0:00

Hello and welcome to Python Bytes, where we

0:02

deliver Python news and headlines directly to your

0:04

earbuds. This is episode 362, recorded November 28th,

0:06

2023. I'm

0:11

Michael Kennedy. And I'm Brian Ocken. And this episode

0:13

is brought to you by Scout APM. We will

0:15

tell you more about them later, but please check

0:18

out what they're offering. It really helps support

0:20

the show. And if you want to connect with

0:22

us on the socials, the place we're all

0:24

hanging out at is mastodon,

0:26

fostodon.org specifically. I'm Adam

0:29

Kennedy. Brian is at Brian Ocken.

0:31

And the show is at Python Bytes. fostodon.org

0:33

is invite only. Of course, you can follow

0:35

us from any place on Mastodon. But if

0:37

you want to be part of it, just

0:39

shoot one of us a message. We can

0:41

send you an invite. Yep. Help you get over there to

0:43

hang out a little bit closer to home for us. And

0:46

finally, if you want to be part of the live

0:48

stream and see how the show is made, see

0:50

what we're doing, even the real mistakes

0:52

that we might edit out, check that

0:55

out PythonBytes.fm live, usually Tuesdays at 11pm.

0:58

And older videos are there as well. Let's kick

1:00

it off. What do you got? This is a gentle

1:03

ease back into the show. Over

1:06

the Thanksgiving break, I was thinking a lot about

1:08

kind of what I wanted to do for the

1:10

next year, starting to, you know, starting to think

1:13

about plans for how to, where

1:15

do I want to focus my time? And

1:17

I ran across this great post by Vadim.

1:22

Let's just say Vadim. Habits

1:25

of great software engineers.

1:27

And it's kind of,

1:29

there's a lot of great stuff

1:31

here. So also, nice index on

1:34

the right, which is cool. So

1:36

focusing beyond the code. So I'm not going to

1:39

go through all these. I'm going to go through,

1:41

there's like 12 or 13 listed, but

1:43

I'll go through a few of them. Focus

1:45

on the code. So I

1:47

don't know if I buy this, but he says, as a developer, you

1:49

code 20% of your time. You

1:52

should excel at coding nonetheless, but it won't be

1:54

enough to be great. I think I code more

1:56

than 20% of the time. I

1:58

would say too. Well, define. code, right?

2:00

Okay, so is coding, is that reading

2:02

code? Is that thinking about code? Is

2:05

that... Oh, the design part. Right. Or

2:07

is it actually I'm hitting keys on

2:09

keyboard? You know what I mean? Yeah,

2:11

that's true. Right? Because people say code

2:13

is read way more than it's written

2:16

and stuff, but I think for me, coding

2:18

is like the act of being in the

2:20

code and adding features or

2:22

evolving it, even if like I'm not keystroking

2:24

it at that moment. So, unclear,

2:27

but if that encompasses all,

2:29

then 20% is way too low agreed.

2:31

Okay. Well, I guess the point of

2:33

this, regardless of the percentage, is focusing

2:35

on the other parts of your job

2:37

also and focusing on,

2:40

you know, not just being a better software

2:42

engineer, but also... Well, being a better

2:44

software engineer encompasses more than just coding

2:47

well and knowing like knowing Python details

2:49

and things like that, like communication skills

2:51

and all that, design skills,

2:53

learning other, learning to read other people's code,

2:55

things like... Helping unblock your teammates who are

2:58

having a challenge. Instead of writing code yourself,

3:00

like I'm gonna help them get that problem

3:02

fixed or something, right? Like... Yeah. Yep.

3:05

All of a sudden. Well, and okay.

3:07

So, number two was efficiency and anti-fragility.

3:10

Not really a fan of the

3:12

anti-fragile word. I don't even really

3:14

know what it means, but I

3:16

guess... You get originally comes from Nassim

3:19

Talib and that whole philosophy, although I haven't

3:21

read a whole lot of that either. Anyway.

3:25

It's kind of about things that get better

3:27

the more they kind of get abused, right?

3:29

Like they sort of build up a resilience.

3:31

Well... I think it originally comes out of

3:34

like economics. Okay. Well, the next one is

3:36

one I really agree

3:38

with is the joy... Number three is the

3:40

joy of tinkering. Build projects, try out frameworks,

3:42

build stuff on the side, keeps

3:45

the spark alive. And

3:47

this is one I couldn't agree more.

3:50

It doesn't... For a while I was

3:52

often thinking about like what could I

3:54

build that I could actually release or

3:56

that I'd be proud of. That's great.

3:59

You could try... to at least do a side project

4:01

that tries to make money or something. But

4:04

if you just want to try out FastAPI,

4:06

for instance, just build something with FastAPI. Just

4:08

come up with a toy project and build

4:10

it and learn it. And it's fun. Even

4:12

if you never apply it to other stuff,

4:14

I've had so many times in my career

4:16

where I've had something that I tinkered with

4:19

just as a toy that suddenly I could

4:21

use. So I'm like, oh,

4:23

yeah, I've built wikis before and stuff like that.

4:26

Go ahead and do that for a

4:28

project. Or I've done

4:30

a flask gap on the side. So if I need

4:32

a flask gap at work, it's going to be pretty

4:34

easy. Things like that. Play

4:37

with toy projects. I say put that on top of

4:39

your list for 2024. More toy projects. I'll

4:43

go through the rest of these pretty

4:45

quick. Number four is knowing the why. Number

4:47

five, thinking and systems. Number

4:49

six, tech detox.

4:51

Recharging away for the

4:54

monitor. This makes you a better programmer.

4:56

Couldn't agree more. The

4:58

art of approximation. I didn't really know what this is

5:00

about. I didn't read this part.

5:02

Knowing important tech numbers. Anyway.

5:05

I actually want to cover that

5:07

as one of our items in

5:10

the future. There's a couple of hosts

5:13

or charts or whatever that somebody put up that

5:15

I think would be really interesting. So this is

5:17

like, what is the relative speed of L2 cache

5:20

versus RAM? RAM versus disk? What

5:22

is the cost of a CPU core

5:25

or a database operation on the

5:27

cloud, etc. But I

5:29

don't like some of the numbers there. I

5:31

think they're way skewed. So I want to come

5:33

up with my own table and then put them

5:35

both forward. So that's why I haven't covered it

5:37

yet. But that's what this is about, I think.

5:39

Okay. Well, what's funny is that the teaser at

5:41

the beginning of this

5:44

post says there's only eight, but there's more if

5:46

you read it. But there's a table of contents

5:48

on the right. So you could just look at

5:50

the table of contents. The last one, I

5:54

think, is the most important for me is

5:57

strong opinions loosely held. For

6:00

this, for me, this means that

6:03

I'm going to have a default approach to

6:05

test for a lot of problems and a

6:07

default way to do things. It just speeds

6:09

things up. It's like, if I don't care

6:12

about it, just do it quickly. You've got

6:14

an opinion how things are done. However, be

6:16

open-minded enough to when somebody approaches you and

6:18

says, hey, there's a better way to do

6:20

this, or keeping your eyes

6:22

open with the tinkering, trying new

6:25

ways to do things, be willing to

6:27

abandon your old, strong opinion and go,

6:29

oh, there's a better way. Cool. My

6:32

new opinion is this now. It pretty much

6:34

sums up our podcast. Yeah,

6:36

exactly. Nice

6:40

post thinking about

6:43

planning for your time allotment for

6:45

the upcoming year. Yeah. Lots

6:48

of feedback on this one. Jeff out

6:50

there says, tech numbers like LEET, 1337. Excuse

6:54

me. Yes. Those are

6:56

the important tech numbers. Those are the important ones. Yeah.

6:59

LEET, NOOB, all those things. And then Grant

7:01

following up on the anti-fragile says, number

7:03

two speaks more to reactor principles that

7:05

came about with the cloud. It could

7:07

be summed up as build with the

7:09

expectation that your K8 pods can be

7:11

nuked mid-task at any point. Yeah,

7:14

exactly. That you might just use

7:16

one VM, but it could die. But if

7:18

you know that it's always getting abused, you

7:20

would build failover and those types of things

7:22

into it, making it, hence, less fragile. I

7:24

think that's kind of that principle there. Yeah.

7:27

Okay. But also, if it's

7:29

a side project, if it's down once in a

7:32

while, it's fine. Yeah. It's

7:34

not a side project. Yeah. Right. Maybe

7:37

a habit of great software engineers. I don't know. I wonder how

7:39

many blogs run on Kubernetes. I

7:42

don't know what the answer is, but I bet too many. The answer is

7:44

too many. Too many. Too many

7:46

run on Kubernetes. Okay. Okay.

7:49

What do you got for us? I got

7:51

eight nines of time on my blog. Okay.

7:55

We probably do, Brian, because ours runs on a

7:57

CDN as a static file. Okay. That's

8:00

pretty good uptime, right? Yeah. And the database

8:02

doesn't need to be migrated for that business.

8:04

Does it? All right, moving on. We've covered

8:06

this enough. Let's talk about a new version

8:09

of Flask. So Flask obviously is a super

8:11

important web framework. Like everyone has heard of

8:13

Flask. It's kind of one of the two

8:15

main pillars that I would think of as

8:17

like the old guard pillars, right? There's Flask

8:20

and Django. And the debate for so many

8:22

years has been like, do you use Flask?

8:24

Do you use Django? What kind of web

8:26

developer are you? Well, Flask 3.0 was released

8:29

just a little while ago. That

8:31

three weeks ago, something like that. There's

8:33

also a 301 that is kind of

8:36

unreleased, but has like very, very minor

8:38

changes. So anyway, the changes for a

8:40

two to three release are not that

8:43

huge. Although I think the numbering probably

8:45

comes from this removed previously deprecated code

8:47

there. That's probably what pushed it from

8:49

a two to three rather than some

8:52

mega new feature. Yeah. Right.

8:54

Because it's like, hey, it's not

8:56

backwards compatible. Miguel Grinberg actually wrote

8:59

an interesting but controversial piece on

9:01

that. I'm not going to link to it, but

9:03

y'all can check that out. I just want to

9:05

cover the release here. But this actually this first

9:07

line, there's a lot of discussion about it, let's

9:09

say. Neutral there.

9:11

So some of the things that are interesting here though,

9:13

Brian, are not so much interesting because they're Flask. They're

9:15

just like, wait, what? You could do that? So

9:19

deprecated the dunder version attributes. So

9:21

frequently people will go to their

9:23

dunder init.py for their package and

9:25

put a dunder version global variable

9:27

and set it to a string

9:29

and then you can say, well,

9:31

what version is the thing? Well,

9:33

it's flask.dunder version is the version

9:35

and that's just a string, right?

9:38

Yeah. Yeah. So the idea is

9:40

that's deprecated. And now the

9:42

way you're supposed to get

9:44

the version of Flask is

9:46

using feature detection or with

9:49

importlib.medadata.version of quote flask,

9:51

which I had nothing like. Okay.

9:55

So cool that that might work, but

9:57

how in the world do you make

9:59

it possible? possible for a

10:01

package that comes out of it. If I wrote a

10:03

package and I wanted to say .version

10:05

of my package in import lib, how do

10:07

I put a number in there that makes

10:09

it the version, right? So do

10:12

I have it here? Yeah, basically you have

10:15

to go, if you use a pyproject.toml, at

10:17

least one way is, and you just set

10:19

the metadata category, set the version to something,

10:21

and that's that. Yeah. Okay. Yeah,

10:24

pretty cool. And so that's where that

10:26

value comes from when you say import

10:29

lib.meditated.version. Pretty cool. What's more

10:31

interesting is deprecated a global variable. How

10:33

do you duplicate a global variable? I

10:35

know how to duplicate a method, a

10:38

class, but a variable? But just delete

10:40

it. No, no, that's

10:42

removing it. You wanted to give a warning

10:44

when you access the variable. Hmm. Okay.

10:48

And the variable value is a string. Like,

10:50

okay, what? So I'm like, I got to

10:52

say that. So I went and I tracked

10:54

down the change log. I checked on the

10:56

PR that made this happen. I'm like, I

10:58

got to see the code. What is this? So

11:01

it says a deprecated dunder version. And

11:03

here you can see somewhere, you

11:05

can just see they literally, like

11:07

you said, they deleted the dunder

11:09

version variable, but they added a

11:12

function to the module called dunder

11:14

get adder. For some reason, I

11:16

thought that only applied to classes, but no, apparently modules. And

11:19

given that it comes in and it asks

11:21

for the version, it says

11:23

no. There's a warning

11:25

and then it calls the import

11:27

lib.version thing on itself. Interesting, huh?

11:29

Yeah, actually, this is a handy thing

11:31

to keep around. Yes, exactly. If

11:34

you're asking for a dunder version, it warns

11:36

and then it does the new thing. Otherwise,

11:38

it just says, no, we don't have one

11:40

of them. But yeah, that's wild, right? Yeah.

11:43

I love how dynamic Python is. It

11:45

was nuts. It was a global variable.

11:47

Now it's a function. Or a... How

11:49

it's a function call to a dunder

11:51

get adder method on the module. I

11:54

think I'm going to replace all

11:56

of my global function declarations as

11:58

get adder. abnormalities

14:00

faster and easier. It ties bottlenecks such

14:02

as memory leaks, slow database queries, and

14:04

background jobs, and of course the dreaded

14:06

n plus one quorum query problem, and

14:08

more directly to source code so that

14:10

you can spend less time debugging and

14:12

more time building. I don't know where

14:14

that fits in the 20% of your

14:17

code example, but I think this

14:19

kind of stuff counts too. You'll love Scout

14:21

because it's built for developers by developers, makes

14:23

it easy to get set up and deliver

14:26

insights in less than four minutes. Seriously, you

14:28

could be done by the time you're done

14:30

listening to this episode or live stream. So

14:32

start your free trial and get instant insights

14:34

today. Visit pythonbytes.fm slash scout.

14:36

Please use our link because it really

14:39

does support the show. If they

14:41

don't know you're coming from us, then they

14:43

might not sponsor us anymore. So please, pythonbytes.fm

14:45

slash scout to check it out. Thank you

14:48

to Scout APM for supporting Python Bytes. Off

14:50

to you, Brian. What's next? Well, next we've

14:52

got something fun. On

14:55

real Python, there's a post Build

14:59

Conway's Game of Life with Python from

15:02

Leodanos Ramos. And

15:04

I just couldn't resist because I think I

15:06

have a feeling that maybe there's some new,

15:08

like younger software folks that maybe aren't familiar

15:10

with Game of Life. You've got to live

15:12

a while to know about the Game of

15:14

Life. Well,

15:17

I hope everybody knows about it, but it's

15:19

a it's kind of fun. And so this

15:21

is just a fairly good walkthrough of if

15:23

you were gonna build Game of Life. And

15:25

it's one of the cool things about it

15:28

is built in on the command line. It's

15:31

using ArcParse, but also Curses. And

15:34

you've got a Curses version in the terminal

15:36

of Game of Life. And how cool

15:38

is that? It's pretty, pretty fun.

15:40

Looking through, I've seen some rather

15:43

awful implementations of Game of

15:45

Life over the years. Like

15:47

a little bit of Pearl Golf going on

15:49

sometimes. But the this

15:52

one looks pretty good. It walks

15:54

through setting up an entire

15:56

project using PyProject.toml, which

15:58

is great. Then talking

16:00

about the different splitting up

16:02

the implementation into a grid

16:04

in one module, the patterns

16:07

in another, different views, and

16:09

then the CLI in another. Just

16:12

the split up looks good for modularizing

16:15

the application. A

16:20

lot of times people want to play like build

16:22

games or something and this is a game that

16:24

you don't actually have to play, it just plays

16:26

itself. I

16:29

just wanted to highlight it, it's just a

16:31

fun walkthrough of implementing Game of Life.

16:34

Yeah, you already talked about these little projects

16:36

in your first thing, right? Here's a cool

16:38

little project that you can play around with

16:40

and I think games are a good way

16:42

to get into coding, not Doom, Half-Life, whatever.

16:44

I've been playing a ton of small ants,

16:46

by the way, also not a good way

16:48

to get into programming. It seems like something

16:50

that would take forever, but small ants is

16:52

awesome. However, I think this is like little

16:55

games, they're easy, you don't have to have

16:57

a huge amount of domain knowledge, you don't

16:59

have to know about databases and connection strings

17:01

and how do I embed the password and

17:03

how do I make it run, all that,

17:05

just have fun. Yeah, and one

17:08

of the fun things about this is

17:10

the hiding of, so it has a

17:12

life grid. So one of the, of

17:14

this project, almost all of the project

17:16

is going to be transferable to other

17:18

stuff, except for the cursors part. I'm

17:20

probably not going to write a whole bunch

17:22

of cursors applications, but that's the same thing

17:24

with most, a lot of games. Unless you

17:26

become a game developer, a lot of the

17:29

game development, it's using some sort of

17:32

user interface stuff that you're probably not going to

17:34

use for a lot of projects, unless you go

17:36

into using games, writing more games, which is great.

17:38

But having that isolated off to one

17:41

little part was actually, it's

17:43

pretty kind of nice to be able to set that.

17:46

Anyway, good implementation. Yeah, it is cool

17:48

to see how you might create that

17:50

little UI as well. I mean, usually

17:52

people are doing Unreal Engine, Unity,

17:55

Pygame, Arcade, one of these is like

17:57

quite, you quite deep down in those.

18:00

frameworks. Yeah. And it's just a little bit.

18:02

One of the things I wasn't sure about

18:04

tonight, so I didn't see any disclaimer at

18:06

the top. I did not run through this,

18:08

but was if I

18:10

didn't know that you could use curses

18:12

on Windows, for instance, so is this

18:14

a maybe somebody in the

18:16

audience knows if you could. Yeah, maybe someone

18:18

knows. It might be because of Windows terminal

18:20

versus command prompts. Oh, okay. Yeah, possibly.

18:23

But that's a guess. Zero

18:25

knowledge. Alright, enough of this fun, Brian. Let's

18:28

get down to business. Okay. All there's business

18:30

that is. This comes from Marco Gorelli.

18:32

And this is just a simple plugin

18:34

for polars. Now, if you don't know

18:37

polars, polars is like and as with

18:39

two major differences, it has a lazy

18:41

fluent API. So you can kind of

18:43

chain commands together and it doesn't actually

18:46

execute until you kind of work through

18:48

them, kind of generators and those sorts

18:50

of things. And it's written in Rust

18:52

to be super, super fast. So this

18:54

is a plugin for polars. And the

18:57

idea is it lets you answer questions

18:59

like how many business days between

19:01

now and then. So if like my

19:03

my payment system says you

19:05

get paid net 30, well, or you

19:07

get paid after 10 business days, but

19:10

one of those days is the 4th

19:12

of July and then there's the weekend

19:14

and then there's like how many days,

19:16

you know, well, calendar math is already

19:18

gnarly. If you want

19:20

to like throw in skipping weekends, skipping holidays,

19:23

even worse, right? So super simple.

19:26

This thing, the extension itself is written

19:28

in Rust. So it's kind of compatible

19:30

with polars. I believe polars has actually

19:32

a Rust API as well as a

19:34

Python API. So maybe this is available

19:36

there. I have zero information about that.

19:38

All you do is you

19:40

can even define your own holidays and weekends

19:43

like every Wednesday, I'm taking it off, whatever.

19:45

So the way it works is

19:47

you just go import polars

19:50

and then polars business somewhere down

19:53

here. You can go to grab

19:55

a column, say offset by and

19:58

just say five B.D. and

20:01

it'll skip by, you know, it'll go

20:03

through all the columns, figure out what

20:05

their dates are, and using vector math,

20:07

convert those to five days ahead by

20:09

however they offset. Like each one individually,

20:11

that's pretty awesome, right? Yeah, fun. Yeah,

20:13

I don't know how that works. That

20:16

sounds complicated. But yeah, so

20:18

you can put in like, what are your weekend days?

20:20

What are your, there's,

20:22

you know, obviously Saturday and Sunday is pretty

20:24

standard. But like, we have restaurants around here

20:26

that are closed on Sunday and Monday, they're

20:29

open on Friday and Saturday. So if

20:31

you're trying to like do business math

20:33

around, like that kind of schedule, like you

20:36

could just say Sunday, Monday for your weekend

20:38

and you're good to go. Nice. Yeah, like,

20:40

one of the things I was thinking about

20:42

was like, education wise, schools

20:44

often have like tons of, so

20:47

many times, so much time off.

20:49

Yeah, like in like, teacher non

20:51

work days, which are whatever

20:54

it's, the teachers are still working. It's just

20:56

the students aren't working. They should call it

20:58

student non work days. Yeah,

21:01

like my daughter is like, Oh, I don't have

21:03

school this week. No school this week. Why? Like,

21:05

oh, it's parent teacher conference. Oh,

21:07

my okay. Yeah, my daughter even calls,

21:10

they call it, her friends

21:12

call it, like no, no

21:14

school November. Yes, exactly. Exactly.

21:17

Anyway, so that would be good

21:19

for that. Or lots

21:21

of other businesses have wacky holidays.

21:24

So people can check this out. It's, it's,

21:26

you know, easy to understand, easy to adopt if it's

21:28

relevant to you. But it seems pretty cool. Yeah. All

21:31

right, Brian, how extra do you feel

21:33

today? I just have one extra. And

21:35

that is last week, you get what

21:38

I got convinced to do a Black

21:40

Friday sale for the complete

21:42

PyTest course. And it is one

21:45

word Black Friday, all caps, just

21:47

that and you get 50% off

21:49

of off the course.

21:51

So check it out. I'm in a

21:54

in over the break, I did a whole bunch on it. So

21:56

I've up to what chapter 13. Oh, chapter

21:58

three. Chapter 13 was

22:00

great. I wanted to talk about that a little bit.

22:02

Chapter 13 was on debugging. And

22:05

instead of just artificially adding errors

22:07

to the code, I decided to

22:09

do a test-driven development

22:12

little thing of writing a couple

22:14

tests first, doing the implementation, and

22:16

resisting the urge to refactor during

22:18

the implementation. And then once everything

22:20

was green, going back and doing

22:22

a refactoring. And then

22:25

explored... I enjoyed learning

22:27

this anyway. So it's

22:29

exploring testing, debugging,

22:32

using PyTest plus PDB for

22:34

the Python debugger. One

22:37

of the cool things about learning PDB is

22:39

you can use it with talks. So you

22:41

can debug an individual environment if maybe your

22:43

code's only failing on 3.8 or 3.9 or

22:45

something. So anyway,

22:47

that's still up. And that's my

22:49

extra. How about you? Very cool.

22:52

Yeah, congrats on getting Chapter 13

22:54

up. And I think debugging 13,

22:56

I think stuff could

22:58

go out. That's perfect. As long as you didn't release a

23:00

chapter on Friday the 13th, everything will be fine.

23:02

Right. I got a

23:05

few things. First, there was like a really

23:07

heartfelt message from Kenneth Wright that he posted

23:09

on X Twitter. And so I thought maybe

23:11

I'd just read that because it's, you know,

23:13

just kind of remind everyone of, you know,

23:15

how nice the community is and why we

23:17

all like to be here. You up for

23:19

that, Brian? Yeah. All right. Kenneth's

23:21

boos on things like requests and many other

23:23

projects says, Dear Python community, I hope this

23:25

letter finds you well and thriving in your

23:27

various endeavors. Python, my name is Kenneth Wright.

23:29

So I've been part of this vibrant community

23:31

for quite some time, contributing projects like Quest,

23:33

Pip, Envy, and among others. Today I

23:35

write to you not just as a fellow

23:37

Python enthusiast, but as someone who, like many

23:40

of you, have faced moments of uncertainty and

23:42

feared my journey with programming, it's a path

23:44

that while rewarding is often fraught with challenges

23:46

and moments of self-doubt. I'm reaching out to

23:48

share these feelings because I believe in the

23:50

power of our community to support and uplift

23:52

each other. In recent times, I found myself

23:55

grappling with a sense of fear, fear of

23:57

not living up to expectations, fear of the

23:59

unknown, or ever-evolving field and perhaps fear not

24:01

being able to contribute as significantly as I

24:03

have in the past. These are feelings

24:05

I'm sure many can relate to. But in

24:07

these times of vulnerability that I'm reminded

24:09

of the strength and compassion of the

24:11

Python community, that we're a group that

24:13

not only shares a love for programming

24:15

but also cares for the well-being of

24:17

its members, your support, encouragement, and willingness

24:19

to share knowledge and experiences have always

24:21

been a source of strength for me.

24:23

So reaching out to ask for your

24:25

understanding and continued support, not just for

24:27

me but for everyone in the community

24:30

facing similar challenges. It goes on. I'll let people read it,

24:32

link to it in the show notes. But I just think

24:34

that's kind of a nice message and I wanted

24:36

to amplify it a bit for Kenneth. Yeah. I

24:39

mean, obviously, he's someone who people, I know

24:41

many people look up to as like, wow,

24:43

look how much he's accomplished. And so if

24:45

you're feeling uncertain or like you're not

24:47

contributing enough or whatever, like if Kenneth also feels

24:49

that way, like, you know, maybe these are just

24:51

feelings people have and not so much things that

24:53

should stop you. All right. So

24:55

that's one. Last time I talked about Python

24:57

3.13, 0, alpha 1 being out. Well,

25:01

this time, since it was so close to when

25:03

3.2 or alpha 2 was coming out, alpha 2

25:05

is here as well. So I just wanted to

25:07

link over to that. Really

25:09

many of the features that they talk about, what's

25:12

coming in 3.13 really just has to

25:14

do with like deprecations and very concrete stuff.

25:16

But I know there's the Faster C Python

25:19

initiative making a lot of changes, although they're

25:21

not even at all mentioned here.

25:23

Just like we remove many deprecated things and

25:25

many of the dead batteries that we talked about

25:27

like Telnet lib, go find another way to

25:29

implement your Telnet client instead of the

25:31

standard library and so on. So

25:33

alpha 3 will be scheduled for

25:36

December 19th. So we can

25:38

round out our year with a third Python

25:40

3.13 alpha release. So

25:43

that's nice and quick. And finally,

25:45

as you touched on as well, thank you

25:47

to all the people who participated in the

25:49

Black Friday thing. We raised

25:51

a bunch of money for charity, which

25:53

is awesome. So really happy to have

25:55

done that. So just a bit of

25:57

a heartfelt thanks on my end for

25:59

people. All right, ready for a joke?

26:01

Yeah. The joke is called select star and

26:03

I think we can all relate to it.

26:06

Okay, so select star being theoretically a bad

26:08

thing. Can I get this image all the

26:10

way over? No, not really. Okay,

26:13

so the question here says, it's

26:15

a person. I think this is

26:17

Will Smith from iRobot kind of

26:19

type thing. Anyway, it doesn't really

26:21

matter. The developer says, can AI

26:23

really write efficient SQL queries? And

26:26

AI looks back at the developers. Can you? Maybe

26:29

not. And

26:32

of course, the title is select star, which is

26:34

like barring joins and other crazy group by stuff,

26:36

like pretty much like the worst thing you can

26:39

do for performance. Like give me everything. I don't

26:41

care if I use it, right? I know there's

26:43

like a 10 hell of a byte per row

26:45

description. I don't need that. Just

26:47

give it to me anyway. I think

26:50

this is like the ideal use case

26:52

for using an AI is optimizing SQL

26:54

queries. Yeah. Yeah.

26:56

Anyway, I know I wouldn't trust. I don't

26:58

trust myself with production SQL. So

27:02

can AI really write efficient SQL

27:04

queries? Can you? Okay,

27:07

well, see some agreement in the audience out there. It's

27:09

all very nice. I want to

27:11

add a just a funny thing that my

27:13

wife shared with me yesterday. And

27:17

as many of you know, or maybe you

27:19

don't know, I used to have short

27:21

hair and over the pandemic grew it

27:23

out. And now I have longish hair.

27:27

I would say that counts as long. And

27:30

so often it's pulled back,

27:32

but it's often not also. So I

27:34

wanted to share this funny thing that

27:36

my wife shared. So I admire those

27:39

with hairstyles. I don't have a hairstyle.

27:41

I have hair. Most

27:43

days it has zero caterpillars in it. That's

27:45

as good as it gets. So

27:49

girl women with like long flowing hair.

27:53

Honestly, the length is about the same as yours

27:55

right now. So that's pretty awesome. And apparently this

27:57

was that came up

27:59

by. cocks. So that's funny. Anyway.

28:01

Yeah, very good. Well, also a very

28:03

good episode. Good to be back with

28:05

you and with everyone. Yeah, thank you. Yeah,

28:07

you bet. Thanks everyone for listening. See you

28:09

later.