РАСШИФРОВКА 🔗
WEBVTT
00:00:00.080 --> 00:00:04.800
Hello and welcome to Python Bytes, where we deliver Python news and headlines directly to your earbuds.
00:00:05.160 --> 00:00:11.040
This is episode 497, recorded September 22nd, 2026.
00:00:11.420 --> 00:00:12.640
I'm Calvin Hendryx-Parker.
00:00:12.880 --> 00:00:13.680
And I'm Michael Kennedy.
00:00:14.100 --> 00:00:14.660
Excellent, Michael.
00:00:15.400 --> 00:00:17.380
I think you get a little message here you want to give.
00:00:17.580 --> 00:00:20.620
This episode is brought to you by Logfire from Pydantic.
00:00:20.860 --> 00:00:23.960
Check them out at pythonbytes.fm/logfire.
00:00:24.140 --> 00:00:28.800
They are super, they're a great company, you know, Samuel and team.
00:00:29.160 --> 00:00:34.400
Pydantic has been a good community member, and this is a really great way to add observability to your app.
00:00:34.400 --> 00:00:35.780
I will tell you more about that later.
00:00:36.100 --> 00:00:36.320
Perfect.
00:00:36.620 --> 00:00:38.320
Right now, connect us, Calvin.
00:00:38.600 --> 00:00:39.000
Yeah, I will.
00:00:39.200 --> 00:00:41.620
You should connect with Michael and I and the show.
00:00:41.740 --> 00:00:44.820
We are over on Mastodon, Bluesky X, and LinkedIn.
00:00:45.420 --> 00:00:52.020
And make sure you can, if you want to check out and be part of the fun, we do live stream this at pythonbytes.fm/live.
00:00:52.240 --> 00:00:54.180
We want to be part of the live studio audience.
00:00:54.420 --> 00:00:57.560
Usually it's on Tuesdays at 7 a.m. Pacific, but we're doing a little later today.
00:00:58.380 --> 00:01:01.140
Older versions are available on the YouTube as well.
00:01:01.380 --> 00:01:11.980
And finally, if you want the Artisanal handcrafted digest of every week's show notes and email form, join our friends of the show list and we will send you that email pronto.
00:01:12.500 --> 00:01:17.340
Speaking of pronto, things that go fast, usually you wanna profile things that make them go faster.
00:01:17.580 --> 00:01:18.300
What do you got for us, Michael?
00:01:18.620 --> 00:01:20.980
You absolutely do wanna profile things to go fast.
00:01:21.120 --> 00:01:22.440
I wanna tell you about Tachyon.
00:01:22.720 --> 00:01:24.380
Now this is interesting on a couple of levels.
00:01:24.740 --> 00:01:28.160
So this is Python 315's new profiler.
00:01:28.240 --> 00:01:28.560
Okay.
00:01:29.020 --> 00:01:31.640
What's interesting about it is it's built in the Python.
00:01:31.810 --> 00:01:38.700
So previously we've had profile and C profile and it still is a bit of a mystery to me while we have two.
00:01:39.180 --> 00:01:41.100
And they weren't the easiest to use either.
00:01:41.540 --> 00:01:43.140
No, they were not, not, they were not that obvious.
00:01:43.580 --> 00:01:48.460
And what the, what I'm going to tell you about with this tachyon thing is going to literally be a game changer.
00:01:48.630 --> 00:01:49.600
The game shall be changed.
00:01:49.860 --> 00:01:50.200
All right.
00:01:50.520 --> 00:01:58.360
So the other ones, it was basically the same details, but one was implemented in C and one was implemented in Python.
00:01:58.880 --> 00:02:12.440
So you kind of think of them at the same, but for the most part, they were tracing profilers, a tracing profiler instruments, every function call, every stack change, all that kind of stuff to say, now we're entering this function.
00:02:12.560 --> 00:02:13.260
Now we're leaving this function.
00:02:13.400 --> 00:02:14.920
Now this function is now calling that function.
00:02:15.020 --> 00:02:16.160
Now that function we're leaving that function.
00:02:16.190 --> 00:02:16.600
You know what I mean?
00:02:16.660 --> 00:02:20.900
Like it's, it's really invasive and that can dramatically slow down your code.
00:02:21.000 --> 00:02:24.380
And what's frustrating is it slows it down in an uneven way.
00:02:24.520 --> 00:02:28.180
Like some functions that are very shatty, you slow down a lot.
00:02:28.260 --> 00:02:32.460
And one that calls something external, that's really equally slow, might barely slow down at all.
00:02:32.760 --> 00:02:32.840
Right.
00:02:32.840 --> 00:02:34.900
And so it also skews your results.
00:02:35.280 --> 00:02:37.480
So people use sampling profilers.
00:02:37.940 --> 00:02:38.160
All right.
00:02:38.380 --> 00:02:40.060
And I don't think the other one did sampling.
00:02:40.220 --> 00:02:40.640
Maybe it did.
00:02:40.660 --> 00:02:49.520
I don't have not used it that way, but we now have this thing called tachyon from Pablo, and Laszlo.
00:02:50.020 --> 00:02:51.480
Sorry, Laszlo, I'm forgetting your last name.
00:02:51.860 --> 00:02:55.100
I, they're on the PEP of for, for, putting this together.
00:02:55.480 --> 00:03:01.780
They worked on, I had them on the podcast on talk Python last week, had a really awesome conversation about this thing called tacky on.
00:03:01.900 --> 00:03:03.660
So this is unique in a couple of ways.
00:03:03.860 --> 00:03:05.040
It has its own logo.
00:03:05.540 --> 00:03:09.900
How many sections of the Python standard library get their own logo?
00:03:10.190 --> 00:03:10.740
I think none.
00:03:11.150 --> 00:03:11.540
I know.
00:03:11.800 --> 00:03:13.200
And they usually don't get a code name.
00:03:13.500 --> 00:03:23.240
Now, Pablo and Lazlo would like this to be addressed, referred to as tacky on not profiling dot sampling, but they've redone the module layout.
00:03:23.420 --> 00:03:26.740
So there's profiling dot tracing, which is C profile renamed.
00:03:27.120 --> 00:03:31.360
And there's profiling dot sampling, which is a statistical profiler.
00:03:31.780 --> 00:03:39.800
And it leverages the new features available in 3.15 that allow profilers to hook in externally into CD Python, the runtime.
00:03:40.160 --> 00:03:41.660
Now you might think, okay, so fine.
00:03:41.720 --> 00:03:44.700
It's a sampling profiler, not a tracing profiler, whatever.
00:03:44.860 --> 00:03:45.360
No big deal.
00:03:45.800
Load full transcript
--> 00:03:48.040
Let me call your attention to the left here, Calvin.
00:03:48.180 --> 00:03:48.960
Do you see this attach?
00:03:49.460 --> 00:03:50.120
Ooh, yes.
00:03:50.660 --> 00:04:01.480
This means I might have a project running on a server in a Docker container, processing real results, suffering some kind of problem that I cannot reproduce in dev or QA.
00:04:01.610 --> 00:04:05.000
I can SSH and then Docker exec over into there.
00:04:05.010 --> 00:04:08.320
And I can say Python dashed in profiling sampling attached to this PID.
00:04:08.720 --> 00:04:09.740
Let it go for 30 seconds.
00:04:10.400 --> 00:04:10.820
Detach.
00:04:10.990 --> 00:04:18.579
And it will have generated a production level profiling PSTAT file that I can flame graph and I can look at and so on.
00:04:18.840 --> 00:04:20.579
That sounds scary, but you're very useful.
00:04:21.000 --> 00:04:22.260
It sounds scary.
00:04:22.480 --> 00:04:29.660
And what I said about the degradation of performance with the tracing one, it would be, they say this has like a 2% to 3% overhead.
00:04:29.940 --> 00:04:30.240
Oh, wow.
00:04:30.520 --> 00:04:31.020
That's not bad at all.
00:04:31.380 --> 00:04:31.500
Okay.
00:04:31.760 --> 00:04:32.320
That's neat, right?
00:04:32.700 --> 00:04:32.920
Yeah.
00:04:33.200 --> 00:04:33.960
It's actually super neat.
00:04:34.320 --> 00:04:34.440
Yeah.
00:04:35.440 --> 00:04:36.320
This is a big deal.
00:04:36.580 --> 00:04:37.720
I mean, I know it's a big deal.
00:04:37.800 --> 00:04:38.760
It's such a big deal.
00:04:38.900 --> 00:04:40.340
Python 315 is going to be a huge deal.
00:04:40.620 --> 00:04:41.880
It's going to be such a big deal.
00:04:41.920 --> 00:04:42.560
It'll be so good.
00:04:42.940 --> 00:04:46.680
There's also run, which is the more standard, like just run this and profile it.
00:04:46.920 --> 00:04:48.500
This attach is not just production though.
00:04:48.700 --> 00:04:51.740
Here's something that drives me out of my mind when I'm trying to do profiling.
00:04:51.860 --> 00:04:54.600
I want to profile how long this web request that takes.
00:04:55.080 --> 00:04:57.580
So I profiles, run the app.
00:04:57.780 --> 00:05:01.500
I go over and I click on the endpoint a few times and then I shut it down.
00:05:01.980 --> 00:05:10.120
99.9% of all the profiling stuff is loading the web framework, connecting the database, doing this, doing like all the junk that is not what you're interested in.
00:05:10.380 --> 00:05:10.560
No.
00:05:10.720 --> 00:05:15.640
So you could start your app and get it just ready to profile, attach it, do the thing and detach it.
00:05:15.920 --> 00:05:17.340
And you lose a lot of that noise.
00:05:17.440 --> 00:05:18.220
I think it's really neat.
00:05:18.520 --> 00:05:21.880
So yeah, there's a whole section on profiling and production and how that might go.
00:05:22.240 --> 00:05:23.720
There's also different profiling modes.
00:05:24.180 --> 00:05:27.080
Wall clock mode, CPU mode, GIL mode.
00:05:27.160 --> 00:05:29.460
So you can understand GIL contention specifically.
00:05:30.460 --> 00:05:34.240
If you're trying to do threading, CPU is, imagine there was no waits in the world.
00:05:34.280 --> 00:05:36.780
If I wait on a database, that doesn't count because that's external.
00:05:37.100 --> 00:05:38.540
I can't address that.
00:05:38.660 --> 00:05:39.760
It's not part of my program.
00:05:40.000 --> 00:05:42.240
So let's pretend that the stuff I'm waiting on doesn't exist.
00:05:42.720 --> 00:05:43.600
That's the CPU mode.
00:05:43.740 --> 00:05:44.800
Wall mode is like traditional.
00:05:45.220 --> 00:05:48.480
And exceptions is just how much time are you spending handling exceptions and errors.
00:05:48.810 --> 00:05:49.920
I think there's a bunch of stuff here.
00:05:50.010 --> 00:05:53.780
Like it's, check out the docs I linked to, the little TOC on the left.
00:05:53.940 --> 00:05:54.340
It's ridiculous.
00:05:54.770 --> 00:05:55.260
It's really cool.
00:05:55.440 --> 00:05:55.620
Yeah, yeah.
00:05:55.880 --> 00:06:03.420
I mean, for people, if you get down to the point where you need to profile, you know you've got a serious performance problem because you don't take profiling typically lightly.
00:06:03.820 --> 00:06:08.080
But this is going to be a huge boost for folks who are like, oh, Python's just too slow and can't do the things.
00:06:08.520 --> 00:06:09.740
It might actually be your code.
00:06:10.100 --> 00:06:17.500
And if you use this tool, you may find you can actually fix a problem be just as fast without having to switch off the python i like that a lot it's amazing yeah i think
00:06:17.540 --> 00:06:21.960
this is really neat it's a big i think it's a bigger step than it initially sounds like for
00:06:22.100 --> 00:06:32.640
python yeah it got its own name and logo in the python dog i mean that's significant that's kind of significant i unheard of definitely unheard of i'm surprised there weren't more boats rocked when
00:06:32.760 --> 00:06:38.160
that that kind of went down you know well pablo's uh a force of nature he got it done i guess so
00:06:38.360 --> 00:06:47.040
i'm glad they worked on that's really yeah me too well hey i've got a new there's been a big week in releases for the Python world in kind of other areas.
00:06:47.280 --> 00:06:53.560
So the first one I wanted to cover was Cloudflare has released Python workers that are now, it's now generally available.
00:06:53.920 --> 00:07:00.400
So for the last two years, I believe it is, that you could actually try out Python workers in the Cloudflare space.
00:07:00.480 --> 00:07:15.260
So if you're not familiar, you can use compute on Cloudflare. And so it puts your compute at the edge and it's been generally JavaScript or TypeScript native because it does, it's all WASM. So it compiles it, runs it at the edge, and it's pretty limited too.
00:07:15.460 --> 00:07:20.900
So one of the things you had to always worry about was RAM consideration,
size of the container.
00:07:21.580 --> 00:07:24.800
It was pretty, you know, cranked down as far as constraints go.
00:07:25.320 --> 00:07:30.660
They just released the Python workers out of beta, so it's now generally available. So now a first-class language.
00:07:31.120 --> 00:07:42.220
And one of the big holdups from before was there's no more manual JavaScript interop. If you were using one of the other Cloudflare services and you were using the beta version of the Python workers.
00:07:42.750 --> 00:07:55.500
Like for example, if you're using R2, which is their object store, if you're using D1, which is their like mini like SQLite like database, if you're using durable objects for WebSockets, you had to convert all those data structures over into JavaScript
00:07:55.980 --> 00:08:01.640
first from Python and then use those services, which pain in the butt, it's just like extra hoops.
00:08:01.860 --> 00:08:05.260
You'd forget how to, you know, feed that overhead going along with that as well.
00:08:05.670 --> 00:08:08.020
But now they're all natively available in Python.
00:08:08.180 --> 00:08:14.300
So you can now just do self.m.queue.send if you want to use their durable objects queue system.
00:08:14.820 --> 00:08:16.160
It all just works out of the box.
00:08:16.600 --> 00:08:17.820
It's all running on Pyodide.
00:08:18.200 --> 00:08:26.600
So it's WASM compiled Python with another big thing that wasn't available in the beta first was the ability for real TCP socket support.
00:08:26.880 --> 00:08:29.720
So in the beta you could not connect to other databases.
00:08:29.980 --> 00:08:37.620
So for example if you wanted to use I believe it's called Hyperscale which is their MySQL Postgres hosted versions of those things, that was a no go.
00:08:38.099 --> 00:08:38.800
Couldn't couldn't do it.
00:08:39.360 --> 00:08:40.320
Now it has support for that.
00:08:40.520 --> 00:08:42.400
It also has support for frameworks.
00:08:43.060 --> 00:08:53.060
So it more now natively supports FastAPI, Django, Flask, a bunch of AI libs. So if you want to do like a line chain or build MCP servers, it's pretty cool.
00:08:53.220 --> 00:08:57.780
So here I've got the docs open right here, which is, you know, the Python worker docs are pretty, pretty nice, pretty clear.
00:08:58.260 --> 00:09:09.640
I've been building lots of little fun, unique tools for myself using Cloudflare workers, which is very handy. So for example, I've replaced a lot of my usage of like Zapier or make.com just with like little one-off workers.
00:09:10.080 --> 00:09:13.120
And now I can write them in Python because I was having to do them in JavaScript before.
00:09:13.740 --> 00:09:14.860
So yeah.
00:09:15.280 --> 00:09:21.500
Now don't get too excited when I say that Django is available on FastAPI or on the Cloudflare workers.
00:09:21.780 --> 00:09:23.400
You still have some limits.
00:09:23.680 --> 00:09:25.000
Actually, I'll pull that up right here.
00:09:25.440 --> 00:09:26.800
The limits are still real.
00:09:27.280 --> 00:09:31.880
You have a memory limit for free and for paid of 128 megabytes of memory.
00:09:32.160 --> 00:09:39.700
I can tell you from my experience using FastAPI Cloud, pretty regularly going up close to 512, sometimes over.
00:09:40.120 --> 00:09:49.840
And so you have to have a pretty slim app to be able to run in that small of a memory footprint. It's got to be really, really stateless and be able to handle that kind of memory usage.
00:09:50.260 --> 00:09:58.260
So deploying a full blown Django app, probably not going to happen inside of these Cloudflare workers, but it's really not the intention or what you should be using them for.
00:09:58.500 --> 00:10:02.460
Think about them as, again, kind of a Zapier drop-in replacement.
00:10:03.080 --> 00:10:09.900
Instead of using Zapier, use Cloudflare because you could write the code. You have more control. Now you've got TCP sockets, so you can connect to real databases.
00:10:10.200 --> 00:10:13.780
You can use their, I actually use their durable objects in some of my projects.
00:10:14.280 --> 00:10:15.140
Super convenient.
00:10:15.460 --> 00:10:22.080
I mean, just the convenience to having some of these pieces all geographically distributed automatically for you is a huge win.
00:10:22.760 --> 00:10:25.280
But just yet to begin, be aware of all the limits.
00:10:25.460 --> 00:10:27.220
Like, again, another thing it's worker size.
00:10:27.500 --> 00:10:33.020
You're limited to 64 megabytes of disk space that you basically have available to you.
00:10:33.180 --> 00:10:35.900
This has improved dramatically, though, since last year.
00:10:36.260 --> 00:10:48.120
Earlier this spring or late last 2025, these limits used to be considerably smaller, like three megabytes and 10 megabytes were the limits up until very, very recently. So that's another big change that has happened.
00:10:48.540 --> 00:10:57.820
That's across all workers, not just the Python workers, but that definitely helps the Python folks be able to and get our code encapsulated down into something small enough that can run in there.
00:10:58.460 --> 00:11:04.200
This has all been done because the underlying platform was formalized as part of like PEP783, the Py and Scriptum.
00:11:04.500 --> 00:11:08.320
So after a year of discussion, we now have generally available Python workers.
00:11:08.560 --> 00:11:15.180
I'm really excited about this because now I can do real Python at the CloudFloor edge with no JavaScript required at all.
00:11:15.460 --> 00:11:15.940
Amazing.
00:11:16.360 --> 00:11:16.940
I'm excited about this as well.
00:11:16.940 --> 00:11:17.800
Have you played with this at all?
00:11:18.320 --> 00:11:23.080
No, no, but playing in a space tangentially near it.
00:11:23.360 --> 00:11:25.820
I've not done anything with edge workers.
00:11:26.260 --> 00:11:29.520
It looks super interesting to me, but I just haven't found the use case for me.
00:11:29.940 -->
00:11:30.080
So,
00:11:30.600 --> 00:11:30.700
no.
00:11:30.740 --> 00:11:38.020
I've got tons of event-driven, like any place where I'd want a cron job to go off and do something, I may put it into a Cloudflare worker.
00:11:38.320 --> 00:11:49.020
I use it for Cloudflare pages quite a bit to publish quick little microsites to show off to a customer or to a prospect to convey something to them that we've worked on, because I can put authentication.
00:11:49.570 --> 00:11:53.960
You can use middleware in the pages to actually cause authentication to trigger.
00:11:54.440 --> 00:11:55.560
It's pretty slick.
00:11:55.970 --> 00:12:01.720
Yeah, one thing that I do find interesting is the Cloudflare D1, and this is closer to where I'm talking.
00:12:01.740 --> 00:12:02.080
Isn't that cool?
00:12:02.380 --> 00:12:06.680
Which is basically SQLite on the edge, but somehow it syncs.
00:12:07.060 --> 00:12:10.320
Maybe with Lightstream behaviors, I'm not entirely sure.
00:12:10.660 --> 00:12:13.880
But it sounds very cool, and I'm actually all here for SQLite these days.
00:12:14.040 --> 00:12:15.460
I've been doing some interesting stuff.
00:12:15.620 --> 00:12:21.360
My number one item next week will be something, trying to think about how dramatic I wanna be here.
00:12:21.560 --> 00:12:25.140
How dramatic is something pretty amazing on SQLite next week.
00:12:25.320 --> 00:12:25.960
Not DuckDB.
00:12:27.100 --> 00:12:28.380
No, DuckDB is an option.
00:12:28.820 --> 00:12:34.880
This is more of a transactional thing that's more well-shaded with row-level behavior than with columns.
00:12:35.300 --> 00:12:36.660
But if it was data science, then yeah, sure.
00:12:36.960 --> 00:12:37.840
See, this is anticipation.
00:12:38.100 --> 00:12:39.120
We'll just have to wait until next week.
00:12:39.120 --> 00:12:39.540
Oh my gosh.
00:12:39.580 --> 00:12:40.620
I'm not even-
00:12:40.720 --> 00:12:42.320
I'm gonna literally be sitting on the edge of my chair.
00:12:42.600 --> 00:12:43.340
It's gonna be hard.
00:12:43.520 --> 00:12:45.880
Until then. - It's worse than Christmas, Let me tell you, but we're going to do it.
00:12:46.600 --> 00:12:47.060
We're going to do it.
00:12:47.160 --> 00:12:51.300
Now, before we move on, though, I do want to tell you about our sponsor, Pydantic Logfire.
00:12:51.460 --> 00:12:54.160
They've been sponsored in the last couple episodes, and we really appreciate that.
00:12:54.580 --> 00:12:56.100
What a cool forum.
00:12:56.340 --> 00:13:08.940
So if you run Python apps in production, especially if these are AI, you're running AI models, using AI models, not necessarily executing models, but, you know, using something you're talking to, then you want to check out Logfire.
00:13:09.260 --> 00:13:10.280
So here's the story.
00:13:10.480 --> 00:13:11.220
It's 2 a.m.
00:13:11.520 --> 00:13:12.220
You know where your kids are?
00:13:12.260 --> 00:13:13.100
No, that was a different one.
00:13:13.620 --> 00:13:14.720
Your AI agent has
failed.
00:13:15.150 --> 00:13:16.760
Was it the model, the tool call, the database?
00:13:17.660 --> 00:13:20.940
Most observability tools can't tell you because they only see part of your stack.
00:13:21.120 --> 00:13:22.540
Pydanic Logfire sees all of it.
00:13:22.780 --> 00:13:29.240
One trace across your agents, LLMs, APIs, and database down to the infrastructure, whether it's services, Kubernetes, and on the hosts.
00:13:29.700 --> 00:13:33.200
It's built on open telemetry with SDKs for Python, TypeScript, and Rust.
00:13:33.470 --> 00:13:35.280
And it works with any OTEL compatible language.
00:13:35.740 --> 00:13:39.160
Every prompt costs right next to your vector searches and API calls.
00:13:39.470 --> 00:13:41.880
You query everything with Postgres compatible SQL.
00:13:42.620 --> 00:13:44.980
And so your coding agent can as well.
00:13:45.360 --> 00:13:49.120
It can use the Logfire MCP server, which is excellent.
00:13:49.540 --> 00:13:50.000
Stop guessing.
00:13:50.420 --> 00:13:50.820
Read the trace.
00:13:51.120 --> 00:13:51.760
PyDenic Logfire.
00:13:52.120 --> 00:13:54.040
AI is still just engineering.
00:13:54.430 --> 00:13:57.620
So visit pythonbytes.fm/logfire today and sign up.
00:13:57.820 --> 00:13:59.120
Get 10 million records for free every month.
00:13:59.150 --> 00:14:00.180
No credit card required.
00:14:00.680 --> 00:14:03.620
You can even click on board with your coding agent.
00:14:03.790 --> 00:14:05.540
I love it so much to copy the prompt.
00:14:05.880 --> 00:14:06.260
It's scary so much.
00:14:06.260 --> 00:14:09.520
You give the Clotter codex and just say, user wants to do this.
00:14:10.040 --> 00:14:10.360
Help them.
00:14:10.520 --> 00:14:12.320
And you and I, we got this.
00:14:12.600 --> 00:14:13.880
and then you'll just have it all set up for them.
00:14:13.960 --> 00:14:15.960
That's a really cool way to onboard your projects these days.
00:14:16.270 --> 00:14:17.820
So thanks to Pydantic for supporting the show.
00:14:18.040 --> 00:14:19.860
It's a good thing we know them and we would trust them.
00:14:20.120 --> 00:14:20.340
Exactly.
00:14:20.800 --> 00:14:24.040
It's very unlikely to have.
00:14:24.420 --> 00:14:24.580
True.
00:14:24.860 --> 00:14:25.300
Very unlikely.
00:14:25.840 --> 00:14:26.660
All right, over to you.
00:14:26.980 --> 00:14:27.180
All right.
00:14:27.440 --> 00:14:31.320
Another announcement this week was that Flet has gone 1.0.
00:14:31.640 --> 00:14:39.040
So if you've not checked it out, it is a Flutter-backed app written in pure Python with no front-end experience needed at all.
00:14:39.220 --> 00:14:48.800
So you can basically write a web app, a mobile app for iOS and Android, because Flutter is basically this cross-platform framework that allows you to do all that.
00:14:49.200 --> 00:14:52.220
And you get stuck and run on macOS, Linux, and the web.
00:14:52.760 --> 00:14:53.160
Cool.
00:14:53.520 --> 00:14:55.260
I've never actually tried it until today.
00:14:55.480 --> 00:15:05.480
I went and built a little quick demo from a previous app I had built and added a web app basically to it by saying, use Flutter, make a web app, and then launch it so I can play with it.
00:15:05.740 --> 00:15:08.960
And so you get one code base that can target many platforms.
00:15:09.220 --> 00:15:11.040
That's more the promise from Flutter.
00:15:11.240 --> 00:15:22.460
The promise from Flutter is you get to do it all just with Python and you don't have to do any Java or Kotlin or, you know, JavaScript front end pieces to it.
00:15:22.680 --> 00:15:37.060
That's a tall order. There's been a couple of different projects over the years that have attempted to do what this says, but this gives you 150 plus built-in UI controls plus support for some custom controls, all wrapping native Flutter packages.
00:15:37.600 --> 00:15:45.520
The mobile version now supports real Python packages. So you can go out real Python packages that might include things like C or Rust dependencies.
00:15:46.000 --> 00:15:52.480
So NumPy, Pandas, Pillow, Cryptography, those kinds of things can actually be used as part of Flut 1.0 now.
00:15:53.020 --> 00:16:01.420
It has a MCP server that you can use as part of the documentation to give your agent knowledge about how to build a Flut app, which is kind of nice.
00:16:01.780 --> 00:16:04.540
And I think it also has skills built into the repo.
00:16:04.900 --> 00:16:11.660
So if you have your agent set up correctly, it can use the skills that come with Flet to build fancy apps very, very quickly.
00:16:11.980 --> 00:16:14.480
So it had its first released back in 2022.
00:16:15.740 --> 00:16:20.160
This kind of 1.0 is signaling that we're production ready and not experimental.
00:16:20.320 --> 00:16:23.320
So actually, if you look at the, they have a blog post released on it today.
00:16:23.880 --> 00:16:26.640
Flet.1.0 is out, or no, it's not today, it's from the 15th.
00:16:26.720 --> 00:16:29.300
So Flet.1.0 is out, means it's production ready to go.
00:16:29.400 --> 00:16:31.480
So it has a CLI and a library.
00:16:31.950 --> 00:16:34.500
So you install the library to be able to use it to build your app in Flet.
00:16:34.690 --> 00:16:38.540
You install the CLI so you can use things like the Flet build, Flet test, Flet run.
00:16:39.520 --> 00:16:40.760
And away you go.
00:16:41.140 --> 00:16:42.960
I also, here's like, I built a little calendar.
00:16:43.250 --> 00:16:45.400
So this is all using Python under the covers.
00:16:46.080 --> 00:16:46.760
And it's a real web app.
00:16:46.930 --> 00:16:49.120
You can see you're running on my framework here in the house.
00:16:49.560 --> 00:16:50.140
Yeah, it looks good.
00:16:50.340 --> 00:16:50.540
Yeah.
00:16:50.870 --> 00:16:56.260
I mean, those are all, I did nothing other than port Flet and told my agent to wrap the CLI app.
00:16:56.320 --> 00:16:58.940
This is all a CLI app I'd written to make a little calendar.
00:16:59.280 --> 00:17:01.700
and told it
to give it next and previous buttons.
00:17:02.260 --> 00:17:06.780
And I wrote no JavaScript is harmed in the creation of this quick little app.
00:17:07.020 --> 00:17:08.380
No semicolons were dropped.
00:17:08.760 --> 00:17:09.680
No, definitely not.
00:17:10.380 --> 00:17:11.540
But I definitely consider it.
00:17:11.880 --> 00:17:14.260
We've had some success with Flutter projects in the past.
00:17:14.600 --> 00:17:20.680
I wasn't happy about the Flutter part, but now I could maybe tolerate it a little better if I'm not having to deal with it because Flutter deals with it for me.
00:17:20.800 --> 00:17:24.880
I thought a lot about to write the Talk Python courses app in, and I ended up choosing Flutter.
00:17:25.319 --> 00:17:26.860
And this is Dart, not Flutter.
00:17:27.120 --> 00:17:27.800
It's been great.
00:17:27.880 --> 00:17:28.720
It's been going for a couple of years.
00:17:28.900 --> 00:17:29.460
I really like it.
00:17:29.940 --> 00:17:30.580
It's performant.
00:17:30.720 --> 00:17:31.440
It feels native.
00:17:31.900 --> 00:17:32.440
It feels native.
00:17:32.720 --> 00:17:33.720
Yeah, I like that.
00:17:34.040 --> 00:17:35.400
Yeah, I just built a new version.
00:17:35.860 --> 00:17:36.080
Sorry, go ahead.
00:17:36.320 --> 00:17:42.400
Yeah, I would say when people get really real about it, I think you'll want to do native versions, but I think until then, Flutter is a great option.
00:17:42.620 --> 00:17:42.660
I mean,
00:17:42.820 --> 00:17:43.780
there's native and then there's native.
00:17:43.900 --> 00:17:45.840
You know, like how native does it have to be?
00:17:46.000 --> 00:17:47.100
Like, we're right in the support lane.
00:17:47.260 --> 00:17:47.460
That's true.
00:17:47.820 --> 00:17:48.000
Right?
00:17:48.660 --> 00:17:53.400
But still, this, I just pushed a new version to the App Store last week.
00:17:53.780 --> 00:17:54.400
So, I don't know.
00:17:54.840 --> 00:17:58.600
All the App Store review times are getting so bad with so many apps being submitted to them.
00:17:59.060 --> 00:18:00.620
I mean, Apple can definitely, yeah.
00:18:01.030 --> 00:18:02.440
I don't mind the new apps.
00:18:02.620 --> 00:18:07.940
It's just Apple, especially, they just, they don't have the money to hire enough reviewers.
00:18:08.260 --> 00:18:09.960
It's really unfortunate, those poor souls.
00:18:10.360 --> 00:18:11.620
No, they just don't hire enough people.
00:18:11.720 --> 00:18:12.280
It's really frustrating.
00:18:12.290 --> 00:18:14.280
Like my review took a week or something like that.
00:18:14.450 --> 00:18:16.580
And like, that's not normal for an app that's already out there.
00:18:16.590 --> 00:18:18.380
For one that's already published, it's definitely not normal.
00:18:18.540 --> 00:18:19.140
Yeah, yeah, exactly.
00:18:20.100 --> 00:18:24.080
Anyway, the download for the Delta was three megs for the new version.
00:18:24.190 --> 00:18:25.100
So it was really, really good.
00:18:25.340 --> 00:18:26.120
All right, so
there's that.
00:18:26.200 --> 00:18:31.700
And then also I had, Fidor Fitzner, the guy behind flight on the show, not too long ago.
00:18:31.790 --> 00:18:34.480
So people want to check that out, but I want to talk about books.
00:18:35.040 --> 00:18:37.020
I want to talk about Marimo books.
00:18:37.260 --> 00:18:47.680
So if you wanted to make a book, let's imagine you want to use not exactly Jupyter, but you want to use Marimo Marimo, and then you wanted to generate a book from it.
00:18:47.960 --> 00:18:48.820
This is pretty cool actually.
00:18:49.180 --> 00:18:50.700
So I like Marimo a lot.
00:18:50.700 --> 00:18:51.040
I don't know.
00:18:51.100 --> 00:18:51.900
You said you do as well.
00:18:52.110 --> 00:18:52.500
I do too.
00:18:52.920 --> 00:18:52.980
Yeah.
00:18:53.280 --> 00:18:53.400
Yeah.
00:18:53.700 --> 00:18:57.460
It's got a really nice UI, really nice feel on what it does.
00:18:57.480 --> 00:19:02.680
But so they brought the guy behind any widgets and now works at Marimo.
00:19:03.060 --> 00:19:11.280
And so you can imagine there's really good support in Marimo now for any widgets, which is interchangeable ways to have widgets for the printer and Marimo and others, I believe.
00:19:11.640 --> 00:19:15.040
And you can have interactive sliders in your book.
00:19:15.400 --> 00:19:17.260
You can either pre-render it or whatever.
00:19:17.880 --> 00:19:21.380
Or if you want to do it online, you can add, here we go, here's our tie-in.
00:19:22.100 --> 00:19:23.800
Wasm render per chapter.
00:19:24.010 --> 00:19:26.100
So you can just publish this right up to Cloudflare.
00:19:26.560 --> 00:19:26.680
Yeah.
00:19:26.910 --> 00:19:27.000
Yeah.
00:19:27.090 --> 00:19:29.920
You could just take this and say, here's your, here's your book.
00:19:30.100 --> 00:19:33.480
And by the way, this part has a cool physics animation that I want to show you.
00:19:33.840 --> 00:19:38.460
So online, we have the Wasm version with all the dynamic aspect of it.
00:19:38.580 --> 00:19:39.500
Material for MK docs.
00:19:39.670 --> 00:19:42.780
They mean, I wonder if it'll take on Sensicle eventually, but who knows?
00:19:43.060 --> 00:19:44.960
You know, cause that's kind of the successor.
00:19:45.800 --> 00:19:47.640
Martin and crew over there are working on that.
00:19:48.100 --> 00:19:56.340
Launch buttons per chapter for launch on GitHub or on Molab, which is the Marimo online cloud hosted lab for notebooks.
00:19:56.820 --> 00:19:57.920
Incremental build cache.
00:19:58.000 --> 00:20:01.380
So stuff goes nice and speedy and goes straight to GitHub pages.
00:20:01.680 --> 00:20:02.060
What do you think?
00:20:02.320 --> 00:20:02.980
This looks really cool.
00:20:03.600 --> 00:20:03.960
Yeah, it does.
00:20:04.260 --> 00:20:05.420
Yeah, bundle onto Marimo.
00:20:05.560 --> 00:20:06.000
I love it.
00:20:06.160 --> 00:20:11.520
This is from LJ Ching, not from the Marimo team.
00:20:11.860 --> 00:20:15.580
So it's an emerging project, but it looks neat to me.
00:20:16.200 --> 00:20:18.660
And so I thought I'd go ahead and shine a bit of a light on it nonetheless.
00:20:19.060 --> 00:20:28.920
I just love the idea of this being maybe the future of education, where you have a book that is fully interactive and the code works and it's published online and it works with the whole sandbox and everything.
00:20:29.120 --> 00:20:30.540
It's just, that's, that's what's needed.
00:20:30.740 --> 00:20:31.160
I love it.
00:20:31.380 --> 00:20:31.720
That's really cool.
00:20:32.300 --> 00:20:32.520
Absolutely.
00:20:33.559 --> 00:20:37.480
Michael, do you have any, I got, I got one extra for us for this go around.
00:20:37.640 --> 00:20:38.000
Run with it.
00:20:38.380 --> 00:20:38.460
Yeah.
00:20:39.080 --> 00:20:45.300
If you have not checked out, Python 315 has a new data type in it called frozen dict.
00:20:45.740 --> 00:20:50.920
So I wanted to just highlight this link over to the RealPython site on their overview of it.
00:20:50.930 --> 00:20:52.120
They did a nice overview.
00:20:52.660 --> 00:20:55.700
What's nice about frozen dicts is that they are immutable.
00:20:56.120 --> 00:21:00.720
And they basically line up to their list and set counterparts that have been around for years.
00:21:00.950 --> 00:21:06.740
It also means you can use a frozen dict as a key in a regular dict because it is hashable.
00:21:07.060 --> 00:21:07.900
Oh, interesting.
00:21:08.200 --> 00:21:08.500
Yes.
00:21:08.680 --> 00:21:11.140
I would have never thought of using a dictionary as a key.
00:21:11.820 --> 00:21:12.900
It already blows my mind.
00:21:13.080 --> 00:21:15.760
You can use a tuple as a key and for sorting and stuff.
00:21:16.020 --> 00:21:16.660
Because it's immutable.
00:21:17.060 --> 00:21:25.420
But yeah, so that was just a quick little one I wanted to throw in there because I think given the excitement that's now brewing around 315, I can't wait.
00:21:25.660 --> 00:21:26.420
Yeah, same here.
00:21:26.840 --> 00:21:34.080
And yeah, just another reason to be excited about 315, which by the way, I don't know if my calendar exactly pulled up, but I believe that is next week.
00:21:34.360 --> 00:21:36.320
October 1st is when that comes out per schedule.
00:21:36.680 --> 00:21:37.320
So that's great.
00:21:37.580 --> 00:21:38.700
All right, now I have a couple.
00:21:39.220 --> 00:21:39.760
Yeah, exactly.
00:21:40.300 --> 00:21:45.700
A couple of things just to better noteworthy here, I thought maybe worth chatting about a bit.
00:21:46.040 --> 00:21:51.040
One, we went back 10 years and there was an update from Microsoft for Windows or something like that.
00:21:51.300 --> 00:21:52.660
There'd be but what, 20 patches?
00:21:53.240 --> 00:21:53.680
Yeah, maybe.
00:21:54.620 --> 00:21:57.860
This year is on track to have just more patches.
00:21:58.420 --> 00:21:58.840
So many more.
00:21:58.910 --> 00:22:04.400
I think last month was a record-setting month with 500 security patches for Windows.
00:22:04.940 --> 00:22:05.780
This year is 1,000.
00:22:05.840 --> 00:22:07.460
Sorry, this month is 1,000.
00:22:07.600 --> 00:22:09.540
That's 2x the max.
00:22:09.740 --> 00:22:12.000
of last month, which was an all-time record.
00:22:12.440 --> 00:22:19.020
This is a serious endorsement, I guess, of using AI to find and solve security problems.
00:22:19.560 --> 00:22:23.580
You know, Microsoft has this MDash program, which is kind of like their own, like, we're going to build our own glass wing.
00:22:23.820 --> 00:22:24.800
Thank you very much sort of thing.
00:22:25.180 --> 00:22:25.900
So I don't know.
00:22:25.910 --> 00:22:27.800
I think this is just noteworthy.
00:22:27.830 --> 00:22:38.020
And if you also look at the Golden Gate macOS 27, there was just a wall of little tiny, this is faster, this has improved, that's better, things that kind of mirrored a similar thing.
00:22:38.160 --> 00:22:45.140
So I think, you know, people ask, where is the explosion of software if actually have increased productivity with AI?
00:22:45.400 --> 00:22:45.500
Well
00:22:45.920 --> 00:22:46.160
Tech debt.
00:22:47.180 --> 00:22:47.660
Yeah, exactly.
00:22:47.820 --> 00:22:49.440
We're still digging out of the holes.
00:22:49.840 --> 00:22:54.080
Anyway, I just wanted to throw this out here, not to bash on Microsoft. This is actually a good thing. - No,
00:22:54.080 --> 00:22:54.760
I think it's a good thing.
00:22:54.980 --> 00:23:03.480
I worry though still that for everyone they patched, how many more did we create along the way? Hopefully less than they're being very, they obviously are cognizant of that problem.
00:23:03.800 --> 00:23:06.480
Yeah, there was a problem they introduced with Excel.
00:23:06.940 --> 00:23:09.640
I think they broke copy and paste and they quickly fixed it.
00:23:09.960 --> 00:23:11.980
But there's not a new security problem, right?
00:23:12.110 --> 00:23:15.240
I mean, if you freeze the software, it's been around for 30 years.
00:23:15.480 --> 00:23:17.300
There's a finite set of problems.
00:23:17.880 --> 00:23:20.320
And the faster we can plow through them, the better, generally.
00:23:20.420 --> 00:23:23.380
As long as the attackers don't outrun the white hats.
00:23:23.680 --> 00:23:25.400
Anyway, okay, so on to the next.
00:23:25.920 --> 00:23:29.460
The next thing is I am speaking of beginning of October.
00:23:29.550 --> 00:23:34.340
On October 3rd, I will be speaking at, right here I am.
00:23:34.340 --> 00:23:34.940
Who is this character?
00:23:35.360 --> 00:23:36.440
I will be speaking at Friday.
00:23:36.440 --> 00:23:36.800
I recognize all of those.
00:23:36.820 --> 00:23:36.880
folks.
00:23:37.120 --> 00:23:38.620
Yes, I didn't know as well.
00:23:38.960 --> 00:23:45.100
We'll go back to the list a second, but I'm basically doing a updated newer version based on my book, Talk Python in production.
00:23:45.380 --> 00:23:59.940
So how can you pick a simpler view of running and hosting your apps and not spend
a ton of money, have it simple enough that one person can run it in their very much part-time efforts and yeah, still get a good outcome. So that's what the talk is about. I'm excited about that.
00:24:00.240 --> 00:24:04.240
And we also have, let's see, Himalafox is going to be there. Awesome.
00:24:05.220 --> 00:24:05.860
Brian Okken.
00:24:06.580 --> 00:24:08.160
People may know him from the show. I'm not sure.
00:24:08.200 --> 00:24:08.280
Maybe.
00:24:10.160 --> 00:24:13.460
So anyway, I hope to see you all there for some of you.
00:24:13.720 --> 00:24:17.520
And I also want to give a quick shout out to PyCon Netherlands.
00:24:18.560 --> 00:24:28.280
So this was sent over by Gareth together and just wanted me to give a quick bit of background for people. So if you're headed to Europe in October, the good stuff is happening in October.
00:24:28.400 --> 00:24:30.100
This is October 15th in Utrecht.
00:24:30.400 --> 00:24:35.580
So you can check it out. It's one day, three tracks, about 150 people and it's an hour by train from Scheifell.
00:24:36.020 --> 00:24:40.920
There's a session for just for community organizers from groups like PyLadies, PyData and Django.
00:24:41.520 --> 00:24:45.960
And yeah, this is a good place because this is where Python was born in the Netherlands.
00:24:46.280 --> 00:24:47.920
So go back to its roots.
00:24:48.240 --> 00:24:50.000
Sometimes it only makes sense if you're Dutch.
00:24:50.940 --> 00:24:51.320
That's right.
00:24:51.760 --> 00:24:52.920
Well, I'll tell you what makes sense.
00:24:53.320 --> 00:24:54.700
You gotta be German for this to make sense.
00:24:54.720 --> 00:24:55.060
Oh, okay.
00:24:55.220 --> 00:24:55.400
Yeah?
00:24:55.600 --> 00:24:55.880
Yeah?
00:24:56.320 --> 00:24:57.520
You got a joke for me?
00:24:57.840 --> 00:24:58.620
I have homework.
00:24:59.100 --> 00:24:59.880
I have homework for you.
00:24:59.920 --> 00:25:02.220
And I know that you've done your homework you are a good student.
00:25:02.850 --> 00:25:03.640
You are a good student.
00:25:04.000 --> 00:25:08.160
So this comes to us from Kyle Lentit, otherwise known as programmers are human.
00:25:08.320 --> 00:25:10.220
And oh my gosh, has he been killing it lately.
00:25:10.590 --> 00:25:13.820
So this is actually one of two things I want to recommend to people.
00:25:14.100 --> 00:25:19.640
And I'm not going to spoil it. We're going to come back to it. But there's a little hint right down here, the bottom right.
00:25:19.840 --> 00:25:28.160
There's another video that I think is even better than this, but this is what the one I grabbed for the show called interview with a big data engineer in 2026.
00:25:28.600 --> 00:25:32.180
I love how the intentional like gaps of like wording and stuff like.
00:25:33.520 --> 00:25:35.740
So this is a video, but how long is it?
00:25:36.080 --> 00:25:36.600
It's eight minutes.
00:25:36.800 --> 00:25:37.980
You should all go watch it.
00:25:38.100 --> 00:25:38.980
It's very funny.
00:25:39.080 --> 00:25:40.520
You will spend on YouTube.
00:25:41.680 --> 00:25:46.420
There's some great one-liners and some serious gold embedded into this video.
00:25:47.200 --> 00:25:49.420
Like eventual consistency is hard.
00:25:49.600 --> 00:25:51.240
So we went with immediately inaccurate.
00:25:51.640 --> 00:25:52.960
One of my favorites from that video.
00:25:53.480 --> 00:25:53.560
Yeah.
00:25:53.740 --> 00:25:55.440
and 70% of the records are accurate.
00:25:55.640 --> 00:25:57.000
That's better than a weather forecast.
00:26:00.100 --> 00:26:07.480
And Kai is just, God, he is good at these jokes, but he's really, really informed in just like the broader space of programming.
00:26:08.030 --> 00:26:12.020
It sounds so real because many of us have experienced these things.
00:26:12.380 --> 00:26:13.720
Yeah, yeah, it really is good.
00:26:14.240 --> 00:26:16.220
And he has another good little section here.
00:26:16.480 --> 00:26:19.080
Salary, job security, holidays, pick two.
00:26:20.140 --> 00:26:21.500
High salary, job security.
00:26:23.320 --> 00:26:24.620
You can have any two of those you want.
00:26:26.180 --> 00:26:30.620
Anyway, yeah, it's just like, I'm just clicking around at every little place I click on this video.
00:26:30.680 --> 00:26:31.760
It's like, yeah, that part was awesome.
00:26:32.180 --> 00:26:32.920
So check it out.
00:26:32.980 --> 00:26:33.580
This is my joke.
00:26:33.880 --> 00:26:34.780
It's not a direct joke.
00:26:34.980 --> 00:26:37.320
It's homework, but trust me, you will enjoy it.
00:26:37.400 --> 00:26:37.720
Click the link.
00:26:37.900 --> 00:26:38.380
It'll be worth it.
00:26:38.580 --> 00:26:38.900
It will.
00:26:39.100 --> 00:26:41.160
And we'll come back to more stuff from Kai later.
00:26:42.100 --> 00:26:42.560
All right, Michael.
00:26:42.660 --> 00:26:45.120
Well, thank you so much again for this week's edition.
00:26:45.340 --> 00:26:46.880
And we will see you all next week.
00:26:47.140 --> 00:26:47.500
You bet.
00:26:47.720 --> 00:26:47.820
Goodbye.