WEBVTT 00:00:00.000 --> 00:00:05.360 Hello and welcome to Python Bytes, where we deliver Python news and headlines directly to your earbuds. 00:00:05.820 --> 00:00:10.700 This is episode 471, recorded March 2nd, 2026. 00:00:11.100 --> 00:00:12.420 It just still sounds weird. 00:00:12.800 --> 00:00:14.160 And I am Brian Okken. 00:00:14.400 --> 00:00:15.280 I am Michael Kennedy. 00:00:15.960 --> 00:00:21.000 And this episode is sponsored by all of you guys and us. 00:00:21.000 --> 00:00:32.060 So everybody that helps send some money our way through taking a course at Talk Python Training or a pytest course at pythontest.com or reads one of our books. 00:00:32.920 --> 00:00:34.480 We're also Patreon supporters. 00:00:34.620 --> 00:00:35.640 Thank you, Patreon supporters. 00:00:35.980 --> 00:00:42.220 If you would like to send out, we get a lot of, we do a lot of research ourselves, but we also get a lot of tips from other people. 00:00:42.220 --> 00:00:51.700 So if you'd like to have feedback, if you disagree with something or want to add to the conversation, go ahead and connect with us on BlueSky, Mastodon. 00:00:52.460 --> 00:00:54.080 I think that's it. 00:00:54.480 --> 00:00:56.840 Or just at pythonbytes.fm. 00:00:56.900 --> 00:00:57.940 There's a contact form there. 00:00:58.540 --> 00:01:06.440 You can also subscribe to the, actually one of the best ways is you can go over to pythonbytes.fm and subscribe to the email list. 00:01:06.720 --> 00:01:12.040 And when you get an email, then you'll know our email address and then you can just send us stuff through there. 00:01:12.680 --> 00:01:13.700 That gets to us. 00:01:14.080 --> 00:01:17.660 Also, if you'd like to join the show live, that's fun. 00:01:17.960 --> 00:01:24.760 Go ahead over to pythonbytes.fm/live and you can find out how to get, watch the show live. 00:01:25.060 --> 00:01:28.140 With that, let's kick it off with the first topic, Michael. 00:01:28.380 --> 00:01:29.240 Let's do it. 00:01:29.500 --> 00:01:36.820 So this first topic is an article that I wrote, more of a design pattern that I would like to advocate for. 00:01:36.960 --> 00:01:37.400 Okay. 00:01:37.400 --> 00:01:39.800 Now, you see, Brian, I have this friend. 00:01:40.220 --> 00:01:41.080 His name is Claude. 00:01:41.080 --> 00:01:43.980 He's really good at writing code, but he gets super distracted. 00:01:44.680 --> 00:01:45.120 Claude code. 00:01:45.400 --> 00:01:55.300 So I've been thinking a lot about what does agentic programming and people leveraging these tools mean for like design patterns and architecture. 00:01:55.520 --> 00:01:57.900 I'm a big fan of thinking about architecture and so on. 00:01:57.900 --> 00:01:58.940 So I wrote this up. 00:01:59.000 --> 00:02:04.340 It got super popular on Reddit and I thought I would share it here as well. 00:02:04.860 --> 00:02:11.300 You know, popular in the sense that like lots of comments, lots of upvotes, but also a lot of like mean comments because it's Reddit. 00:02:11.460 --> 00:02:12.360 So that's how it goes. 00:02:12.360 --> 00:02:14.440 But good comments too. 00:02:14.640 --> 00:02:17.000 First of all, I want to put this out as a disclaimer. 00:02:17.180 --> 00:02:23.020 As you think about this, don't try to adjudicate whether or not people should be using AI for coding. 00:02:23.180 --> 00:02:28.940 Don't try to adjudicate whether AI coding should exist, whether it is bad or good. 00:02:29.020 --> 00:02:29.760 It exists. 00:02:29.980 --> 00:02:30.740 People are using it. 00:02:30.740 --> 00:02:34.460 If that's true, what should we do as an industry? 00:02:34.700 --> 00:02:36.400 How should we consider, how should we move forward? 00:02:36.460 --> 00:02:37.500 How should we think about this, right? 00:02:37.520 --> 00:02:39.960 Are there things that would make that better or worse, right? 00:02:40.000 --> 00:02:41.440 You yourself can decide you're not using it. 00:02:41.460 --> 00:02:44.580 But as an industry, I think it's pretty clear that it's being highly adopted. 00:02:44.760 --> 00:02:48.260 So that's kind of the goal that I had when I wrote this article. 00:02:48.580 --> 00:02:48.800 Yeah. 00:02:49.060 --> 00:02:52.400 And so the title is Raw Plus DC. 00:02:52.740 --> 00:02:57.620 So as in Raw Database Queries and DC is in Data Classes. 00:02:57.620 --> 00:03:01.540 And the subtitle is The ORM Pattern of 2026? 00:03:02.560 --> 00:03:06.060 And I'm putting that out there because I think this is a pattern that is much, 00:03:06.260 --> 00:03:08.880 much more aligned to the way AI coding works. 00:03:09.200 --> 00:03:10.380 Why do people like ORMs? 00:03:10.500 --> 00:03:14.680 And technically for this, I use what's called an ODM, an Object Document Mapper, 00:03:14.800 --> 00:03:18.880 because I was using the example I extracted it from was written against MongoDB. 00:03:18.880 --> 00:03:23.520 But whatever, like it's immaterial, whether it's Mongo or Postgres, okay? 00:03:23.720 --> 00:03:24.460 Or SQLite. 00:03:24.460 --> 00:03:30.540 People, developers love ORMs and ODMs because they provide type safety, 00:03:30.740 --> 00:03:33.020 especially with like document databases. 00:03:33.020 --> 00:03:37.400 They constrain what can and cannot go into the database, right? 00:03:37.540 --> 00:03:40.140 You got to match the class and then the class goes in the database. 00:03:40.260 --> 00:03:43.320 You don't just stick dictionaries in there, which is otherwise the case. 00:03:43.600 --> 00:03:48.800 They prevent SQL injection or Mongo query injection because they always, 00:03:49.360 --> 00:03:51.500 by the way they work, use parameterized queries. 00:03:51.860 --> 00:03:52.740 They're a nice abstraction. 00:03:52.740 --> 00:03:56.860 Also for developers, they provide really good IDE auto-completion. 00:03:57.160 --> 00:03:57.940 So I have a function. 00:03:58.240 --> 00:04:02.960 I give it a type or even the type is just discovered through type inference that it returns this class. 00:04:03.080 --> 00:04:06.480 When I get one back and I say thing from the database dot, 00:04:06.580 --> 00:04:08.700 it gives me a list of all the stuff I can do in the database. 00:04:08.880 --> 00:04:10.000 All these things are awesome, right? 00:04:10.360 --> 00:04:10.460 Yeah. 00:04:10.460 --> 00:04:11.760 But here's the thing. 00:04:12.480 --> 00:04:16.940 I ran to this literally this morning working on this project that I've not yet released, 00:04:17.020 --> 00:04:18.120 but we'll talk about pretty soon. 00:04:18.420 --> 00:04:27.160 These AI tools are trained so much more on raw queries than they are on any particular ORM. 00:04:27.160 --> 00:04:30.040 So for example, I was using Beanie on this project. 00:04:30.400 --> 00:04:34.500 Beanie is downloaded 1.4 million times a month, which actually is really impressive. 00:04:34.860 --> 00:04:36.820 So Beanie is the ODM for talking to MongoDB. 00:04:37.160 --> 00:04:44.440 The underlying library that lets you talk in the native query syntax is downloaded 74 million times per month. 00:04:44.660 --> 00:04:45.940 That's 53 times more. 00:04:45.940 --> 00:04:49.880 And there's other libraries for talking directly to Mongo as well, right? 00:04:49.980 --> 00:05:00.020 So there's like 50, 75 times more bits of code out there that are actually just talking directly to the database than using this particular ORM. 00:05:00.360 --> 00:05:00.520 Okay. 00:05:00.700 --> 00:05:02.300 And that's true for SQL as well. 00:05:02.440 --> 00:05:04.700 Like for just sticking with Mongo, right? 00:05:04.840 --> 00:05:13.700 PyMongo is the Python version of this, but there's also people doing the native query syntax in Node, Deno, Bun, PHP, .NET, whatever, right? 00:05:13.700 --> 00:05:15.000 Like all these other languages. 00:05:15.000 --> 00:05:27.140 So if you actually do that sort of cross product, it's probably a thousand times more that AIs see raw database queries than they see any particular ORM, even if that ORM is popular, right? 00:05:27.180 --> 00:05:32.940 Because it's for a technology and then it's sliced down into who uses that ORM rather than across all programming. 00:05:33.200 --> 00:05:40.480 The other thing is, so these ORMs go unmaintained, right? 00:05:40.480 --> 00:05:46.340 I mean, Django ORM probably won't go unmaintained, but they come and they go in sort of their popularity. 00:05:46.480 --> 00:05:51.340 So for example, Beanie, it had only like three releases in the entire year of 2025. 00:05:51.840 --> 00:05:54.840 Its last release was four months ago to close some minor issue. 00:05:54.840 --> 00:06:01.740 And it's got a bunch, like almost a hundred open issues, even though it's downloaded over a million times a month, it's kind of gone unmaintained. 00:06:01.840 --> 00:06:05.320 And before that I was using Mongo engine, which is way more unmaintained than Beanie. 00:06:05.400 --> 00:06:06.980 Like Beanie was the exciting new one, right? 00:06:07.180 --> 00:06:09.680 And so what do I have to do if this changes? 00:06:09.760 --> 00:06:13.980 If I decide that it's too unmaintained, I have to completely rewrite it in a different ORM. 00:06:13.980 --> 00:06:17.940 But if it was written in a regular, just straight query syntax, that doesn't change, right? 00:06:18.000 --> 00:06:24.280 That's been identical for SQL since the databases came out, identical for Mongo since Mongo came out. 00:06:24.340 --> 00:06:25.620 I mean, they add to it a little bit, right? 00:06:25.660 --> 00:06:26.740 Like if they add a feature or something. 00:06:26.880 --> 00:06:29.140 But generally, it's going to be the same. 00:06:29.480 --> 00:06:30.860 So I don't have to worry about it. 00:06:30.920 --> 00:06:31.720 These things are going on the same. 00:06:31.840 --> 00:06:31.980 Go ahead. 00:06:32.160 --> 00:06:33.400 I want to interject a little bit. 00:06:33.500 --> 00:06:34.780 You threw out Django. 00:06:34.780 --> 00:06:46.740 I was going to comment that Django might be an outlier there in that a lot of, I know there are people that use other ORMs or raw SQL with Django. 00:06:46.900 --> 00:06:51.380 But I think it's probably a lion's share of the Django users that use the Django ORM. 00:06:51.660 --> 00:06:54.080 And I don't think that's going away anytime soon. 00:06:55.120 --> 00:06:57.200 I totally 100% agree with you. 00:06:57.340 --> 00:07:01.580 But what if you decide you want to change your project to a different web framework? 00:07:01.580 --> 00:07:17.400 Well, I would probably say the same argument is that you could probably switch to this raw SQL plus data classes from Django, the Django ORM, because there's a ton of Django stuff out there. 00:07:17.480 --> 00:07:20.000 I'm pretty sure that the models are trained on lots of Django. 00:07:20.300 --> 00:07:21.620 Yeah, they were pretty well trained. 00:07:22.040 --> 00:07:24.540 So this maintained argument is not go for everything. 00:07:24.680 --> 00:07:27.340 But it's definitely another angle, right? 00:07:27.340 --> 00:07:37.840 Yeah, and also, I think that you probably touched on this already, but just one less dependency is one less thing you have to keep up on. 00:07:38.100 --> 00:07:38.860 Yes, absolutely. 00:07:39.340 --> 00:07:42.160 And it's not usually just one dependency, right? 00:07:42.180 --> 00:07:45.020 It's like this thing depends on four others, which depend on three. 00:07:45.140 --> 00:07:46.620 And now there's a CVE in one of them. 00:07:46.700 --> 00:07:47.400 You're like, ah, man. 00:07:47.740 --> 00:07:52.140 Okay, so let's – but there's reasons people like these ORMs. 00:07:52.260 --> 00:07:56.040 I am a huge fan of ORMs because they give you back classes that are great. 00:07:56.040 --> 00:08:05.960 So my proposed way of working is create a nice data access layer, put functions in there that are very clear, like create user, update password, whatever. 00:08:06.340 --> 00:08:10.040 And then inside, let it be raw queries. 00:08:10.420 --> 00:08:12.480 Let Claude write however Claude wants. 00:08:12.780 --> 00:08:15.660 But when stuff comes out of there, it has to return some kind of class. 00:08:15.780 --> 00:08:17.740 So I propose data classes, okay? 00:08:18.060 --> 00:08:22.240 And I think data classes are pretty interesting here for a couple reasons. 00:08:22.240 --> 00:08:23.860 One, they're typed, right? 00:08:23.940 --> 00:08:26.000 They look like – you can tell them use slots. 00:08:26.160 --> 00:08:27.760 That's just good for memory and performance. 00:08:28.000 --> 00:08:32.620 They're typed just like if you use Pydantic or if you use any ORM generally. 00:08:33.260 --> 00:08:34.340 So that's really good. 00:08:34.620 --> 00:08:37.140 They're also built in, so no dependencies, as you pointed out. 00:08:37.200 --> 00:08:39.620 And they're very, very lightweight, which is great. 00:08:39.900 --> 00:08:44.240 Now, they have type safety in the terms of like type checking type safety. 00:08:44.240 --> 00:08:47.540 But they don't have runtime type safety like, say, Pydantic, right? 00:08:47.780 --> 00:08:47.960 Yeah. 00:08:48.200 --> 00:08:53.660 So if you want – another very reasonable option would be to do a Pydantic plus raw queries. 00:08:53.940 --> 00:08:58.960 I didn't use Pydantic because this means every query that comes back from the database gets validated. 00:08:59.140 --> 00:09:04.080 And there's a whole bunch of machinery that runs when you're reading from the database. 00:09:04.240 --> 00:09:07.340 But if the data in the database is wrong, it's already too late. 00:09:07.620 --> 00:09:09.340 It doesn't matter if your class says it's too bad. 00:09:09.400 --> 00:09:10.240 It's like, no, it's saved. 00:09:10.300 --> 00:09:10.880 It's too bad. 00:09:10.900 --> 00:09:11.360 It's too late. 00:09:11.360 --> 00:09:16.060 And so maybe on the outside layer, you could use Pydantic, but in the data access layer, you could use something lighter. 00:09:16.200 --> 00:09:17.760 That was my – that's my take. 00:09:18.240 --> 00:09:22.600 When I posted this on Reddit, people probably like 50 times like, well, why aren't you just using Pydantic? 00:09:22.780 --> 00:09:24.020 It's like, well, Pydantic is great. 00:09:24.080 --> 00:09:25.440 It doesn't have to be used for everything. 00:09:25.820 --> 00:09:28.940 So anyway, I'm going to follow up with one more thing that's pretty cool around that. 00:09:29.320 --> 00:09:36.940 But yeah, so if you look at the performance really quick here as well, I did a nice graph of a bunch of different things. 00:09:37.000 --> 00:09:38.400 What's like 1,000 reads look like? 00:09:38.440 --> 00:09:40.060 What does 1,000 writes and so on? 00:09:40.060 --> 00:09:49.880 And this graph is based on doing just raw queries without any class component because that's really the foundational – like they all have to do that, right? 00:09:50.060 --> 00:09:52.000 So how much overhead does each style add? 00:09:52.260 --> 00:10:00.420 And it's not exactly perfect because this raw data class style can't be faster than writing itself. 00:10:00.580 --> 00:10:02.460 In a lot of places, it's literally the same code. 00:10:02.500 --> 00:10:04.320 So it's just like fluctuations in the machine. 00:10:04.320 --> 00:10:09.720 But if you look at, say, Beanie, which is based on Pydantic, so this is more or less just Pydantic. 00:10:10.040 --> 00:10:11.900 What would Pydantic look like in this model? 00:10:13.060 --> 00:10:16.680 It's a lot of times pretty close, but sometimes it's like five times slower. 00:10:16.820 --> 00:10:23.100 If you look at Mongo Engine, which is another one, 15 times slower, 10 times slower, it's a big difference. 00:10:23.360 --> 00:10:24.540 It's quite a big difference. 00:10:24.540 --> 00:10:28.480 So this style is actually almost native speed. 00:10:28.640 --> 00:10:35.720 There's just a few times where the serialization maybe doubles the time to read, say, 1,000 or 100 orders or something like that, but barely. 00:10:36.080 --> 00:10:37.120 Anyway, I think it's pretty interesting. 00:10:37.340 --> 00:10:37.720 What do you think? 00:10:37.720 --> 00:10:49.600 Yeah, I think that one of the reasons to go with an ORM was speed of development because a lot of people don't know SQL queries that well. 00:10:49.920 --> 00:10:53.680 But a good SQL query is not that hard to read. 00:10:53.760 --> 00:10:54.500 Well, it can be. 00:10:54.620 --> 00:10:58.720 But generally, they can be decent to read. 00:10:59.020 --> 00:11:02.080 Trying to figure out how to write it might be tricky. 00:11:02.080 --> 00:11:10.920 So that's where I think your argument of around using agents and stuff to help with this, they're going to know how to write a decent one. 00:11:11.020 --> 00:11:16.200 And you can probably, even if it's ugly, you might be able to say, can you clean that up so it's easier to read? 00:11:16.820 --> 00:11:17.880 Yeah, or explain it to me. 00:11:18.120 --> 00:11:18.280 Yeah. 00:11:20.900 --> 00:11:22.940 But anyway, I think it's valid. 00:11:23.440 --> 00:11:24.580 It's definitely valid. 00:11:24.660 --> 00:11:28.200 I think it's more valid now than it has been in the last couple decades. 00:11:28.640 --> 00:11:28.980 So, yeah. 00:11:28.980 --> 00:11:32.160 Yeah, just this morning, I asked Claude to do this thing with Beanie. 00:11:32.500 --> 00:11:35.180 I said, oh, it would be way better if I just used a projection. 00:11:35.740 --> 00:11:41.280 And for projections in Beanie, you have to create a subclass or a different class and then use that. 00:11:41.340 --> 00:11:42.640 And it's like, oh, I could just use dictionaries. 00:11:42.700 --> 00:11:43.180 Then it crashed. 00:11:43.260 --> 00:11:44.300 Oh, I don't think I can do that. 00:11:44.340 --> 00:11:46.640 Let me try the aggregation pipeline through Beanie. 00:11:46.720 --> 00:11:47.320 And that didn't work. 00:11:47.400 --> 00:11:49.100 I'm just like, could you just do it directly? 00:11:49.160 --> 00:11:50.040 And it's like, oh, yeah, that works. 00:11:50.900 --> 00:11:53.000 It's just, you know, it's kind of a little bit clunky. 00:11:53.000 --> 00:12:00.920 One other thing that was really popular in the conversation about this article was what Kiva Birb is asking is, what about migrations? 00:12:01.040 --> 00:12:03.520 Not relevant to ODMs, but certainly relevant to ORMs. 00:12:03.900 --> 00:12:06.100 You do what everyone else does if you write straight SQL. 00:12:06.340 --> 00:12:08.040 You write your migrations yourself. 00:12:08.400 --> 00:12:13.360 But the thing is, like, my assumption is this pattern is most adapted to agentic coding. 00:12:13.360 --> 00:12:16.880 If you're just doing raw straight coding yourself, stick with ORMs. 00:12:17.200 --> 00:12:17.820 That'll handle it. 00:12:17.980 --> 00:12:21.840 But if you're doing agentic, like, Claude Code can write a heck of a migration. 00:12:22.020 --> 00:12:22.720 It doesn't care. 00:12:22.980 --> 00:12:23.940 It's going to be great at that. 00:12:24.180 --> 00:12:24.500 Yeah. 00:12:24.800 --> 00:12:28.760 I guess I've just sort of leaned on migrations a lot. 00:12:29.020 --> 00:12:32.080 And I actually don't really know what they do under the hood. 00:12:32.460 --> 00:12:33.480 Yeah, I mean, what are migrations? 00:12:33.640 --> 00:12:42.800 They are basically, they are a statically generated, like, pre-generated SQL script that says, add this field, change this field, change this type, copy these things. 00:12:42.800 --> 00:12:47.960 It's a set of SQL steps for transforming the schema of the database. 00:12:48.400 --> 00:12:57.060 And typically the way migrations work is it looks at the old version of the class and the new version of the class and goes, what do I got to do to write the SQL to make one go from one to the other? 00:12:57.420 --> 00:12:58.660 You can just ask Cloud to do that. 00:12:59.200 --> 00:13:08.200 I also really like the comment about have a data access layer, because I think that that's a good idea, even if you are using an ORM. 00:13:08.600 --> 00:13:09.200 Yeah, thanks. 00:13:09.920 --> 00:13:10.180 All right. 00:13:10.200 --> 00:13:11.880 Well, that is plenty on that. 00:13:11.880 --> 00:13:15.400 I mean, you should definitely test your code if you're going to take on this way of programming. 00:13:15.760 --> 00:13:16.200 Yeah. 00:13:16.500 --> 00:13:17.320 Nice transition. 00:13:17.660 --> 00:13:24.760 I've been spending some time in the last week cleaning up, giving some love to pytestCheck. 00:13:24.940 --> 00:13:31.220 So pytestCheck is a little plugin I wrote a really long time ago, many, many years ago. 00:13:31.220 --> 00:13:37.480 But the last release before recently was in November, I think. 00:13:38.000 --> 00:13:41.320 So yeah, October, November, something like that. 00:13:41.920 --> 00:13:45.720 So I've decided to spend some time getting it back up to date. 00:13:45.720 --> 00:13:53.240 And I did, I like all the stuff on GitHub, the activity, pulse activities thing. 00:13:53.640 --> 00:13:55.920 So there's been three merged pull requests. 00:13:56.280 --> 00:14:00.300 Sorry to everybody for taking so long, but it was only, I guess I only annoyed three people. 00:14:00.300 --> 00:14:05.440 And there is, yeah, I closed a bunch of issues, eight closed issues. 00:14:05.440 --> 00:14:13.260 Now, I did get to the point where I had only one enhancement request and no actual issues and no pull requests. 00:14:13.720 --> 00:14:18.240 That, you know, that doesn't last very long for projects that are being used. 00:14:18.240 --> 00:14:22.640 And I hadn't checked the little stats thing recently. 00:14:23.120 --> 00:14:28.840 And it's used by almost 1.3-ish thousand projects. 00:14:29.060 --> 00:14:29.980 That's fun. 00:14:30.980 --> 00:14:38.060 Anyway, I just wanted to bring it up that that's still in the live project. 00:14:38.520 --> 00:14:40.300 And I'm going to keep it up. 00:14:40.780 --> 00:14:41.740 I'm still having fun with it. 00:14:41.780 --> 00:14:42.860 I still use it every day. 00:14:42.960 --> 00:14:44.340 So that's one of the reasons. 00:14:44.340 --> 00:14:48.340 The changes, there's not a lot of extra features. 00:14:48.540 --> 00:14:55.160 There was a support for raises parameter when you're doing xfail. 00:14:55.580 --> 00:15:00.620 That works now, which that's really kind of a bug fix, but it was extra feature. 00:15:01.160 --> 00:15:02.680 So that was a pull request. 00:15:02.760 --> 00:15:03.460 So thank you. 00:15:03.920 --> 00:15:08.020 But some documentation things that I've been noticing this at work, too. 00:15:08.380 --> 00:15:12.480 If you have a new feature and you don't document it, it's like invisible work. 00:15:12.520 --> 00:15:13.740 Why did you do that work? 00:15:13.740 --> 00:15:16.220 So I added some documentation things. 00:15:17.000 --> 00:15:22.800 But a lot of it was around my Pi because a lot of people are using my type checkers and 00:15:22.800 --> 00:15:24.000 using Pi test check. 00:15:24.600 --> 00:15:26.900 So I updated a lot of stuff. 00:15:27.060 --> 00:15:33.740 But I guess I have got a question or kind of an open thing. 00:15:33.880 --> 00:15:38.540 I was going to try to make it like strict compatible, but I don't think I like it. 00:15:38.540 --> 00:15:46.520 So with strict compatible, you have to do things like return none and have a lot of extra decorations. 00:15:46.880 --> 00:15:47.680 It's a test. 00:15:47.880 --> 00:15:50.080 We know that Pi test tests return none. 00:15:50.200 --> 00:15:50.960 They have to. 00:15:51.200 --> 00:15:53.440 I think this is overkill most of the time. 00:15:53.440 --> 00:15:58.280 So for most of I think it's cleaner just to not show that. 00:15:58.400 --> 00:16:02.440 These are supposed to be examples of how to use it, how to use a tool. 00:16:02.740 --> 00:16:05.520 I think it's fine to not have a lot of that stuff. 00:16:05.700 --> 00:16:10.040 So I still do type checking on the examples, but I'm not doing strict. 00:16:10.040 --> 00:16:14.540 So anyway, that was my two cents on that, I guess. 00:16:14.680 --> 00:16:23.940 I think it's fine to not have type safe or type clean types, clean strict type checking results on example code. 00:16:24.280 --> 00:16:25.280 I think it can be too much. 00:16:25.520 --> 00:16:29.820 I think it has certainly a value, but there's plenty of times where it's just too much. 00:16:29.920 --> 00:16:32.600 And these type checkers can definitely get out of control. 00:16:32.760 --> 00:16:38.660 Like I had something similar with Flask where I had a Flask view method and I had a custom decorator. 00:16:38.660 --> 00:16:47.800 And the custom decorator didn't declare its types in a way that it was identical to this really complicated, you know, app at app dot get sort of signature. 00:16:48.020 --> 00:16:52.600 And I got these huge crazy warnings saying this function has a mismatch on the signature. 00:16:52.760 --> 00:16:54.140 I'm like, it's called by Flask. 00:16:54.240 --> 00:16:55.260 No one calls this function. 00:16:55.400 --> 00:16:58.780 I don't care about expressing what its signature. 00:16:58.900 --> 00:16:59.860 It works fine. 00:16:59.920 --> 00:17:00.680 Just leave it alone. 00:17:00.800 --> 00:17:01.280 You know what I mean? 00:17:01.540 --> 00:17:01.680 Yeah. 00:17:01.680 --> 00:17:02.660 It can go way too far. 00:17:03.620 --> 00:17:07.280 There's some things that I need to also things I need to look at. 00:17:07.280 --> 00:17:13.360 Because like for an example, for a test, doing something like assert one equals two makes sense. 00:17:13.460 --> 00:17:14.440 You know that's going to fail. 00:17:14.740 --> 00:17:16.100 That's a great example, I think. 00:17:16.440 --> 00:17:19.500 But like the mypyth says that that's an error. 00:17:19.860 --> 00:17:24.480 And because what, int constants aren't comparable? 00:17:24.660 --> 00:17:27.360 I can't remember the error, but it's weird. 00:17:28.180 --> 00:17:28.660 Interesting. 00:17:29.060 --> 00:17:29.300 Yeah. 00:17:29.300 --> 00:17:32.960 It can be over that last one or two percent is just, it's folly. 00:17:33.340 --> 00:17:34.060 That's what I say. 00:17:35.200 --> 00:17:37.540 So anyway, thanks to everybody. 00:17:37.860 --> 00:17:40.900 And yeah, if you don't, this isn't a dead project. 00:17:41.240 --> 00:17:41.240 00:17:41.500 --> 00:17:41.820 Yay. 00:17:41.960 --> 00:17:42.080 Cool. 00:17:42.340 --> 00:17:42.460 Yeah. 00:17:42.660 --> 00:17:42.820 Yeah. 00:17:42.820 --> 00:17:43.280 It looks awesome. 00:17:43.500 --> 00:17:45.940 And over a thousand projects using it. 00:17:45.940 --> 00:17:46.400 That's pretty good. 00:17:46.580 --> 00:17:46.780 Yeah. 00:17:46.860 --> 00:17:48.540 And some of those have lots of stars. 00:17:49.000 --> 00:17:48.980 So. 00:17:49.340 --> 00:17:49.680 Nice. 00:17:50.040 --> 00:17:51.580 Transitive stars are piling up. 00:17:51.680 --> 00:17:52.000 Feel it. 00:17:52.000 --> 00:17:54.280 Anyway. 00:17:54.520 --> 00:17:54.760 Okay. 00:17:55.000 --> 00:17:55.600 What you got next? 00:17:55.880 --> 00:18:02.300 What I got next is I want to talk about what, something I learned through that whole discussion 00:18:02.300 --> 00:18:06.220 around that raw plus DC design pattern that I talked about. 00:18:06.500 --> 00:18:10.280 Somebody, somebody said, look, you are so close, but where's your validation? 00:18:10.280 --> 00:18:11.460 I'm like, you know what? 00:18:11.580 --> 00:18:14.600 I'm not really sure I need runtime validation when I'm reading from the database. 00:18:14.820 --> 00:18:15.820 Maybe writing to it. 00:18:16.020 --> 00:18:16.260 Maybe. 00:18:16.560 --> 00:18:20.900 But anyway, I learned about a couple of really cool tools that you can use. 00:18:20.900 --> 00:18:24.220 So for data classes, there's this thing called the data class wizard. 00:18:24.500 --> 00:18:26.580 Ooh, a wizard has a cool wizard hat. 00:18:26.820 --> 00:18:30.080 It's simple, elegant wizarding tools for Python's data classes. 00:18:30.460 --> 00:18:32.460 So this is actually a super neat project here. 00:18:32.740 --> 00:18:35.980 So, and fun docs, behold the power of data class wizard. 00:18:36.340 --> 00:18:38.820 So what you can do, there's different ways you can use it. 00:18:38.900 --> 00:18:42.860 You can use it through what are called mixins, which are like, they add some functionality, 00:18:43.260 --> 00:18:45.040 you know, through inheritance. 00:18:45.780 --> 00:18:45.960 Okay. 00:18:46.180 --> 00:18:49.560 You know, maybe like a from JSON function, something like that. 00:18:49.560 --> 00:18:54.220 But also lets you specify other interesting things. 00:18:54.740 --> 00:18:59.320 Like, I didn't even know you could do this in the class definition, but basically pass 00:18:59.320 --> 00:19:04.000 configuration values to a class, the class constructor for the type, I guess. 00:19:04.200 --> 00:19:07.460 Anyway, you can say, hey, I have a regular data class. 00:19:07.460 --> 00:19:14.420 But in fact, I would like it to be unconcerned with the case of the inbound data. 00:19:14.520 --> 00:19:18.620 So somebody doesn't say C# and they write Pascal case things, right? 00:19:18.680 --> 00:19:24.280 Like lowercase and then uppercase second bit or, you know, camel case type of things like 00:19:24.280 --> 00:19:26.120 we do for classes for their fields. 00:19:26.120 --> 00:19:31.700 But I want to express them as, you know, snake case is active Tupa with underscores and not 00:19:31.700 --> 00:19:33.280 capital A, capital T sort of thing. 00:19:33.400 --> 00:19:35.000 You can just say, oh, that doesn't matter. 00:19:35.300 --> 00:19:38.700 And if I ever dump it out, put it back out in camel case, because that's what the people 00:19:38.700 --> 00:19:41.020 who are working with this API want. 00:19:41.280 --> 00:19:46.440 So this data class wizard is like a, it's like data classes, but it adds that type of functionality 00:19:46.440 --> 00:19:46.980 to it. 00:19:46.980 --> 00:19:47.180 Right? 00:19:47.420 --> 00:19:47.660 Cool. 00:19:47.660 --> 00:19:50.860 Which is super cool because my example, I said, oh, well, you just got to write a from 00:19:50.860 --> 00:19:53.940 JSON function that like turns it into, they're like, why would you write that? 00:19:53.960 --> 00:19:54.740 You could use the wizard. 00:19:55.000 --> 00:19:56.560 Like, cause I didn't know about the wizard. 00:19:56.680 --> 00:19:57.680 Tell me about this thing. 00:19:58.300 --> 00:20:00.140 But there's other ways to use it as well. 00:20:00.400 --> 00:20:05.300 You can have just a regular data class and mix in the JSON wizard mix in. 00:20:05.420 --> 00:20:08.160 So then you get to and from JSON, kind of like I just described. 00:20:08.340 --> 00:20:09.280 This is super cool. 00:20:09.460 --> 00:20:11.660 You can have a, an env wizard. 00:20:11.940 --> 00:20:14.960 So I can then have a data class that derives from env wizard. 00:20:14.960 --> 00:20:17.840 And then it reads as you just instantiate it. 00:20:17.840 --> 00:20:19.640 It reads from the environment variables. 00:20:20.100 --> 00:20:20.940 That's neat, right? 00:20:21.260 --> 00:20:21.500 Yeah. 00:20:21.760 --> 00:20:25.400 So you just have a config file and has all these variables and you just say, guess what? 00:20:25.440 --> 00:20:28.920 Those come out of my environment or doctor variables or whatever. 00:20:29.340 --> 00:20:30.840 And yeah, pretty neat. 00:20:31.100 --> 00:20:34.940 They've got the JSON wizard, the list wizard. 00:20:34.940 --> 00:20:40.220 Cause sometimes you get a set of JSON and it starts as a list instead of curly braces. 00:20:40.220 --> 00:20:41.440 And then it's hard to parse. 00:20:41.440 --> 00:20:45.020 So, files, Toml, Toml and YAML. 00:20:45.160 --> 00:20:49.680 You can also get ones that'll read and write Toml and YAML data classes that basically naturally 00:20:49.680 --> 00:20:50.460 know that type. 00:20:50.800 --> 00:20:54.100 And if you just keep going, it says you don't have to use inheritance either. 00:20:54.280 --> 00:20:58.600 You can just create standard data classes, even hierarchical data classes. 00:20:58.600 --> 00:21:02.120 Like here has an A class, which has a list of B types. 00:21:02.120 --> 00:21:03.740 And those are also data classes. 00:21:03.740 --> 00:21:08.640 And then you can just use the from dict from data class wizard. 00:21:08.800 --> 00:21:11.000 Say it's this type and here's the dictionary. 00:21:11.160 --> 00:21:13.740 And then boom, it's like a little factory method sort of thing. 00:21:14.060 --> 00:21:17.040 Oh, that's especially great if the classes come from somebody else. 00:21:17.280 --> 00:21:17.660 Exactly. 00:21:17.840 --> 00:21:20.960 And it also does really interesting parsing. 00:21:21.080 --> 00:21:27.060 I can't remember where it is in their documentation, but it does, coercion or of types. 00:21:27.260 --> 00:21:29.400 I don't know exactly how they refer to it there. 00:21:29.400 --> 00:21:34.620 But if I've got, say I have a list of integers like Pydantic, it does, or at least used to 00:21:34.620 --> 00:21:35.740 do like one. 00:21:35.740 --> 00:21:39.160 And then the quote two is the second entry it'll, and it knows it's supposed to be an 00:21:39.160 --> 00:21:39.300 end. 00:21:39.340 --> 00:21:40.620 It'll turn that into an end and stuff. 00:21:40.700 --> 00:21:44.960 So it kind of brings a lot of what Pydantic offers to data classes, but only when you want 00:21:44.960 --> 00:21:45.180 them. 00:21:45.420 --> 00:21:45.580 Cool. 00:21:45.700 --> 00:21:49.440 You're like in this place, I want to parse it and validate it, but not other times, 00:21:49.520 --> 00:21:50.240 you know what I mean? 00:21:50.320 --> 00:21:50.940 Stuff like that. 00:21:51.040 --> 00:21:54.320 So anyway, I present to you the data class wizard. 00:21:54.500 --> 00:21:55.140 It's pretty cool. 00:21:55.380 --> 00:21:55.600 Yeah. 00:21:55.700 --> 00:21:57.920 And hopefully it stays maintained. 00:21:58.180 --> 00:21:58.540 Okay. 00:21:58.540 --> 00:21:59.320 That's a good question. 00:21:59.400 --> 00:21:59.740 Let's go. 00:21:59.860 --> 00:22:01.000 Let's, let's go to the GitHub. 00:22:01.320 --> 00:22:04.620 It says 3.14, I think tested at least. 00:22:05.020 --> 00:22:05.380 Yeah. 00:22:05.520 --> 00:22:05.820 Okay. 00:22:05.980 --> 00:22:07.640 Two months ago it had an update. 00:22:07.820 --> 00:22:09.300 It only has two open pull requests. 00:22:09.300 --> 00:22:10.100 So that's pretty good. 00:22:10.900 --> 00:22:12.080 It doesn't have 200. 00:22:12.760 --> 00:22:13.160 Yeah. 00:22:13.200 --> 00:22:13.520 Anyway. 00:22:13.860 --> 00:22:13.980 Yeah. 00:22:14.040 --> 00:22:14.500 Pretty neat. 00:22:14.560 --> 00:22:15.260 I'm liking it. 00:22:15.260 --> 00:22:18.920 I haven't really had a chance to use it because I just learned about it, but I like the idea 00:22:18.920 --> 00:22:19.420 of it a lot. 00:22:19.620 --> 00:22:19.860 Yeah. 00:22:20.020 --> 00:22:20.820 That looks fun. 00:22:21.440 --> 00:22:27.560 I have got a, another project that is in a similar boat that I haven't, I haven't really 00:22:27.560 --> 00:22:28.380 tried it that much. 00:22:28.380 --> 00:22:30.000 I downloaded it and installed it. 00:22:30.220 --> 00:22:34.340 and what it is, is SQL light. 00:22:34.340 --> 00:22:36.860 Oh, SQL, IDO, SQL. 00:22:36.860 --> 00:22:37.900 SQL IDO. 00:22:37.900 --> 00:22:39.200 I think it's gotta be SQL IDO. 00:22:39.200 --> 00:22:39.640 That's the fun. 00:22:39.640 --> 00:22:39.940 That's the fun. 00:22:39.940 --> 00:22:41.060 SQL IDO. 00:22:41.060 --> 00:22:46.340 So this is, this is a new little toy from, Adam Hill. 00:22:46.580 --> 00:22:52.240 And, I noticed, I can't remember what, I think I saw a social media post on it or 00:22:52.240 --> 00:22:52.540 something. 00:22:52.740 --> 00:22:58.400 Anyway, it's a macOS SQL light browser built for normal people, apparently. 00:22:58.400 --> 00:23:01.280 so, but it looks good. 00:23:01.280 --> 00:23:04.460 It looks in with the little, all the emojis. 00:23:04.460 --> 00:23:08.460 I'm thinking, wow, I think he used, maybe some agents to help build it. 00:23:08.460 --> 00:23:10.160 And that's kind of what I want to talk about. 00:23:10.160 --> 00:23:13.140 But, but I do like, it, it does look good. 00:23:13.140 --> 00:23:16.140 the, the, oops, error loading page. 00:23:16.340 --> 00:23:16.840 I don't know. 00:23:16.920 --> 00:23:21.640 Can't, I can't look at the over the overview pictures that much more, but, it looks 00:23:21.640 --> 00:23:27.900 like a very nice, interface to, you know, look at your SQL light databases on Mac. 00:23:28.000 --> 00:23:28.860 So that's cool. 00:23:29.380 --> 00:23:30.980 so why am I bringing this up? 00:23:30.980 --> 00:23:37.180 I'm bringing it up kind of because, there was a caveat with installing it and it said, 00:23:37.180 --> 00:23:43.940 um, it's a, it's a, it's, you could download the DMG and then you drag that to your app 00:23:43.940 --> 00:23:49.920 folder or just double click it and then you start querying, but there is, it's not verified. 00:23:49.920 --> 00:23:51.220 So it's not through the app store. 00:23:51.540 --> 00:23:55.080 it's, it doesn't have a developer certificate yet. 00:23:55.080 --> 00:23:59.000 so why, why am I okay with like trying this out? 00:23:59.120 --> 00:24:01.540 Well, I think it trust is a big thing. 00:24:01.660 --> 00:24:02.720 I trust Adam Hill. 00:24:02.800 --> 00:24:03.940 I've seen a lot of his work. 00:24:04.000 --> 00:24:05.420 I've seen him comment on stuff. 00:24:05.420 --> 00:24:06.720 He works on open source projects. 00:24:07.360 --> 00:24:10.060 and so there's some inherent trust there. 00:24:10.060 --> 00:24:13.140 If you don't know who Adam Hill is, you don't have that of course. 00:24:13.480 --> 00:24:16.680 but there's, there's other things I was looking at. 00:24:16.800 --> 00:24:22.260 So I was looking at the source code a little bit and he's openly saying there's agents there. 00:24:22.580 --> 00:24:24.400 there's an eight. 00:24:24.620 --> 00:24:27.480 Well, cause there's an agent's file or something talking about. 00:24:27.540 --> 00:24:27.700 Yeah. 00:24:27.920 --> 00:24:28.400 Interesting. 00:24:28.580 --> 00:24:29.780 It looks like it was built with Gemini. 00:24:30.120 --> 00:24:30.360 Okay. 00:24:30.400 --> 00:24:32.860 There's not a Claude.md, but there's a Gemini.md. 00:24:33.080 --> 00:24:33.300 Yeah. 00:24:33.300 --> 00:24:36.880 And that's just pointing to the agent, his agent at MD file. 00:24:36.880 --> 00:24:42.080 and then there's, there's like some other stuff on, like there's a directory with 00:24:42.080 --> 00:24:44.040 some to-do list ideas and stuff. 00:24:44.120 --> 00:24:48.800 And I think, I think just being like not hiding it, I think is good. 00:24:48.800 --> 00:24:54.520 And also another thing I checked out is the contributor because there's, I don't, so all 00:24:54.520 --> 00:24:55.800 the contributors are him. 00:24:56.140 --> 00:25:03.340 So I, I, I'm thinking that even if like there's a bunch of code written by agents, he's 00:25:03.340 --> 00:25:04.700 at least looking at it enough. 00:25:04.860 --> 00:25:06.760 And I don't know. 00:25:06.800 --> 00:25:07.800 I'm still on the fence. 00:25:07.860 --> 00:25:13.280 I don't know how I feel about people giving agents control of, commits, on a project 00:25:13.280 --> 00:25:13.800 or not. 00:25:13.800 --> 00:25:15.860 I think I'm not here for it. 00:25:16.000 --> 00:25:16.460 You're not. 00:25:16.740 --> 00:25:16.980 No. 00:25:17.040 --> 00:25:23.060 So, but like what I would, if I would was, what, what would I be okay with? 00:25:23.060 --> 00:25:29.200 having somebody, do this stuff and then commit themselves because you would 00:25:29.200 --> 00:25:30.460 assume that they're looking at it. 00:25:30.460 --> 00:25:32.440 But what if you just do that on a branch? 00:25:32.440 --> 00:25:35.920 If you had, if you had the agent working on a branch committing stuff, you could always 00:25:35.920 --> 00:25:36.740 just throw it away. 00:25:36.740 --> 00:25:38.380 And if you're happy with it, you can merge it. 00:25:38.380 --> 00:25:40.020 I don't know how it's different. 00:25:40.420 --> 00:25:40.960 so. 00:25:41.040 --> 00:25:41.960 I, I hear you. 00:25:42.100 --> 00:25:42.640 I hear you. 00:25:42.800 --> 00:25:44.260 I certainly think your branch makes it better. 00:25:44.580 --> 00:25:46.320 A feature branch, some exploration branch. 00:25:46.540 --> 00:25:48.780 But it would still show up as a contributor in the history. 00:25:48.980 --> 00:25:51.300 so that's, I guess that's where I'm going with that. 00:25:51.500 --> 00:25:51.700 Yeah. 00:25:51.740 --> 00:25:58.180 By default, if you ask Claude to create a GitHub, a Git message, it'll put, or PR, 00:25:58.340 --> 00:26:00.580 it'll put co-contributor at Claude. 00:26:00.900 --> 00:26:01.180 Yeah. 00:26:01.180 --> 00:26:06.480 So I still, I'm not, I'm still not sure how I feel about all that, but, but I do, 00:26:06.740 --> 00:26:08.980 there's, there's just, there's other things. 00:26:08.980 --> 00:26:10.580 So I'm thinking, I trust the developer. 00:26:10.580 --> 00:26:15.060 I trust that I look, I look at the contributor list and I look at some of the commits. 00:26:15.220 --> 00:26:16.640 what else am I looking at? 00:26:16.660 --> 00:26:21.240 I'm looking to make sure that there's tests, even though, I can't read a Swift test very 00:26:21.240 --> 00:26:21.460 well. 00:26:21.460 --> 00:26:21.880 I don't know. 00:26:22.020 --> 00:26:24.520 I mean, there's at least some testing going on. 00:26:24.860 --> 00:26:30.360 so, so that, that, I guess that's, that's where I'm, I'm at with, some of 00:26:30.360 --> 00:26:30.800 these things. 00:26:30.860 --> 00:26:36.260 I, and I bring this up partly because I review a lot of projects, you know, for this, for 00:26:36.260 --> 00:26:38.780 this, podcast looking at projects. 00:26:38.820 --> 00:26:43.920 And if it looks like there are signs like a bunch of emojis, that's not necessarily bad, 00:26:43.920 --> 00:26:46.400 but it's a sign that somebody had agents help them with it. 00:26:46.400 --> 00:26:49.320 And actually a lot of people are really bad at writing read me. 00:26:49.320 --> 00:26:52.080 So I'm okay with people having help writing read me. 00:26:52.260 --> 00:26:53.980 but I'm looking for tests. 00:26:53.980 --> 00:27:00.000 I'm looking for what the contributor things I'm looking for a lot of stuff, and trusting 00:27:00.000 --> 00:27:02.040 the developers a big thing still, I think. 00:27:02.280 --> 00:27:08.700 So yeah, a hundred percent with a command book, my app for managing like long running 00:27:08.700 --> 00:27:09.700 terminal like commands. 00:27:09.840 --> 00:27:10.120 Yeah. 00:27:10.360 --> 00:27:13.360 I had to go through the same thing, had to get it digitally signed. 00:27:13.720 --> 00:27:16.560 Both the app bundle itself is digitally signed. 00:27:16.560 --> 00:27:21.760 And then you have to go through a separate process to sign digitally sign the DMG as well, 00:27:21.760 --> 00:27:23.540 which is an utter pain for sure. 00:27:23.540 --> 00:27:28.260 So it is a hassle, but if you have an Apple developer account, it's not too bad. 00:27:28.320 --> 00:27:32.000 It's way less bad than iOS and app store review and those kinds of things. 00:27:32.000 --> 00:27:34.300 It's just running it through like a series of tools. 00:27:34.400 --> 00:27:38.620 So if Adam Hill, if you're out there listening and you want, a quick bit of help, happy 00:27:38.620 --> 00:27:40.600 to help you get over that last hump there. 00:27:40.820 --> 00:27:41.980 Okay, cool. 00:27:42.260 --> 00:27:43.380 it doesn't say yet. 00:27:43.400 --> 00:27:44.640 So I think he's planning on it. 00:27:44.640 --> 00:27:47.520 So I know it's just, it's brutal, but like, here's another thing. 00:27:47.520 --> 00:27:50.740 I like, I ask, I ask Claude, I'm like, all right, Claude, you got to help me. 00:27:50.800 --> 00:27:51.860 I I'm lost here. 00:27:51.900 --> 00:27:54.040 I do not know how to notarize this thing. 00:27:54.080 --> 00:27:54.980 It's which is what it's called. 00:27:54.980 --> 00:27:57.900 And then like, do I have to notarize the DMG as well? 00:27:57.920 --> 00:27:58.740 And how do I pack it? 00:27:58.820 --> 00:28:02.120 And I'm like, okay, here's a big long list of steps I got to set up. 00:28:02.120 --> 00:28:04.300 And I got like, it was pretty helpful actually. 00:28:04.820 --> 00:28:09.060 And that another thing, I guess a little bit of a tangent, right away from the trust thing. 00:28:09.060 --> 00:28:12.120 I think this seems like a tool that Adam wanted. 00:28:12.600 --> 00:28:17.260 And because of agents, he found time in his schedule to build it. 00:28:17.260 --> 00:28:22.580 and we wouldn't, without it, we probably wouldn't have this tool as something we can 00:28:22.580 --> 00:28:23.080 look at. 00:28:23.080 --> 00:28:28.920 now trying to get people to stay maintaining projects that they can open source in a few 00:28:28.920 --> 00:28:31.980 hours, over the time is a different story. 00:28:31.980 --> 00:28:34.460 But anyway, I guess enough on that. 00:28:34.460 --> 00:28:40.460 I guess I do think we're entering an entirely different era of highly personalized software. 00:28:40.820 --> 00:28:44.960 I made this for me because I could do it in an hour and exactly it solves my problem. 00:28:44.960 --> 00:28:47.280 maybe it's interested in you, but it's not built to be general. 00:28:47.280 --> 00:28:48.000 It's built for me. 00:28:48.240 --> 00:28:48.580 Yeah. 00:28:48.860 --> 00:28:49.040 Yeah. 00:28:49.040 --> 00:28:51.960 There's going to be some of that, but why open source it at that point? 00:28:51.960 --> 00:28:53.200 yeah, fair. 00:28:53.740 --> 00:28:54.140 Anyway. 00:28:54.400 --> 00:28:56.140 Well, that would be extra work, wouldn't it? 00:28:56.520 --> 00:28:56.840 Yeah. 00:28:56.920 --> 00:28:57.740 Do we have any extras? 00:28:58.300 --> 00:28:58.840 I do. 00:28:58.900 --> 00:29:00.040 Do you got any while your screen's up? 00:29:00.220 --> 00:29:00.640 Yeah, sure. 00:29:00.700 --> 00:29:00.980 I'll do. 00:29:01.160 --> 00:29:03.320 and it's all on the same sort of topic. 00:29:03.320 --> 00:29:09.740 So, an article on, htmx.org blog, called yes. 00:29:09.760 --> 00:29:15.060 And from Carson gross, this says, I teach computer science at Montana university, 00:29:15.200 --> 00:29:18.000 Montana state university, and father of three sons. 00:29:18.000 --> 00:29:23.460 And he gets the question given AI, should I still consider being a computer programmer? 00:29:23.460 --> 00:29:27.720 Or I would translate that is into is a CS degree still worth it? 00:29:27.720 --> 00:29:30.040 and his response is yes. 00:29:30.040 --> 00:29:33.880 And, and I do, I, it's a great article. 00:29:34.140 --> 00:29:35.760 enjoy reading it. 00:29:35.840 --> 00:29:41.780 the, the top level thing is, programming is fundamentally problem solving using computers 00:29:41.780 --> 00:29:45.540 and learning to control complexity while solving this problems. 00:29:45.540 --> 00:29:48.320 I would say that that's, that's computer science. 00:29:48.320 --> 00:29:51.980 but that's, and then a bunch of other stuff. 00:29:51.980 --> 00:29:57.460 I think it's experience helps even, and, and background knowledge helps even if you're 00:29:57.460 --> 00:30:03.540 using AI to help, then that's the same kind of a similar sentiment over on, Simon 00:30:03.540 --> 00:30:06.740 Willison's blog, hoard the things, you know, how to do. 00:30:06.820 --> 00:30:12.460 one of his recommendations here is, a big part of a big part of the skill building 00:30:12.460 --> 00:30:16.980 softwares understanding what is possible and what isn't, and having at least a rough idea 00:30:16.980 --> 00:30:19.140 of how those things can be accomplished. 00:30:19.140 --> 00:30:21.200 And this helps with directing agents. 00:30:21.200 --> 00:30:25.280 You wouldn't ask it to do things that, you know, is going to be a completely ridiculous, 00:30:25.280 --> 00:30:32.180 um, I mean, an agent might go off and do something dumb, but if you ask it to, you just, 00:30:32.180 --> 00:30:37.040 you do have to have experience to know what dumb is and what it possible is, I guess. 00:30:37.040 --> 00:30:40.140 So, yeah, I agree with Carson that yes. 00:30:40.140 --> 00:30:45.960 And I think I basically the short version is I believe Jevon's paradox is real and that 00:30:45.960 --> 00:30:49.400 we're going to have more software rather than no software developers. 00:30:49.400 --> 00:30:54.820 And we need people to oversee and guide and make decisions and trade-offs about that. 00:30:54.960 --> 00:30:55.060 Right. 00:30:55.080 --> 00:30:57.640 It's very hard for AI to make proper trade-offs. 00:30:57.800 --> 00:31:02.800 Like, yes, I know I could, yes, I know I could make this an app, but my company doesn't have 00:31:02.800 --> 00:31:04.680 a culture of putting apps out. 00:31:04.880 --> 00:31:08.040 And we actually have this other backend system we want to, you know what I mean? 00:31:08.040 --> 00:31:12.600 There's just like all this information that it doesn't know, or like in our industry, 00:31:12.600 --> 00:31:14.200 here's how we do this. 00:31:14.200 --> 00:31:18.820 And there's just a bunch of stuff that humans can bring plus the ability to break it down 00:31:18.820 --> 00:31:21.260 into smaller problems and then give that to an AI. 00:31:21.260 --> 00:31:24.400 You're way more likely to be successful than just like, Hey, guess what? 00:31:24.400 --> 00:31:25.780 I need a CRM build that. 00:31:25.780 --> 00:31:26.740 I'll be back by lunch. 00:31:26.740 --> 00:31:31.320 Like you're going to get some form of something, but it's again, it's not highly personalized. 00:31:31.320 --> 00:31:32.500 It's just super generic. 00:31:32.500 --> 00:31:35.740 And if that's all you want, just go get a CMS from like some SAS and you're done. 00:31:36.400 --> 00:31:39.520 I think one of the really interesting things is like, why do we build software at all? 00:31:39.640 --> 00:31:42.240 You know, there's this term software engineer, right? 00:31:42.400 --> 00:31:43.220 How well it applies. 00:31:43.220 --> 00:31:46.940 I don't know, but a lot of times, especially in the early days of like talking about architecture 00:31:46.940 --> 00:31:50.460 and engineering, it was all tied back to like, well, how do civil engineers do it? 00:31:50.700 --> 00:31:53.720 And they have all these, these formalities of how they work. 00:31:53.720 --> 00:31:58.180 And maybe that should be brought into computer engineering and programming and software engineering 00:31:58.180 --> 00:31:58.620 and so on. 00:31:59.500 --> 00:32:00.780 There's a really big difference. 00:32:01.020 --> 00:32:06.680 Civil engineers built a bridge 50 times because there are 50 places the bridge need to be built. 00:32:06.900 --> 00:32:11.540 If you build a piece of software that solves the same problem once, like you just run it again, 00:32:11.540 --> 00:32:15.920 you copy it and run it, you build new software because you're solving a problem that in some 00:32:15.920 --> 00:32:17.240 way has never been solved before. 00:32:17.600 --> 00:32:18.980 So there's a difference. 00:32:19.360 --> 00:32:22.140 There's like this, this, the part where it's never been solved before. 00:32:22.240 --> 00:32:25.960 That's where the human brings in the stuff to the AI and the AI goes, well, there's all 00:32:25.960 --> 00:32:27.460 these Lego blocks that we already know. 00:32:27.520 --> 00:32:28.380 Like, all right, put those together. 00:32:28.380 --> 00:32:31.040 But for this, I think that's what software engineering is going to become. 00:32:31.040 --> 00:32:31.440 Yeah. 00:32:31.660 --> 00:32:33.480 I mean, I think it has become that already. 00:32:33.740 --> 00:32:37.100 And electrical engineering has been that way for a long time. 00:32:38.040 --> 00:32:38.480 So, yeah. 00:32:38.660 --> 00:32:42.760 I mean, already with like packages on PyPI and things like that, it's kind of like that. 00:32:42.820 --> 00:32:46.920 But like with AI, it's instead of, I need an auth library so I can build a login page. 00:32:46.980 --> 00:32:48.520 It's like, I make a login page. 00:32:48.760 --> 00:32:50.060 Now click that together with it. 00:32:50.120 --> 00:32:51.580 You know, it's just like another level up. 00:32:51.640 --> 00:32:53.840 But that doesn't mean that people necessarily go away. 00:32:54.080 --> 00:32:54.780 I hope not. 00:32:55.040 --> 00:32:56.460 At least for another 20 years. 00:32:56.760 --> 00:32:56.920 Yeah. 00:32:56.920 --> 00:32:59.120 Give it, give it till, till Brian and I retire then. 00:32:59.120 --> 00:33:01.700 All right, I got some fun extras as well. 00:33:02.040 --> 00:33:02.280 Okay. 00:33:02.660 --> 00:33:08.520 So, first of all, how would you like to unplug your Python and get a bit of an MTV vibe? 00:33:09.020 --> 00:33:09.760 But it's PyTV. 00:33:10.240 --> 00:33:11.040 Yeah, it's so fun. 00:33:11.460 --> 00:33:16.160 So, our friends at JetBrains and a bunch of our friends who are speaking at this conference, 00:33:16.460 --> 00:33:20.660 but are not necessarily at JetBrains, are putting on a free online Python conference, 00:33:20.920 --> 00:33:22.900 the best Python, straight into your living room. 00:33:23.220 --> 00:33:26.120 I feel like Max Headroom should kick off the keynote, honestly, don't you? 00:33:26.720 --> 00:33:28.260 I love Max Headroom, yeah. 00:33:28.260 --> 00:33:29.040 I know I did too. 00:33:29.400 --> 00:33:31.400 But there's a bunch of great people speaking here. 00:33:31.760 --> 00:33:37.700 Carol Willing, Sarah Boyce, Marlene, Gina, Carlton, a lot of folks we know. 00:33:38.200 --> 00:33:39.580 And so, yeah, check this out. 00:33:39.640 --> 00:33:41.140 It's a free online conference. 00:33:41.900 --> 00:33:43.740 And is it this week? 00:33:43.960 --> 00:33:44.320 When is it? 00:33:44.360 --> 00:33:44.760 Let me look. 00:33:44.920 --> 00:33:46.040 It must have a date, surely. 00:33:46.360 --> 00:33:46.580 I know. 00:33:46.620 --> 00:33:47.320 I think it is this week. 00:33:47.380 --> 00:33:48.040 Yeah, March 4th. 00:33:48.040 --> 00:33:49.460 So, we're recording on March 2nd. 00:33:49.520 --> 00:33:50.340 That would be Wednesday. 00:33:50.760 --> 00:33:54.000 So, everybody be on top of your listening to the show or miss out. 00:33:54.140 --> 00:33:54.320 Okay. 00:33:54.600 --> 00:33:55.200 So, that's one. 00:33:55.560 --> 00:33:56.460 Quick shout out to that. 00:33:56.840 --> 00:34:00.380 Next is, is AI a fad and a bubble? 00:34:00.600 --> 00:34:04.220 Or is it an actual thing with real value that's going to change the world? 00:34:04.460 --> 00:34:04.700 Yes. 00:34:04.700 --> 00:34:11.400 So, remember we talked about the Sasspocalypse where like some of the legal companies and 00:34:11.400 --> 00:34:15.940 other stuff where like their stock went way down because Anthropic published 13 markdown 00:34:15.940 --> 00:34:16.220 files. 00:34:16.600 --> 00:34:24.200 Well, IBM, IBM, big blue stock crashes 11% because Anthropic released some kind of markdown file 00:34:24.200 --> 00:34:25.260 that helps convert Cobalt. 00:34:25.260 --> 00:34:28.580 Because Anthropic threatens the Cobalt empire. 00:34:28.980 --> 00:34:30.580 I think there's a lot of interesting stuff here. 00:34:30.720 --> 00:34:35.920 I think Cobalt is a very interesting analogy for like pre and post agentic written code. 00:34:36.040 --> 00:34:39.000 I can see in 20 years are like, that's some pre AI code. 00:34:39.100 --> 00:34:42.020 We need some humans that know how to type still to come in here and look at this. 00:34:42.280 --> 00:34:42.360 Right. 00:34:42.360 --> 00:34:45.920 I think there's going to be that kind of like difference as we go on. 00:34:45.960 --> 00:34:46.680 And so, I don't know. 00:34:46.680 --> 00:34:47.640 I just wanted to throw that out there. 00:34:47.700 --> 00:34:48.860 It's an interesting read for folks. 00:34:49.640 --> 00:34:49.800 Yeah. 00:34:49.800 --> 00:34:52.400 The language stuff is an interesting thing. 00:34:52.400 --> 00:35:00.100 With things like Anthropic around, I am less scared of looking at code bases like Swift 00:35:00.100 --> 00:35:01.140 or Cobalt. 00:35:01.340 --> 00:35:04.200 I haven't looked at Cobalt, but I wouldn't be afraid now. 00:35:04.360 --> 00:35:04.780 I might be able to just. 00:35:04.780 --> 00:35:05.260 A hundred percent. 00:35:05.560 --> 00:35:05.780 Yeah. 00:35:06.140 --> 00:35:11.900 You need like 20% the capability you used to need to get something done or to get like, 00:35:12.160 --> 00:35:13.160 I don't understand this. 00:35:13.320 --> 00:35:14.040 Explain it to me. 00:35:14.140 --> 00:35:15.340 Like, okay, now it makes sense. 00:35:15.400 --> 00:35:17.080 Whereas before, I was just like, I don't understand it. 00:35:17.260 --> 00:35:18.720 Well, especially if you write a test for it. 00:35:18.720 --> 00:35:23.140 So, you write it like maybe there's a bug in some Cobalt code and you say, and you write 00:35:23.140 --> 00:35:23.800 a test for it. 00:35:23.900 --> 00:35:27.280 And now the agent, you can kind of trust the agent to fix it. 00:35:27.360 --> 00:35:29.020 It'll do good review, of course. 00:35:29.180 --> 00:35:29.660 But yeah. 00:35:29.900 --> 00:35:30.080 Yeah. 00:35:30.260 --> 00:35:30.640 Okay. 00:35:31.060 --> 00:35:31.840 Quick update. 00:35:32.220 --> 00:35:37.480 I'm absolutely in love with my 40 inch curved monitor that I told people about for. 00:35:37.660 --> 00:35:41.060 When I first got it, I was a little wigged out by how big and rappy around it is. 00:35:41.060 --> 00:35:46.940 It took some getting used to, but this like 5k monitor that is so giant is absolutely perfect 00:35:46.940 --> 00:35:49.100 for all the type of work we do these days. 00:35:49.320 --> 00:35:53.560 Next up, update est, update est, not ist. 00:35:53.820 --> 00:35:54.820 I tried to search for it. 00:35:54.880 --> 00:35:56.100 Update ist, I-S-T. 00:35:56.180 --> 00:35:56.920 No, it's E-S-T. 00:35:57.360 --> 00:35:59.720 So here's a really cool macOS bit of software. 00:36:00.120 --> 00:36:05.940 And what you do is you just install it and it looks for all different kinds of apps that 00:36:05.940 --> 00:36:07.620 are out of date on your system. 00:36:08.120 --> 00:36:10.160 It looks for auto updating apps. 00:36:10.160 --> 00:36:11.860 It looks for stuff from the app store. 00:36:11.960 --> 00:36:12.920 It looks at homebrew. 00:36:13.120 --> 00:36:16.200 It looks at CLI tools and it manages all of them. 00:36:16.520 --> 00:36:18.780 So you go, oh, look, this thing's actually, that was been depending on. 00:36:18.840 --> 00:36:19.420 It's actually out of date. 00:36:19.480 --> 00:36:20.180 Let me push the button. 00:36:20.500 --> 00:36:21.360 And super cool. 00:36:21.560 --> 00:36:22.740 It's got a really nice UI. 00:36:22.940 --> 00:36:25.400 It costs, it has money, but very little. 00:36:25.520 --> 00:36:27.880 It costs $13 once for your entire life. 00:36:28.020 --> 00:36:30.400 So anyway, it's pretty, seems pretty fair to me. 00:36:30.660 --> 00:36:31.520 So very cool app. 00:36:31.780 --> 00:36:33.240 Also Swift UI app, I believe. 00:36:33.500 --> 00:36:34.600 I like Adam Hills. 00:36:34.940 --> 00:36:36.300 Wait, how do you update it though? 00:36:36.600 --> 00:36:37.320 Oh my God. 00:36:37.360 --> 00:36:38.520 I believe it does auto update. 00:36:38.520 --> 00:36:38.880 Okay. 00:36:38.880 --> 00:36:40.920 If it doesn't, that would be deeply bad. 00:36:41.080 --> 00:36:47.560 I think it uses the same thing that Command Book uses, which is called Sparkle, which is one of the most common ways to do it. 00:36:47.560 --> 00:36:51.920 And actually it looks at all the Sparkle update feeds of all the apps installed on your system too, which is cool. 00:36:51.920 --> 00:36:52.360 Okay. 00:36:52.360 --> 00:36:55.480 People may have heard of Bartender, which is another Mac app. 00:36:55.480 --> 00:36:55.880 That's cool. 00:36:55.960 --> 00:36:59.760 Like compresses all the little widget thingies you got at the top on the right by the clock. 00:36:59.880 --> 00:37:00.960 I love all those things. 00:37:00.960 --> 00:37:05.140 So I go way overboard and I needed something to manage it, especially on small like recording screens. 00:37:05.140 --> 00:37:09.920 So I had Bartender, they got bought in like a sort of underhanded way. 00:37:09.920 --> 00:37:14.080 Like they got sold, but then nobody said they were sold for like months. 00:37:14.080 --> 00:37:15.460 It was like, they wouldn't announce. 00:37:15.920 --> 00:37:16.660 Anyway, it was weird. 00:37:16.740 --> 00:37:18.460 So a lot of people moved to open source stuff. 00:37:18.560 --> 00:37:25.840 So there's a one called Ice for the Mac menu bar, but it had its last updates like six months ago and it's clearly abandoned. 00:37:25.840 --> 00:37:27.340 So you got ice. 00:37:27.540 --> 00:37:29.500 Well, what if it gets stuck and frozen? 00:37:29.600 --> 00:37:30.300 You got to thaw it out. 00:37:30.380 --> 00:37:34.440 So now there's a popular one called Thaw, a fork of ice, the menu bar manager. 00:37:34.580 --> 00:37:38.160 So anyway, here's a free open source menu bar manager for Mac. 00:37:38.260 --> 00:37:39.040 That's pretty cool. 00:37:39.140 --> 00:37:41.700 Following that weird chain of events over the last couple of years. 00:37:42.460 --> 00:37:44.500 I don't even remember what I have. 00:37:45.100 --> 00:37:49.040 On this computer, I still have ice running because I haven't had a chance to update it. 00:37:49.160 --> 00:37:51.420 I just learned about this week, but it's getting thaw. 00:37:51.580 --> 00:37:53.380 It's getting thawed out when I get done with this recording. 00:37:53.660 --> 00:37:54.080 Getting thawed. 00:37:54.700 --> 00:37:55.060 Yeah. 00:37:55.060 --> 00:37:58.900 I have something that minimizes it, but I can't remember what it is. 00:38:00.380 --> 00:38:01.120 I don't know. 00:38:01.420 --> 00:38:01.640 Yeah. 00:38:01.940 --> 00:38:02.340 Okay. 00:38:03.020 --> 00:38:04.480 Well, shall we make a joke? 00:38:04.860 --> 00:38:05.260 Yeah. 00:38:05.480 --> 00:38:06.100 Something funny. 00:38:06.100 --> 00:38:07.480 I know you got kids. 00:38:07.560 --> 00:38:08.120 I have kids. 00:38:08.540 --> 00:38:13.160 Anyone out there who has kids, well, maybe roommates as well, but especially kids. 00:38:13.260 --> 00:38:14.200 Like they're so messy. 00:38:14.400 --> 00:38:15.280 This joke will appreciate. 00:38:15.420 --> 00:38:19.020 This is like a programmer take on that situation. 00:38:19.220 --> 00:38:22.620 And the title of the joke is The House is Read Only. 00:38:23.040 --> 00:38:23.460 Okay. 00:38:23.760 --> 00:38:24.680 The House is Read Only. 00:38:25.300 --> 00:38:28.860 So here's a woman named Kelly on X that posted. 00:38:29.140 --> 00:38:30.920 It says, I have cleaned the house. 00:38:31.140 --> 00:38:32.620 The house is now read only. 00:38:32.800 --> 00:38:34.420 Please do not edit the house. 00:38:36.340 --> 00:38:36.740 Yes. 00:38:37.100 --> 00:38:37.480 Yes. 00:38:37.800 --> 00:38:38.140 Kids. 00:38:38.420 --> 00:38:41.300 I just put the dishes in the dishwasher. 00:38:41.300 --> 00:38:43.800 Why is the sink full of dishes again? 00:38:43.960 --> 00:38:45.940 I just went up to the back to clean up the other stuff. 00:38:46.020 --> 00:38:47.180 I came back in and now it's messy again. 00:38:47.220 --> 00:38:48.060 I can't take it anymore. 00:38:48.440 --> 00:38:49.360 The house is read only. 00:38:49.460 --> 00:38:50.060 It's so good. 00:38:50.300 --> 00:38:51.600 I need this for my mother-in-law. 00:38:51.860 --> 00:38:52.060 Yeah. 00:38:52.060 --> 00:38:56.280 If you don't pay the mortgage, you don't get to put holes in the walls. 00:38:56.680 --> 00:38:56.960 Just saying. 00:38:57.300 --> 00:38:58.000 Houses read only. 00:38:58.280 --> 00:38:59.300 Please do not edit the house. 00:39:02.800 --> 00:39:03.200 Nice. 00:39:03.700 --> 00:39:04.260 That's funny. 00:39:04.260 --> 00:39:06.000 Yeah. 00:39:06.200 --> 00:39:09.460 It's been a crazy week outside of programming. 00:39:09.600 --> 00:39:11.220 So I'm trying to concentrate on programming. 00:39:12.500 --> 00:39:16.380 Escape to the well-known world of structured text. 00:39:16.500 --> 00:39:16.880 Let's go. 00:39:17.120 --> 00:39:17.380 Anyway. 00:39:17.680 --> 00:39:18.980 It was a fun episode. 00:39:19.140 --> 00:39:21.660 Thanks everybody for showing up and we'll see you all next week. 00:39:21.860 --> 00:39:22.020 Yeah. 00:39:22.220 --> 00:39:22.620 Thank you. 00:39:22.620 --> 00:39:22.900 Bye.