx, rr. I

49:04

don't think we have to use that with my dad, but

49:07

if I am going to talk to anybody else, I

49:09

don't really know how to prepare for that. It's

49:12

hard enough to learn Morse, so I don't want to learn somebody's-

49:14

Language. ... genography. I

49:17

don't know how often that stuff is actually used, probably pretty commonly,

49:20

because Morse is extremely slow.

49:23

Obviously, abbreviating things makes sense,

49:25

but I don't know any of those. I

49:28

might just have to print out a thing and

49:30

make postures.

49:33

Thank you for asking. It is continuing

49:36

slowly, as

49:37

most of my projects do.

49:42

Let's see.

49:43

I should mention the book, because

49:45

I told O'Reilly that I had

49:48

a podcast with listeners who

49:50

might be interested in the book, and

49:53

it should be out in March. I

49:56

have finished the first draft, and-

50:00

So, you know, hear me whining about how

50:02

hard it is to write about motors anymore. Now

50:05

I can whine about my tech reviewers, which

50:08

I won't do because they've all been super

50:10

nice and very helpful and have found

50:12

things and

50:14

have often said, maybe you should show them

50:16

work here instead of just jumping from

50:19

here's what it is, here's

50:21

the equation, here's the answer, like actually show

50:23

it out. So thank you to the

50:25

tech reviewers. When

50:28

the book comes out, we will be

50:30

having some giveaways. There will

50:33

be some discounts for

50:34

patrons

50:37

or newsletter subscribers

50:38

or people who ask for them. I

50:40

don't know. But that's all in the

50:43

March timeframe.

50:44

It's still on early release

50:46

on the O'Reilly learning system. I

50:49

can get a 30 day subscription

50:52

for that if for anybody who wants it. But

50:56

again, it'd be better if you're actually interested

50:58

in my book instead of just the O'Reilly learning system,

51:01

you might want to wait until like March when

51:03

it goes from having diagrams

51:06

that I have scribbled

51:06

out to professional

51:09

diagrams. No, scribble is the cool part.

51:13

There is the one that says, please make this monster

51:15

look scary. And the monster

51:17

I drew is just not scary. It's

51:19

not even attempting to be scared.

51:22

So yeah,

51:25

that's closer to done. And

51:30

the tech review stage is hard

51:32

because I know that

51:35

these people are helping me and they're

51:37

really nice people. And I

51:39

mean, they're going above and beyond. And

51:42

there's still always the temptation

51:44

to argue.

51:45

You know, like,

51:47

I totally disagree with you because I wrote it that way

51:49

for a reason. And that's not really true. I

51:51

just wrote it that way because I didn't think of what they were saying. So

51:54

it's like all ego in

51:56

code reviews. You kind of have to get

51:58

around it. realize that it

52:02

doesn't, they're not attacking

52:04

me. They're trying to help me make it better. And

52:07

they are succeeding when I

52:09

let them.

52:11

And then

52:12

class part is doing their asynchronous

52:15

cohort, which

52:17

has turned out to be kind

52:19

of weird

52:22

and not much less work

52:24

than a regular cohort. So

52:27

that part wasn't successful.

52:29

I don't know

52:32

what's going to happen there. I don't know if we're going to do another

52:34

one. Stay

52:35

tuned, I guess. Stay

52:36

tuned, I guess, but don't hold your breath. For

52:40

the book, I am starting to look for conferences

52:42

to present at. To

52:45

do, of course, not

52:47

very many in person or

52:50

mostly local to the Bay Area if it is in person.

52:53

And

52:54

being paid is really nice and

52:56

keynote positions are really spicy and

52:59

all that. But I am working on

53:01

a couple of pretty cool topics. Should

53:05

go on some podcasts.

53:07

I should actually, actually.

53:09

I

53:09

hear people like podcasts.

53:11

Did you do that on purpose? Of course I did.

53:15

Do you know what podcast I'm on?

53:16

Yeah, the building something,

53:18

the builders, builders quadrangle.

53:21

The builders circle. I was

53:24

close. The critique

53:26

with Sarah. I was on. It's a

53:28

podcast devoted to people with

53:30

startups and how

53:34

to make their startups work. And

53:37

she's a mechanical engineer and she's talked

53:39

to a lot of electrical engineers, but

53:42

she didn't have a lot of firmware experience

53:44

or really understand the whole process.

53:47

So it's like I talked for an

53:49

hour, hour and a half solid very

53:51

quickly in order

53:54

to download everything.

53:57

So, yes, if you would like to hear

53:59

more of me talking. about embedded

54:01

systems, which nominally is what this show

54:03

is supposed to be about. Oh, we didn't talk about any of

54:05

it today. Please check out some

54:07

other show. Please check out the Builder Circle.

54:11

And there'll be a link in the

54:12

show.

54:14

The episode doesn't come out until the

54:16

28th.

54:17

Yeah, I think that's right.

54:19

What else is going on? There's

54:22

other stuff. There must

54:24

be other stuff.

54:27

Does there have to be other stuff?

54:29

Yeah, there should be other stuff.

54:31

Because

54:31

then if there is another stuff, I have to go do some

54:33

work. Oh, I see. Or

54:36

exercise. Actually, given

54:38

the time, we both need to exercise. Okay.

54:41

Well...

54:43

And since we got a dog who is actually

54:45

a lump of coal,

54:47

we can't

54:48

take her for a walk because she's too afraid.

54:51

Instead, we will have to

54:54

exercise boring ways. I'm taking

54:56

a picture of the dog in the podcast down here.

54:59

I see. All right. Well, I'm

55:02

sorry to everyone.

55:04

Thank you to Christopher for

55:06

producing and co-hosting.

55:07

What did you call me?

55:09

Thank you to Christopher. Oh,

55:11

okay. I heard Christopher. Thank

55:13

you to Christopher for producing

55:15

and co-hosting. Thank you to

55:18

Nordic for their sponsorship. Thank

55:20

you to our Patreon listener Slack group for

55:23

their support and for

55:25

their questions. You

55:28

can enter the contest or

55:30

talk to us at showatembedded.fm or

55:33

hit the contact link on Embedded FM. You

55:36

can do a comment, but we don't respond as

55:38

quickly. We might

55:40

miss them. We might miss them. Especially on YouTube.

55:43

If you need pictures of our dog, let

55:46

us know. If

55:48

you would like to imagine what a picture is, all

55:50

you have to do is go outside, look at

55:52

the sun for about one second with just

55:54

one eye, and then wherever you

55:56

look after that, you will see our dog. It's

55:59

kind of a fun thing to do. like avoid in space.

56:01

Okay.

56:06

And now for some Winnie the Pooh.

56:10

Let's see. Let's see. We had

56:13

the nothing poo bear,

56:16

nothing. We can't all and some of us don't. That's

56:18

all there is to it. And

56:20

then pooh

56:22

hummed and sang Coddlestone Pie.

56:25

All right, here we are. That's

56:28

right, said your thing.

56:31

Umpty tiddly empty too.

56:34

Here we go gathering nuts and

56:36

may enjoy yourself.

56:38

I am said

56:40

pooh. Some can

56:42

said your boy.

56:45

What's the matter?

56:47

Is

56:48

anything

56:49

the matter?

56:51

You seems so sad.

56:53

Eeyore.

56:55

Sad?

56:56

Why should I be sad? It's my

56:59

birthday. The happiest

57:01

day of the year.

57:04

Your birthday said pooh and

57:06

great surprise.

57:08

Of course it is. Can't you see?

57:11

Look at all the presents I have had. He

57:14

waved a foot from side to side.

57:16

Look at the birthday cake,

57:18

candles and pink sugar.

57:21

Pooh looked first to the right and

57:24

then to the left. Presents said

57:26

pooh. Birthday

57:27

cake said pooh.

57:29

Where?

57:31

Can't you see them?

57:33

No said pooh. Neither

57:36

can I said Eeyore.

57:39

Joke

57:40

he explained. Ha ha. Pooh

57:44

scratched his head being a little puzzled by

57:46

this all.

57:47

But is it really your birthday

57:50

he asked? It is.

57:52

Oh well many happy returns

57:54

of the day Eeyore and

57:57

many happy returns to you.

57:59

Poo, Bear.

58:01

But it isn't my birthday. No,

58:05

it's mine. But you

58:07

said many happy returns.

58:09

Well,

58:11

why not?

58:12

You don't always want to be miserable on

58:14

my birthday, do you?

58:17

Oh, I see, said Poo.

58:20

It's bad enough,

58:22

said Eeyore, almost breaking down. Being

58:25

miserable myself, what with no presents

58:28

and no cake and no candles and no proper

58:31

notices. Me at all,

58:33

but if everyone is going to be miserable too.

58:36

This was too much for Poo. Stay

58:39

there, he's called to Eeyore as he turned

58:41

and hurried back home. As quickly as

58:43

he could for he felt he must get Eeyore

58:45

present of some sort at once and

58:47

he could always think of the proper one after.