WEBVTT 00:00:00.001 --> 00:00:05.560 Welcome to a very special episode. This is the 100th episode of Talk Python To Me. 00:00:05.560 --> 00:00:10.400 It's the perfect chance to take a moment and look at where we've come from and where we're going. 00:00:10.400 --> 00:00:15.820 Not just with regard to this podcast, but for Python in general. And who better to do this with 00:00:15.820 --> 00:00:21.200 than Python's inventor himself, Guido Van Rossum. In this episode, we discuss how Guido got into 00:00:21.200 --> 00:00:27.600 programming, where Python came from and why, and Python's bright future with Python 3. This is 00:00:27.600 --> 00:00:32.360 Talk Python To Me, episode 100, recorded January 18th, 2017. 00:00:55.200 --> 00:01:01.980 Welcome to Talk Python To Me, a weekly podcast on Python, the language, the libraries, the ecosystem, 00:01:01.980 --> 00:01:06.740 and the personalities. This is your host, Michael Kennedy. Follow me on Twitter, where I'm at, 00:01:06.740 --> 00:01:11.980 mkennedy. Keep up with the show and listen to past episodes at talkpython.fm. And follow the 00:01:11.980 --> 00:01:18.720 show on Twitter via at Talk Python. This episode is brought to you by Rollbar and Hired. Thank them 00:01:18.720 --> 00:01:26.440 both for supporting the show. Check them out at Rollbar and at Hired underscore HQ on Twitter and 00:01:26.440 --> 00:01:33.320 tell them thank you. Hey, everyone. I just want to take a moment and reflect on this milestone of 100 00:01:33.320 --> 00:01:39.580 episodes and say a big thank you to everyone out there who's listening. The reason this podcast is 00:01:39.580 --> 00:01:44.800 successful, the reason I've kept doing it, is because so many of you tell me that you appreciate what I'm 00:01:44.800 --> 00:01:49.040 doing, that you enjoy all the guests that I have on the show. So I want to say thank you because 00:01:49.040 --> 00:01:55.180 without you, obviously, I would not have 100 episodes. I get to live basically my dream job. 00:01:55.180 --> 00:02:00.460 I talk to all these brilliant people in the tech industry and I get to share it with you and get 00:02:00.460 --> 00:02:06.060 all the great feedback that so many of you give me. So thanks. And we really have a special guest 00:02:06.060 --> 00:02:13.040 guest this episode with Guido and a look at Python over the years, the past, present and future. And I 00:02:13.040 --> 00:02:16.920 really hope you enjoy that. If you're out there thinking, hey, I really love the show and I'd like 00:02:16.920 --> 00:02:22.080 to support it. There's a couple of things you can do that are really easy. One, if you give us a review 00:02:22.080 --> 00:02:27.840 on iTunes, that actually makes a big difference how we rank within iTunes. So that's kind of like Google 00:02:27.840 --> 00:02:33.220 rank for podcasts. That would be great. If you want to do more than that, one of the really great ways to 00:02:33.220 --> 00:02:38.680 support me and what I'm doing would actually be to buy one of my classes or recommend my training 00:02:38.680 --> 00:02:44.800 content to your employer or your team at work. If you're into that, check that out at training.talkpython.fm. 00:02:44.800 --> 00:02:50.380 And there's even a little Patreon link if you just want to give a dollar or two a week. So thank you 00:02:50.380 --> 00:02:56.720 everybody for making this possible. Thank you for helping me reach 100 episodes and been great to share 00:02:56.720 --> 00:03:02.160 them with you. And I hope you've really enjoyed them yourself. All right. With all that said, here's Guido. 00:03:03.160 --> 00:03:07.960 Guido, welcome to Talk Python. Glad to be here, Michael. I'm honored that you're coming on my 00:03:07.960 --> 00:03:13.040 100th show to celebrate this special episode. And I know everyone in the community is really going to 00:03:13.040 --> 00:03:19.520 appreciate this look at the history and the present and future of Python with you. Well, let's have it. 00:03:19.520 --> 00:03:24.240 All right. Absolutely. So we're going to dig into a whole bunch of things about Python. But before we 00:03:24.240 --> 00:03:27.060 get there, let's start with your story. How'd you get into programming in the first place? 00:03:27.060 --> 00:03:33.840 Well, that was a long time ago. In high school, I did not know what a computer was. I believe I had 00:03:33.840 --> 00:03:40.520 not even ever heard of the word. I was an electronics hobbyist, though. And I got started, I think, 00:03:40.520 --> 00:03:47.980 around the age of 10, building very simple analog circuits, things like a little radio receiver that 00:03:47.980 --> 00:03:55.440 made from a kit. And I gradually discovered simple digital electronics and integrated circuits were 00:03:55.440 --> 00:04:01.480 becoming available to hobbyists like me, even with my very small amount of pocket money. 00:04:02.320 --> 00:04:07.800 And that's where I was when I graduated from high school. Then I went to university to study 00:04:07.800 --> 00:04:13.820 mathematics, the University of Amsterdam. And it was like a completely different world. They had a 00:04:13.820 --> 00:04:18.740 mainframe in the basement, and there were programming classes. And the languages that I remember were 00:04:18.740 --> 00:04:25.660 Algal 60 and Pascal. And I was basically instantly hooked, even though the first year I remember the 00:04:25.660 --> 00:04:29.660 only way I could input programs to the computer was through punch cards. 00:04:30.000 --> 00:04:35.140 Yeah. If you start programming, basically via hardware, and then you have this machine that 00:04:35.140 --> 00:04:39.900 you can just feed anything, the whole world opens up, right? Even if it's punch cards, it's like it can 00:04:39.900 --> 00:04:41.840 do anything I ask it to do almost, right? 00:04:41.840 --> 00:04:46.900 I was so happy that I didn't have to sort of solder stuff together anymore. 00:04:46.900 --> 00:04:48.340 That's right. 00:04:48.340 --> 00:04:50.700 Because that was always my weak point. 00:04:50.700 --> 00:04:56.940 Oh, it's such a difference. So you got started really in the early days, and when the term hacker 00:04:56.940 --> 00:04:58.620 meant something entirely different, right? 00:04:58.760 --> 00:05:03.700 I don't think I knew the word hacker. That was like decades later that people 00:05:03.700 --> 00:05:05.720 told me I had been a hacker. 00:05:05.720 --> 00:05:11.540 Oh, I didn't know. How interesting. Okay, excellent. And then you got into, 00:05:11.540 --> 00:05:16.380 you started working on programming languages. How'd you go from playing with mainframes and 00:05:16.380 --> 00:05:19.080 punch cards to working on things like ABC and stuff? 00:05:19.080 --> 00:05:24.620 Well, I guess I developed an interest in learning different programming languages. That probably 00:05:24.620 --> 00:05:30.660 started out when my first year there were two different languages being taught, I think, 00:05:30.660 --> 00:05:38.300 Algol and Pascal. And then I hung out with a bunch of physics students and their favorite language was 00:05:38.300 --> 00:05:45.200 Fortran. So right from the start, there was this discussion about Algol, no Fortran, no Algol, 00:05:45.200 --> 00:05:52.760 Pascal. And somehow that interested me. I always stayed on the Algol and Pascal side, actually. 00:05:52.760 --> 00:05:57.880 I was told that when I was in college that Fortran was the most important language I'd ever learned in 00:05:57.880 --> 00:06:02.420 my career and I should just focus on that. And I was pleading to take some C and C++. I'm like, 00:06:02.420 --> 00:06:07.200 can I do where the, nope, Fortran is where you got to start. You can do those as an elective afterwards. 00:06:07.200 --> 00:06:15.140 Oh gosh, yeah. Well, it wasn't, in our university, it was more like a split between the math department and the physics department. 00:06:15.140 --> 00:06:25.320 The natural sciences were using Fortran because they were sort of processing measurements and math people were more interested in sort of pure computer science. 00:06:25.320 --> 00:06:26.540 Yeah, very interesting. Yeah. 00:06:26.540 --> 00:06:35.620 The professors actually sort of crossed that bridge. So later, I actually, I encountered the author of ABC, 00:06:35.620 --> 00:06:45.240 Lombard Meartens. I encountered him in my personal life, sort of an extracurricular activity where I was helping some volunteer group doing programming. 00:06:45.240 --> 00:06:50.440 And he was also helping that same group. He was something higher in the organization. 00:06:50.440 --> 00:06:59.520 And he realized that I was a good programmer and had interest in programming languages and their design and implementation and had some, 00:06:59.520 --> 00:07:06.220 some interesting skills there. And when I was about to graduate, he just offered me a job. 00:07:06.360 --> 00:07:07.540 And that was that. 00:07:07.540 --> 00:07:15.880 That's fantastic, right? You're probably thinking, how am I going to find a job? What am I going to apply? What am I going to do? Oh, it just landed in my lab. How wonderful. 00:07:15.880 --> 00:07:25.040 And then I learned really what the job of a language designer and implementer is. And I didn't get to design any part of ABC. 00:07:25.040 --> 00:07:34.200 I got to discuss it with Lambert and other team members endlessly. But basically, the design was already complete when I joined the team. 00:07:34.620 --> 00:07:42.000 And all they needed was someone who would implement it. But by fighting every aspect of the language that I didn't understand, 00:07:42.000 --> 00:07:51.340 I prompted Lambert and others to explain what their reasoning process in the language design phase was. 00:07:51.340 --> 00:07:55.740 And that helped me learn how to be a language designer. 00:07:55.740 --> 00:07:59.540 Yeah, I'm sure it did, because they designed it more or less in the abstract, right? 00:07:59.540 --> 00:08:06.420 And they said, all right, now it's time for a rubber to hit the road. And you'd actually make this thing work that we've specced out, right? 00:08:06.420 --> 00:08:07.800 Yeah, much like that. 00:08:07.800 --> 00:08:12.580 Yeah. And how much of that experience do you feel like made it possible for you to actually create Python? 00:08:12.580 --> 00:08:18.840 Like, to me, thinking of I'm going to create a language, I'm going to create the CPython implementation and the standard library and all that. 00:08:18.840 --> 00:08:20.700 It's very daunting and a very big challenge. 00:08:20.840 --> 00:08:26.380 But do you feel like you kind of got a first round practice at it, doing this thing at ABC? 00:08:26.380 --> 00:08:34.540 Absolutely. Without having been on the ABC team for four years, I would never have been able to do that. 00:08:34.540 --> 00:08:37.660 I wouldn't have felt comfortable. I wouldn't have known how to. 00:08:38.500 --> 00:08:43.600 I wouldn't have known enough about language implementations. 00:08:43.600 --> 00:08:58.860 I mean, Python really is sort of the next version of ABC with all the things that were great about ABC retained and all the things I thought were not so successful in ABC removed. 00:08:58.860 --> 00:09:02.620 And a very small number of my own ideas replacing them. 00:09:02.620 --> 00:09:10.460 I see. So Python was sort of your, let me do this now knowing what I know, let me make a better version of something similar to this. 00:09:10.460 --> 00:09:14.600 Correct. Yeah, that's exactly what I was really thinking. 00:09:14.600 --> 00:09:20.420 And I sort of, I managed to present it to management in a slightly more objective fashion. 00:09:20.420 --> 00:09:23.180 How do you pitch it? 00:09:23.180 --> 00:09:30.260 Well, part of it was that management was not very closely involved in my day-to-day activities. 00:09:30.420 --> 00:09:32.800 There was a certain amount of software that had to be written. 00:09:32.800 --> 00:09:41.940 And however it was written was great, as long as we had sort of the running applications to prove it at the end. 00:09:41.940 --> 00:09:51.480 I was more or less at liberty to invent a different strategy for sort of eventually building that software faster. 00:09:51.480 --> 00:09:54.020 That's fantastic. And that's some of the origins of Python. 00:09:54.020 --> 00:09:56.840 What happened to ABC? It's not around. 00:09:56.840 --> 00:10:00.100 And Python is one of the most popular programming languages in the world. 00:10:00.100 --> 00:10:02.320 Like, why did those things take different paths? 00:10:02.320 --> 00:10:06.760 I've written about this in my old Python history blogs a few times. 00:10:06.760 --> 00:10:15.780 The ABC project actually four years into it, at least for me four years into it, was canceled by upper management at CWI. 00:10:16.120 --> 00:10:24.220 The reason being that there was no observable sort of user uptake. 00:10:24.220 --> 00:10:28.420 There were very few people interested in the language. 00:10:28.420 --> 00:10:32.240 There were even fewer people who are actually using it. 00:10:32.240 --> 00:10:36.340 And the team just couldn't move that needle. 00:10:36.340 --> 00:10:41.180 And part of that was that this was well before the Internet. 00:10:41.180 --> 00:10:43.700 And there was a little bit of Usenet. 00:10:43.700 --> 00:10:50.960 But it was difficult to distribute a language implementation and get people to use it. 00:10:50.960 --> 00:10:52.060 Right. There's no GitHub. 00:10:52.060 --> 00:10:54.700 There's not even SourceForge, right? 00:10:54.700 --> 00:10:55.920 There's no web. 00:10:55.920 --> 00:10:59.800 There's so many of the pieces that make these work. 00:11:00.060 --> 00:11:01.420 It's worse than that, even. 00:11:01.420 --> 00:11:07.540 There was no electronic way to distribute the source code at all when we got started. 00:11:07.720 --> 00:11:17.140 I remember taking a long vacation to the United States with a nine-track computer tape in my luggage. 00:11:17.140 --> 00:11:22.500 And taking that tape to two or three different places in the U.S. 00:11:22.500 --> 00:11:30.540 where there were people interested in using ABC so that they could load that tape onto their computer and get the sources. 00:11:31.180 --> 00:11:35.700 Because the amount of source code was larger than you could possibly send as an email. 00:11:35.700 --> 00:11:40.700 And I think attachments hadn't been invented yet or were still limited in size. 00:11:40.700 --> 00:11:43.020 Yeah, you have to, like, base 64 and code it. 00:11:43.020 --> 00:11:44.560 Just put it as text or something, right? 00:11:44.560 --> 00:11:47.820 I'm going to send you a thousand emails numbered 1, 2, 3, 4. 00:11:47.820 --> 00:11:49.400 Yeah. 00:11:49.400 --> 00:11:53.840 Even the first Python distribution had to suffer through some of that. 00:11:53.840 --> 00:12:03.760 But by then, in 91, it was about 20 compressed maximum-sized messages to some Usenet source code group. 00:12:03.760 --> 00:12:05.980 Yeah, Usenet was just starting to become popular. 00:12:05.980 --> 00:12:08.960 And the internet was starting to be a thing. 00:12:08.960 --> 00:12:12.260 But web browsers didn't really come out until, like, 93, 94. 00:12:12.260 --> 00:12:14.120 So it was quite early days in 91. 00:12:14.120 --> 00:12:16.700 Mm-hmm. 00:12:16.700 --> 00:12:19.740 Talk Python To Me is partially supported by our training courses. 00:12:19.740 --> 00:12:22.720 How does your team keep their Python skills sharp? 00:12:23.080 --> 00:12:26.860 How do you make sure new hires get started fast and learn the Pythonic way? 00:12:26.860 --> 00:12:30.980 If the answer is a series of boring videos that don't inspire, 00:12:30.980 --> 00:12:35.840 or a subscription service you pay way too much for and use way too little, listen up. 00:12:35.840 --> 00:12:39.760 At Talk Python Training, we have enterprise tiers for all of our courses. 00:12:39.760 --> 00:12:44.320 Get just the one course you need for your team with full reporting and monitoring. 00:12:44.320 --> 00:12:47.420 Or ditch that unused subscription for our course bundles, 00:12:47.420 --> 00:12:50.960 which include all the courses and you pay about the same price as a subscription. 00:12:50.960 --> 00:12:52.040 Once. 00:12:52.320 --> 00:12:59.340 For details, visit training. talkpython.fm/business or just email sales at talkpython.fm. 00:12:59.340 --> 00:13:03.820 Did you envision Python being open source from the beginning? 00:13:03.820 --> 00:13:05.840 What was your thinking around that? 00:13:06.100 --> 00:13:07.220 I would say yes. 00:13:07.220 --> 00:13:13.500 ABC actually, in the sense that the concept even existed, was meant to be open source. 00:13:13.500 --> 00:13:17.060 We were not interested in selling it. 00:13:17.060 --> 00:13:19.800 We were just interested in promoting the language. 00:13:19.800 --> 00:13:26.740 If the words open source had existed in the early 80s, we would have said ABC is open source. 00:13:27.120 --> 00:13:32.820 As it was, I don't think we had even realized that there was some kind of need of a license. 00:13:32.820 --> 00:13:33.300 Right. 00:13:33.300 --> 00:13:33.840 That's amazing. 00:13:33.840 --> 00:13:36.400 Just the nomenclature didn't even exist to describe. 00:13:36.400 --> 00:13:36.860 Yeah. 00:13:36.860 --> 00:13:38.620 You had to use words in a description. 00:13:38.620 --> 00:13:40.040 This is a thing we're giving away. 00:13:40.040 --> 00:13:41.080 You don't have to... 00:13:41.080 --> 00:13:41.640 So on, right? 00:13:41.720 --> 00:13:47.460 By the time I started on Python, there were a few models that were pretty solid. 00:13:47.460 --> 00:13:55.140 I remember, I think the big model that I just copied almost literally was the MIT license 00:13:55.140 --> 00:13:58.160 that went on X Windows. 00:13:58.160 --> 00:14:00.420 Now, you're not supposed to call it X Windows. 00:14:00.420 --> 00:14:01.360 It was X11. 00:14:02.160 --> 00:14:09.100 But that was a big, essentially open source piece of free software that was very sort of 00:14:09.100 --> 00:14:16.960 intentionally open source because the authors wanted to unify windowing software across different 00:14:16.960 --> 00:14:18.360 hardware platforms. 00:14:18.360 --> 00:14:19.840 There were all these different flavors. 00:14:19.840 --> 00:14:22.700 How do you write apps that run on all of them and things like that? 00:14:22.700 --> 00:14:23.620 That was a big problem, huh? 00:14:23.620 --> 00:14:24.360 Exactly. 00:14:24.360 --> 00:14:24.960 Okay. 00:14:24.960 --> 00:14:31.580 So how do you think having Python open source has helped or maybe even hindered Python over 00:14:31.580 --> 00:14:32.080 the years? 00:14:32.080 --> 00:14:33.340 Oh, it's only helped. 00:14:33.340 --> 00:14:40.100 I mean, if it hadn't been open source, people would not have been interested in picking it 00:14:40.100 --> 00:14:46.920 up because it's one thing to download an application that's not free software, that's not open source 00:14:46.920 --> 00:14:49.940 and use it to process some of your data. 00:14:49.940 --> 00:14:58.520 But it's quite different to start writing your own software using a language that hasn't proven 00:14:58.520 --> 00:14:59.340 itself yet. 00:14:59.340 --> 00:14:59.920 Right. 00:15:00.720 --> 00:15:05.620 So if maybe you're willing to pay for one of the top three most, the tooling and whatnot 00:15:05.620 --> 00:15:09.920 for the top three most popular languages, because you know that's kind of where a lot of the 00:15:09.920 --> 00:15:10.740 momentum is. 00:15:10.740 --> 00:15:17.500 But in order to break into that space, being open and free really allowed you to wedge yourself 00:15:17.500 --> 00:15:18.060 in there, right? 00:15:18.140 --> 00:15:24.760 And it was also the model that some languages that I felt I was competing with most directly, 00:15:24.760 --> 00:15:28.640 like Perl and Tickle TK, those were also open source. 00:15:28.880 --> 00:15:29.980 I do not recall any names. 00:15:29.980 --> 00:15:35.460 I do not recall any names, but I recall there were several similar level scripting languages 00:15:35.460 --> 00:15:42.640 being designed and distributed at the time that were not open source and very intentionally 00:15:42.640 --> 00:15:47.400 so, where there was someone who had a brilliant idea for making a better scripting language. 00:15:47.400 --> 00:15:52.980 And for all intents and purposes, their language probably was better than what was available. 00:15:52.980 --> 00:15:59.380 But their model for funding their work was sell copies of the interpreter. 00:16:00.140 --> 00:16:02.360 And that just never worked. 00:16:02.360 --> 00:16:02.720 Yeah. 00:16:02.720 --> 00:16:07.280 In those early days, it was still not entirely clear what business models would work for the 00:16:07.280 --> 00:16:09.100 developer community and what wouldn't. 00:16:09.100 --> 00:16:10.340 People were really experimenting. 00:16:10.340 --> 00:16:14.040 And I'm sure some things were lost because bad choices were made. 00:16:14.040 --> 00:16:17.800 But I'm really glad Python is still going strong. 00:16:17.800 --> 00:16:25.400 Did you ever imagine back in 1991 that open source and Python would be where they are today? 00:16:25.400 --> 00:16:30.880 I, in general, suffer from a terrible lack of imagination and vision. 00:16:30.880 --> 00:16:38.720 So at no point in Python's history have I ever adequately predicted where Python would be 00:16:38.720 --> 00:16:40.420 five years from there. 00:16:40.420 --> 00:16:44.780 So no, I had never thought that this would happen this particular way. 00:16:44.780 --> 00:16:46.720 To me as well, it's just amazing. 00:16:46.720 --> 00:16:51.860 Just even over the last five years, the way things have changed is so amazing. 00:16:51.860 --> 00:16:53.460 GitHub has come into existence. 00:16:54.020 --> 00:17:00.760 We're seeing companies that used to be fiercely proprietary become much more embracing of 00:17:00.760 --> 00:17:01.360 open source. 00:17:01.360 --> 00:17:07.640 I recently saw the CEO of the Linux Foundation or the head of the Linux Foundation standing 00:17:07.640 --> 00:17:12.400 next to Satya Nadella at a Microsoft conference saying they love each other. 00:17:12.400 --> 00:17:12.620 Yeah. 00:17:12.620 --> 00:17:16.300 This is a different place than we were a while ago. 00:17:16.300 --> 00:17:18.080 I think it's great for everyone, though. 00:17:18.080 --> 00:17:20.520 I think it's a very, very positive path forward. 00:17:20.520 --> 00:17:23.800 I'm very happy with how this all has turned out. 00:17:23.800 --> 00:17:28.700 I'm hopeful that a lot of technology will continue to be open source. 00:17:28.700 --> 00:17:29.720 Yeah, I think it will. 00:17:29.720 --> 00:17:35.260 And I think it's great to see companies taking open source and building business models alongside 00:17:35.260 --> 00:17:36.540 of it that are sustainable. 00:17:36.540 --> 00:17:43.140 Companies like Continuum or Scraping Web or these guys that have a great popular open source 00:17:43.140 --> 00:17:49.540 project and somehow they're adding value on top of it, but they're not abandoning open source. 00:17:49.540 --> 00:17:54.680 Without open source, you have to do all the work yourself as the company that owns it. 00:17:55.220 --> 00:18:03.360 And now, maybe if you're a large company like Google or IBM or Microsoft or Apple, you don't mind because you have tons of developers. 00:18:03.360 --> 00:18:18.680 But for anyone who is smaller, the value of a community is so tremendous because you sort of, you will still be able to make money on a variety of consulting and support projects. 00:18:19.040 --> 00:18:23.480 So that's how open source developers support themselves generally. 00:18:23.480 --> 00:18:24.760 I'm actually an exception. 00:18:24.760 --> 00:18:29.740 I'm just employed by some large software developer that uses a lot of Python. 00:18:29.740 --> 00:18:32.040 That's been my personal model. 00:18:32.040 --> 00:18:35.820 But yeah, companies like Continuum or Canonical. 00:18:35.820 --> 00:18:36.640 Canonical. 00:18:36.640 --> 00:18:51.340 Everything they produce is open source, but they make a lot of money through handholding of customers who don't want to sort of hire their own software developers. 00:18:51.340 --> 00:18:56.620 That's a model that works for many types of open source software. 00:18:56.620 --> 00:19:02.120 I think it's great to see people being successful in that model and experimenting with other ones as well. 00:19:02.120 --> 00:19:05.320 Let's talk a little bit about language design and trade-offs. 00:19:05.320 --> 00:19:05.920 Sure. 00:19:05.920 --> 00:19:12.060 Python has been growing in popularity pretty dramatically over the last 5, 10 years. 00:19:12.060 --> 00:19:14.620 It had been around for 15 years before. 00:19:14.620 --> 00:19:20.700 And it seems like not only is it still relevant and popular, but that popularity and relevance is growing. 00:19:20.860 --> 00:19:33.540 And I think part of that is expanding into different areas, like the adoption of Python in the data science space, I think, has brought many new people to the Python ecosystem where Python is their primary language now. 00:19:33.540 --> 00:19:38.160 How do you trade off sort of serving these different environments or these different ecosystems? 00:19:38.160 --> 00:19:43.860 Somebody wants in a language as a data scientist might be very different than what somebody wants as a web developer. 00:19:43.860 --> 00:19:51.960 On the language design side, I don't usually take applications into account that much. 00:19:51.960 --> 00:20:03.740 I've seen some language designs where people proudly announced, well, in our language, a URL is a standard data structure that is built into the compiler. 00:20:03.740 --> 00:20:06.260 They say that as if that's a good thing. 00:20:06.260 --> 00:20:09.580 And all it buys you is that you can leave out some string quotes. 00:20:09.580 --> 00:20:11.740 And internally, it usually is still a string. 00:20:11.740 --> 00:20:12.600 Right, exactly. 00:20:12.840 --> 00:20:13.580 It's a gimmick. 00:20:13.580 --> 00:20:18.460 And Python has always presented itself as a general purpose language. 00:20:18.460 --> 00:20:20.360 And you can do many things in Python. 00:20:20.360 --> 00:20:28.880 And I didn't design Python for web development, obviously, because the language is older than the concept of web development. 00:20:28.880 --> 00:20:32.400 And ditto for data science. 00:20:32.400 --> 00:20:48.420 And for web development, it turns out enough people from different backgrounds are interested in doing simple web development using Python that we ended up with a bunch of stuff in the standard library. 00:20:48.620 --> 00:20:57.940 But still, often the most successful APIs for even for web development are actually third-party packages. 00:20:58.400 --> 00:21:02.920 And what the standard library provides is more low-level than that. 00:21:03.020 --> 00:21:07.200 Like the standard library has to provide things like sockets. 00:21:07.640 --> 00:21:18.880 And in fact, one funny story is that I think in the first year that Python existed, before we made it open source, actually, I was teaching myself how sockets worked. 00:21:19.380 --> 00:21:25.280 Because sockets were sort of a new thing in our environment at that point. 00:21:25.280 --> 00:21:27.600 We had a bunch of Unix machines. 00:21:27.600 --> 00:21:32.540 And I had never really known how the networking on those machines worked before. 00:21:32.920 --> 00:21:37.660 And then some colleagues started writing little C programs that used sockets. 00:21:37.660 --> 00:21:38.860 And it was so cool. 00:21:38.860 --> 00:21:42.300 And the programs always crashed because they didn't do the right error checking. 00:21:42.300 --> 00:21:45.380 And I wanted to know what those sockets were about. 00:21:45.380 --> 00:21:53.360 And my way of teaching myself was, oh, I'll write a Python extension that wraps the socket API. 00:21:53.360 --> 00:22:00.920 And so I just read the man pages and say, okay, well, there is like the socket call and the bind call and the listen call. 00:22:01.540 --> 00:22:08.660 And I wrapped each of those in as low-level an extension as possible with proper error checking. 00:22:08.660 --> 00:22:12.620 Because Python has always had this philosophy, if something goes wrong, you get an exception. 00:22:12.620 --> 00:22:26.180 And then I started combining those calls in sort of random combinations and figuring out what errors I would get when and what forms of simple programs actually worked. 00:22:26.180 --> 00:22:30.420 And that's how Python socket module came into existence. 00:22:30.760 --> 00:22:38.060 And later, once the worldwide web started being promoted, I joined a variety of mailing lists around that. 00:22:38.440 --> 00:22:42.800 And started writing my own little web servers and started writing my own little web servers and clients. 00:22:42.800 --> 00:22:45.420 And eventually that turned into URL app. 00:22:45.420 --> 00:22:55.840 But what people actually use for web stuff is third-party frameworks like Twisted or for web serving, pure web serving. 00:22:55.840 --> 00:22:58.220 It's like Django or Flask. 00:22:58.220 --> 00:23:00.900 For web clients, it's requests. 00:23:02.000 --> 00:23:08.560 And so the standard library doesn't even contribute that much beyond the sort of the low-level sockets. 00:23:08.560 --> 00:23:09.120 Sure. 00:23:09.120 --> 00:23:15.900 So I totally agree that it's the packages that have absolutely made Python successful. 00:23:15.900 --> 00:23:23.240 I mean, just looking on PyPI.org right now, we've got 96,000 plus packages. 00:23:23.680 --> 00:23:30.620 And that's really a testament to how amazing the whole ecosystem and community is, right? 00:23:30.620 --> 00:23:37.700 How do you decide when something should be in the standard library or something should be an external package? 00:23:37.700 --> 00:23:39.900 And have these ever moved in or out? 00:23:40.440 --> 00:23:45.460 Yeah, I have to admit that the standard library is still fairly uneven. 00:23:45.460 --> 00:23:54.620 Because long ago, say 15 years ago, I had much less of a filter about what things would be good standard library modules. 00:23:54.620 --> 00:23:58.980 I had very strong backwards compatibility requirements. 00:23:58.980 --> 00:24:06.500 Like once a module is in the standard library, it can sort of grow, but it can't just change in an incompatible way. 00:24:06.500 --> 00:24:10.380 Or at least that's a major project with deprecations and all that. 00:24:10.380 --> 00:24:16.160 But I didn't have much of a bar for including stuff in the standard library. 00:24:16.160 --> 00:24:24.420 And that started out with various hobby projects that I wrote myself and played with for two months in 1991 or so. 00:24:24.420 --> 00:24:29.180 That were still in the Python 2 standard library around 2010. 00:24:29.180 --> 00:24:34.920 Many of those things finally got ripped out for Python 3. 00:24:34.920 --> 00:24:36.880 But of course, Python 2 is still there. 00:24:37.500 --> 00:24:41.800 More recently, sort of, as Python matured and became more popular. 00:24:41.800 --> 00:24:45.780 And from my perspective, it's just sort of been steady exponential growth. 00:24:45.780 --> 00:24:52.440 So I can't tell you whether it was 5 or 10 or 20 years ago that Python suddenly started becoming popular. 00:24:52.440 --> 00:25:04.700 So the current rules for inclusion in the standard library is a combination of something that is useful for multiple application areas. 00:25:04.700 --> 00:25:11.940 A new API for web development will not make it into the standard library just because that's one area. 00:25:12.180 --> 00:25:16.840 It has to be something that's useful for a wide variety of applications. 00:25:16.840 --> 00:25:19.000 It doesn't necessarily have to be all applications. 00:25:19.000 --> 00:25:24.940 But something like, well, sockets are obviously something that is useful across the board. 00:25:24.940 --> 00:25:25.320 Right. 00:25:25.320 --> 00:25:32.060 Many things in the standard library that really belong there also have to do with sort of the language itself. 00:25:32.060 --> 00:25:39.140 Like introspection tools, partial functions, those kind of things are good standard library things. 00:25:39.140 --> 00:25:40.240 The disk module. 00:25:40.240 --> 00:25:41.540 For example, yeah. 00:25:41.540 --> 00:25:41.980 Yeah. 00:25:41.980 --> 00:25:52.300 Things that are bad for inclusion in the standard library is usually almost any piece of code that is under active development. 00:25:52.560 --> 00:25:59.440 Because Python only issues a sort of feature release every 18 months or more. 00:25:59.440 --> 00:26:05.080 And that's just a really slow pace and you can't even get people to upgrade quickly. 00:26:05.080 --> 00:26:11.640 So if someone has a new idea for, let's take an example. 00:26:11.640 --> 00:26:19.660 We have async.io, which is in the standard library, but it doesn't have a built-in web framework. 00:26:19.660 --> 00:26:23.840 Well, why is there no async.io based web framework in the standard library? 00:26:23.840 --> 00:26:36.440 Because the async.io based web framework that exists is under very active development and constantly changes and not always incompatible ways. 00:26:36.440 --> 00:26:46.940 And so people would just be much worse off being stuck with whatever was the async.io based web framework around Python 3.5. 00:26:46.940 --> 00:26:50.440 You would actually hinder it by putting it in Python, right? 00:26:50.440 --> 00:26:51.280 Exactly. 00:26:51.280 --> 00:26:55.800 You would force it to freeze its API so that could never change in a breaking way. 00:26:55.800 --> 00:26:59.340 And it couldn't release more than every 18 months. 00:26:59.340 --> 00:27:05.260 And probably some third-party package would come along, mimic that, but iterate faster and be better anyway. 00:27:05.260 --> 00:27:05.800 Yeah. 00:27:05.800 --> 00:27:06.300 Okay. 00:27:06.300 --> 00:27:06.300 Okay. 00:27:06.300 --> 00:27:23.560 This portion of Talk Python To Me has been brought to you by Rollbar. 00:27:23.880 --> 00:27:27.080 One of the frustrating things about being a developer is dealing with errors. 00:27:27.080 --> 00:27:27.660 Ah. 00:27:27.660 --> 00:27:35.260 Relying on users to report errors, digging through log files trying to debug issues, or a million alerts just flooding your inbox and ruining your day. 00:27:35.260 --> 00:27:42.480 With Rollbar's full-stack error monitoring, you'll get the context, insights, and control that you need to find and fix bugs faster. 00:27:43.020 --> 00:27:44.200 It's easy to install. 00:27:44.200 --> 00:27:48.440 You can start tracking production errors and deployments in eight minutes or even less. 00:27:48.440 --> 00:27:57.800 Rollbar works with all the major languages and frameworks, including the Python ones, such as Django, Flask, Pyramid, as well as Ruby, JavaScript, Node, iOS, and Android. 00:27:58.480 --> 00:28:07.020 You can integrate Rollbar into your existing workflow, send error alerts to Slack or HipChat, or even automatically create issues in Jira, Pivotal Tracker, and a whole bunch more. 00:28:07.020 --> 00:28:10.400 Rollbar has put together a special offer for Talk Python To Me listeners. 00:28:10.400 --> 00:28:16.080 Visit rollbar.com slash Talk Python To Me, sign up, and get the bootstrap plan free for 90 days. 00:28:16.080 --> 00:28:19.120 That's 300,000 errors tracked all for free. 00:28:19.120 --> 00:28:22.840 But hey, just between you and me, I really hope you don't encounter that many errors. 00:28:23.420 --> 00:28:28.840 Loved by developers at awesome companies like Heroku, Twilio, Kayak, Instacart, Zendesk, Twitch, and more. 00:28:28.840 --> 00:28:30.460 Give Rollbar a try today. 00:28:30.460 --> 00:28:33.060 Go to rollbar.com slash Talk Python To Me. 00:28:33.060 --> 00:28:48.000 Has anything been brought in to Python that was originally external that you can think of? 00:28:48.000 --> 00:28:52.320 I don't have specific examples in mind, but it's definitely happened. 00:28:52.540 --> 00:28:55.260 That definitely occasionally happens. 00:28:55.260 --> 00:29:05.260 It's usually more of a threat where we say, if someone comes with an idea and they say, this functionality should really be in the standard library. 00:29:05.260 --> 00:29:09.840 And nowadays, we usually say, well, that looks more like an application. 00:29:09.840 --> 00:29:12.180 And how do we know that it's actually useful? 00:29:12.680 --> 00:29:23.320 Prove us that the design you have in mind works by first releasing it as a third-party package on PyPI and show us how popular that is. 00:29:23.320 --> 00:29:26.520 And then there are also maintenance requirements. 00:29:26.520 --> 00:29:35.300 We would require that a contributor commit to several years of keeping that code up to date and fixing bugs and stuff. 00:29:35.300 --> 00:29:35.760 Sure. 00:29:36.080 --> 00:29:45.520 And we've had things that actually were kicked out of the standard library because they were too much of a maintenance burden for the core development team. 00:29:45.520 --> 00:29:52.220 I think the last time that happened might have been BSDDB or BerkeleyDB, which was a very large package. 00:29:52.420 --> 00:29:57.340 It's currently much happier as a third-party package than it ever was in the standard library. 00:29:57.340 --> 00:30:00.760 Yeah, I'm sure it can, like we said, grow much faster and so on. 00:30:00.760 --> 00:30:01.900 There are databases in there. 00:30:01.900 --> 00:30:03.600 For example, SQLite ships with Python. 00:30:03.600 --> 00:30:18.420 SQLite is one of those things that is so popular and so versatile and so useful for so many different application domains that that was definitely the right decision to include that. 00:30:18.540 --> 00:30:21.540 Also, SQLite itself is incredibly stable. 00:30:21.540 --> 00:30:22.760 Yeah, that's one of the things. 00:30:22.760 --> 00:30:24.500 It's not changing much these days. 00:30:24.500 --> 00:30:25.320 It's very reliable. 00:30:25.320 --> 00:30:26.180 That's great. 00:30:26.180 --> 00:30:26.600 Yeah. 00:30:26.600 --> 00:30:35.380 So speaking of contributors and having people commit to a certain amount of support over time if something's going to come into the standard library, 00:30:35.380 --> 00:30:41.080 how do you ensure that the core development community invites and retains the best contributors? 00:30:41.080 --> 00:30:48.200 You know, I have a lot of respect for the core developers, but how do you make sure that ecosystem is healthy and vibrant? 00:30:48.420 --> 00:30:51.080 I don't think we're doing that very consciously. 00:30:51.080 --> 00:31:03.860 We have a nominal mentorship program in place, but it's mostly used to get people thinking about contributing to open source projects in general. 00:31:04.300 --> 00:31:09.020 I don't think that's where we get most of our new core developers. 00:31:09.020 --> 00:31:26.020 In practice, new core developers almost always happen because somebody has an itch to scratch and they happen to be a really good programmer or at least sort of made a serious study of Python and start contributing. 00:31:26.680 --> 00:31:38.240 And the people who review their code, and sometimes that's just one or two core devs who sort of mentor that one person, find that their contributions are of high value. 00:31:38.340 --> 00:31:54.780 And then at some point, the mentor or one of the mentors proposes, hey, Python dev, core developers, what do you think of giving person X commit bit so that they can commit their own code after it's been reviewed? 00:31:55.000 --> 00:31:55.660 Yeah. 00:31:55.660 --> 00:32:12.020 There's often real discussion about the sort of potential new contributors maturity, not just in terms of their pure programming chops and how well they know CPython or Python or whatever their area of contribution is. 00:32:12.160 --> 00:32:16.020 But also, do they have the right character? 00:32:16.020 --> 00:32:21.980 Are they likely to sort of not commit something when the reviewer says that it's not ready? 00:32:21.980 --> 00:32:22.420 Yeah. 00:32:22.420 --> 00:32:26.880 I suspect one aspect of that is there's a lot of people that come into programming. 00:32:26.880 --> 00:32:37.260 They see a shiny new thing, you know, something with Node.js or some other technology that seems like it's just going to take over the industry and then it doesn't necessarily. 00:32:37.480 --> 00:32:45.420 So there's probably a level of maturity of experiencing tech evolution over the long term so that you can bring that to the language, right? 00:32:45.420 --> 00:32:47.440 That would be too high a bar. 00:32:47.440 --> 00:32:54.780 You can't say people can only start contributing once they've gone through one technology boom and bust cycle. 00:32:54.780 --> 00:33:02.000 You need them to have a certain character that is okay with the pace of things and the thoroughness. 00:33:02.000 --> 00:33:07.000 And you definitely want them to have a very good attention to detail. 00:33:07.380 --> 00:33:07.460 Sure. 00:33:07.460 --> 00:33:19.700 I guess you definitely want a different level of attention to detail and being meticulous when you're working on the internals of CPython rather than if you're working on some package that's used by a thousand people a month. 00:33:19.700 --> 00:33:20.380 Right? 00:33:20.380 --> 00:33:22.360 These have different requirements for that kind of stuff. 00:33:22.360 --> 00:33:27.900 So you've been a champion of diversity in the whole Python ecosystem. 00:33:27.900 --> 00:33:37.420 And I just wanted to point that out and say thank you because I think it's really making the Python space better in some tangible ways than other environments. 00:33:37.740 --> 00:33:44.160 Just by way of a story, I went to a conference in London last year and I took my 16-year-old daughter. 00:33:44.160 --> 00:33:45.960 And this was not a Python conference. 00:33:45.960 --> 00:33:47.080 It was decidedly not. 00:33:47.080 --> 00:33:49.660 But I was speaking on Python to these folks. 00:33:49.660 --> 00:33:50.780 Say, hey, you should also learn Python. 00:33:50.780 --> 00:33:51.300 This is cool. 00:33:51.660 --> 00:33:53.160 And I brought my daughter. 00:33:53.160 --> 00:33:54.120 We went to the speaker dinner. 00:33:54.120 --> 00:33:57.100 And there was 28 or 29 speakers there. 00:33:57.100 --> 00:33:58.980 And I think one or two women. 00:33:58.980 --> 00:34:02.560 And my daughter looked at me after a while and goes, where are all the women? 00:34:03.220 --> 00:34:06.560 And I said, you know, this is a sad part of the tech industry. 00:34:06.560 --> 00:34:08.980 And when I go to Python conferences, I don't feel that way. 00:34:08.980 --> 00:34:10.440 And, you know, I just think that's great. 00:34:10.440 --> 00:34:15.720 I've personally always been a feminist, although never a radical one. 00:34:15.720 --> 00:34:20.600 Encouraging women has been a pretty natural thing for me. 00:34:20.600 --> 00:34:25.640 I think there was a specific series of events. 00:34:25.640 --> 00:34:28.140 I'm not entirely sure when it was. 00:34:28.200 --> 00:34:33.020 But I remember that there was some upset maybe around a decade ago at OSCOM. 00:34:33.020 --> 00:34:48.060 Some people started pointing out that the open source community was, despite all its sort of pride in, well, we care about results, rough consensus and working code and all that. 00:34:48.060 --> 00:34:50.160 There was not a very diverse community. 00:34:50.620 --> 00:34:58.520 And the number of women that someone counted at OSCOM, I believe, was well below even the already low industry level. 00:34:58.520 --> 00:35:01.260 Somehow that was a wake-up call for me. 00:35:01.260 --> 00:35:05.240 There had always been a few active women in the Python community. 00:35:05.240 --> 00:35:10.280 And I had never really counted how many there were or if it was always the same, two or three. 00:35:10.520 --> 00:35:25.620 But because of that discussion at OSCOM that I did not participate or even witness in person, but that's through various blogs and peripheral vision came to my attention, I thought, hmm, what's the situation in Python community? 00:35:25.620 --> 00:35:27.940 Actually, we're not doing great. 00:35:28.480 --> 00:35:39.520 So I sort of made a mental note of, well, maybe we ought to think of how to increase the participation, the diversity in the Python community. 00:35:39.520 --> 00:35:54.640 And I think that the main consequence of that was that when there were actually women who came to the Python community or were already part of the Python community and said, we would like to do something for women. 00:35:54.640 --> 00:35:57.220 We would like this community to be more welcoming. 00:35:57.220 --> 00:35:59.380 I thought, that's a great idea. 00:35:59.380 --> 00:36:00.420 Let's do that. 00:36:00.420 --> 00:36:11.660 Rather than responding in a way that some other tech communities have responded by feeling threatened or, oh, we don't have to do anything. 00:36:11.660 --> 00:36:14.880 If you're a good coder, you're a good coder, and then you're welcome. 00:36:14.880 --> 00:36:19.580 And if you're not a good coder, you're not welcome, regardless of whether you're a man or a woman. 00:36:19.580 --> 00:36:26.720 That's actually a fair amount of bullshit in that attitude because those people don't realize how much bias there is. 00:36:26.720 --> 00:36:42.460 So I've always been very open to the PSF and PyCon and various groups to try and reverse that bias by giving sort of diversity funding to PyCon attendees, for example. 00:36:42.460 --> 00:36:46.260 I think I should call out specifically Jessica McKellar. 00:36:46.260 --> 00:36:47.300 Yeah, I was going to bring her up. 00:36:47.300 --> 00:36:53.020 Who has been a champion of this for a long time, has given some fabulous keynotes about this. 00:36:53.760 --> 00:36:59.620 And all I had to do was be supportive there, I felt, which came naturally. 00:36:59.620 --> 00:37:08.460 I remember in real world politics, since I was a voting age, I've always made a point of voting for a woman whenever I could. 00:37:08.460 --> 00:37:12.020 I think the progress that's been made is really fabulous. 00:37:12.020 --> 00:37:14.580 And you're right that Jessica deserves a lot of credit for that. 00:37:14.580 --> 00:37:16.060 She's been fabulous at it. 00:37:16.060 --> 00:37:21.380 We're in a much better place than we were before with regard to that. 00:37:21.460 --> 00:37:23.240 Although I still think there's more work to do. 00:37:23.240 --> 00:37:23.900 Absolutely. 00:37:23.900 --> 00:37:25.740 I'm super happy to see the direction. 00:37:25.740 --> 00:37:27.280 There's always more work to do. 00:37:27.280 --> 00:37:29.760 And diversity is not just about women either. 00:37:29.760 --> 00:37:30.620 Absolutely. 00:37:30.620 --> 00:37:31.400 Let's remember that. 00:37:31.400 --> 00:37:31.740 Yep. 00:37:31.740 --> 00:37:32.060 Yep. 00:37:32.060 --> 00:37:32.500 Absolutely. 00:37:32.500 --> 00:37:33.080 Yeah. 00:37:33.080 --> 00:37:36.080 It was just the top of my mind because my daughter was so struck. 00:37:36.080 --> 00:37:40.660 I took her to this tech conference thinking, oh, I'd really like to show her how cool the tech scene is and stuff. 00:37:40.660 --> 00:37:47.280 And she did come away with that, but she also came away with a little bit of the feeling like maybe this is not for women, which made me a little bit. 00:37:47.280 --> 00:37:49.300 I just felt bad about it, right? 00:37:49.300 --> 00:37:50.520 Oh, that's terrible. 00:37:50.520 --> 00:37:51.140 Yeah. 00:37:51.140 --> 00:37:51.840 Yeah, it is. 00:37:51.840 --> 00:37:52.140 It is. 00:37:52.820 --> 00:37:53.340 All right. 00:37:53.340 --> 00:37:55.000 So let's shift gears a little bit. 00:37:55.000 --> 00:38:11.200 And, you know, one of the things that there's a lot of talk about right now and I think is turning a corner in a very positive way, but has definitely been a hot topic lately and over the last few years is Python 3 and the migration towards Python 3. 00:38:11.400 --> 00:38:12.220 Ha, yes. 00:38:12.220 --> 00:38:23.220 I love your attitude when I see you speak at conferences and do keynotes and there's a great 2.8 sign with like a slash through it just saying, look, we're moving forward. 00:38:23.220 --> 00:38:24.900 We're not going back, you guys. 00:38:24.900 --> 00:38:27.080 How do you feel we're doing on that? 00:38:27.080 --> 00:38:35.300 That certainly has been a much harder, more arduous journey than anybody had really anticipated. 00:38:35.300 --> 00:38:38.720 I think right now we're well over the hump. 00:38:38.720 --> 00:38:40.320 The destination is in sight. 00:38:40.620 --> 00:38:49.860 It's clearly better on the other side, but the mountain range was much bigger and colder, whatever, scarier than we had anticipated. 00:38:49.860 --> 00:39:01.820 And I think that honestly, the mistake that all of us in the Python core and actually the whole Python community, the mistake we made was underestimating Python's popularity. 00:39:02.440 --> 00:39:25.340 We sort of thought of Python as this sort of relatively small language with a relatively small number of dedicated followers who would all jump at the occasion of making their code more readable and converting to this new version of the language that was so clearly better in many ways. 00:39:25.680 --> 00:39:34.920 And what we just underestimated how much code people had written in Python 2 and really old versions of Python 2 and that. 00:39:34.920 --> 00:39:54.060 How much documentation there was that would have to be updated, not just the standard library and the core Python documentation and the reference manual, but all the hundreds of Python books and websites and answer sites and stack overflow questions and this and that. 00:39:54.160 --> 00:40:05.620 And we just initially, when we first started talking about Python 3 and actually the first time that the term came up was around the year 2000. 00:40:05.860 --> 00:40:09.740 It then took seven or eight years before we did anything about that. 00:40:09.740 --> 00:40:22.220 But initially, from our perspective and certainly from my perspective, but that everyone who participated in Python 3 had sort of had the same experience. 00:40:22.220 --> 00:40:26.000 Everybody was excited about this change. 00:40:26.000 --> 00:40:28.060 We're going to fix the language. 00:40:28.060 --> 00:40:29.480 There are all these Python words. 00:40:29.480 --> 00:40:36.440 Andrew Cushling, who was an important early core developer, had very influential posts about Python words. 00:40:36.440 --> 00:40:45.480 And he enumerated maybe a dozen of sort of key issues with the language that we were hoping we could somehow improve. 00:40:45.480 --> 00:40:49.440 And one example was in 2000, we introduced Unicode. 00:40:49.440 --> 00:40:53.740 And we introduced it in a way that would be incredibly backwards compatible. 00:40:53.740 --> 00:41:10.460 And by the time, three or four years later, when that design had completely been settled, we started finding out that applications actually became more brittle because they didn't expect the Unicode to pop up in places where it happened. 00:41:10.460 --> 00:41:15.120 And that was one of the things that we wanted to fix in Python 3. 00:41:15.640 --> 00:41:29.820 But we had no idea that there were people with millions of lines of Python code that was all interrelated and written by people who were no longer with that team or that company or that project. 00:41:29.820 --> 00:41:32.720 And what it would take. 00:41:32.820 --> 00:41:52.400 I mean, had we known we could have taken a different tack, we could have made Python 3 somehow, we could have changed a few features in Python 3 to allow somehow Python 2 and Python 3 code to coexist in the same virtual machine. 00:41:52.400 --> 00:41:52.820 Sure. 00:41:52.820 --> 00:41:56.760 Somehow make it just seriously deprecated, but not gone. 00:41:56.760 --> 00:41:57.600 Things like that. 00:41:57.600 --> 00:41:58.440 Some of the features, right? 00:41:58.580 --> 00:41:58.800 Yeah. 00:41:58.800 --> 00:42:03.960 And had we really wanted that, we could have done that. 00:42:03.960 --> 00:42:20.640 But we underestimated the difficulty it would be for the average Python user to convert their code because we thought, well, someone has a few scripts here and a few scripts there and maybe a thousand line library that they're using. 00:42:20.640 --> 00:42:24.000 But in fact, all the numbers were 10 or 100 times larger. 00:42:24.000 --> 00:42:25.760 Making it so much harder to switch. 00:42:25.760 --> 00:42:26.580 Right. 00:42:26.880 --> 00:42:27.160 Yeah. 00:42:27.160 --> 00:42:27.200 Yeah. 00:42:27.200 --> 00:42:34.500 And once we realized that this was not an ideal situation, there wasn't any chance of sort of backing out. 00:42:34.500 --> 00:42:39.340 And there was also no chance of suddenly accelerating the conversion process. 00:42:39.340 --> 00:42:54.880 So we've done the best we felt we could do, which included actually backporting many things to Python 2.7 in terms of libraries that were available and backporting even more things on PyPI. 00:42:54.880 --> 00:43:01.180 Like you can use the enum 3.4 package, I think, in Python 2.7. 00:43:01.180 --> 00:43:03.340 Then you can use enums in Python 2. 00:43:03.340 --> 00:43:12.900 But we also wanted to give the community and the users a sort of a clear message about the future of Python. 00:43:12.900 --> 00:43:16.820 And that's where the sort of no 2.8 banner came from. 00:43:17.180 --> 00:43:22.440 You can always say, well, maybe we should have done this differently, but I'm not sure that it would have necessarily been different. 00:43:22.440 --> 00:43:32.600 You could have said, well, we're going to leave our leave basically some 2.7 or Python 2 in Python 3 and just put new features and clean up around it. 00:43:32.640 --> 00:43:37.000 But then you still have all this old code that still is written in the 2.7 style. 00:43:37.000 --> 00:43:41.780 And the fact that people adopt the three features, you know, maybe they wouldn't. 00:43:41.780 --> 00:43:48.400 And it would have been more of a hindrance rather than just going, all right, we're just going to have to make this break and just jump the gap to get there. 00:43:48.400 --> 00:43:49.640 It's a complex problem. 00:43:49.640 --> 00:43:51.100 There's no perfect solution. 00:43:51.100 --> 00:43:55.260 And you also can't just stop evolving the language. 00:43:55.260 --> 00:43:55.940 Absolutely. 00:43:55.940 --> 00:44:00.020 I feel like Python 3 is going so fast and doing so much. 00:44:00.020 --> 00:44:01.760 It's really positive. 00:44:01.760 --> 00:44:02.340 Thank you. 00:44:02.340 --> 00:44:03.120 Yeah, you're welcome. 00:44:03.120 --> 00:44:16.640 A term that I started using, I got this from Matthias, I'm forgetting his last name, I'm sorry, from some of the Jupyter projects, is referring to Python 3 as Python and Python 2 as legacy Python. 00:44:16.640 --> 00:44:20.080 And I think that's an interesting way to think about it. 00:44:20.080 --> 00:44:29.380 You can try to change the language in the hope that people will subliminally be sort of influenced. 00:44:29.380 --> 00:44:31.580 I don't know how effective that is. 00:44:31.580 --> 00:44:42.580 I think the highway administration or whatever it's called have attempted to remove the word accident from our language and replace it with crash. 00:44:43.160 --> 00:44:46.000 That's wishful thinking in my view. 00:44:46.000 --> 00:44:53.220 And I think it's pretty consistently just say Python 2 and Python 3 whenever the distinction is important. 00:44:53.220 --> 00:44:58.320 You know, to round out this part of the discussion, I definitely feel like we've crossed the boundary. 00:44:58.320 --> 00:44:58.860 Absolutely. 00:44:59.140 --> 00:44:59.780 It's gained momentum. 00:44:59.780 --> 00:45:04.320 I see more and more projects that say either we're Python 3 first or Python 3 only. 00:45:04.320 --> 00:45:06.500 And if people want to backport it, that's fine. 00:45:06.500 --> 00:45:08.120 And this is the right path. 00:45:08.340 --> 00:45:18.480 The really good news is that basically all important libraries work as well with Python 3 as they do with Python 2 or better. 00:45:18.480 --> 00:45:27.900 Which means that the early problem with Python 3 adoption was that, well, if you had a thousand lines of pure Python code, that was very easy to port. 00:45:28.160 --> 00:45:36.700 But if you had a thousand lines of Python 2 that depended on seven different packages, you had to wait for those packages to be ported. 00:45:36.700 --> 00:45:38.900 That has taken a long time. 00:45:38.900 --> 00:45:43.740 And that was, I think, one of the issues that we underestimated. 00:45:43.860 --> 00:45:47.060 But that problem has now pretty much been solved completely. 00:45:47.060 --> 00:45:52.220 NumPy, Django, Flask, everything you can think of works with Python 3. 00:45:52.220 --> 00:45:53.940 Even DateUtil. 00:45:53.940 --> 00:46:01.260 If you want to start writing application code from scratch, there's nothing to stop you from using Python 3. 00:46:01.260 --> 00:46:02.280 Yeah, and that's really great. 00:46:02.280 --> 00:46:05.180 I do feel like that was maybe the single biggest barrier. 00:46:05.180 --> 00:46:17.660 Because even if people had the intention of switching to Python 3, but they depended upon these libraries and they just couldn't run, well, they're like, well, you're going to throw their hands up and say, well, I can't convert all these libraries. 00:46:17.660 --> 00:46:18.420 I got work to do. 00:46:18.420 --> 00:46:18.720 Yeah. 00:46:18.720 --> 00:46:19.080 Yeah. 00:46:19.080 --> 00:46:22.380 So let's do one more Python 3 topic here. 00:46:22.380 --> 00:46:25.580 Let's touch on a little bit of your favorite Python 3 features. 00:46:25.580 --> 00:46:35.020 The two that I've been, I think, most directly involved in were AsyncIO until two years ago. 00:46:35.020 --> 00:46:42.520 Since then, AsyncIO has matured to the point where I don't have to be involved all that much directly. 00:46:42.520 --> 00:46:51.400 Although I carefully encouraged and reviewed the development towards Async and the wait, which has been absolutely marvelous. 00:46:51.840 --> 00:47:05.160 The other favorite of mine is function annotations and what we now have in Python 3.6, the variable annotations, PEP 5.26, and the whole type checking area. 00:47:05.160 --> 00:47:08.900 That's sort of, I think those are my favorite features. 00:47:08.900 --> 00:47:22.340 One other thing that I'm personally very happy with is the proper distinction between bytes and text in Python 3, as opposed to the messy way of dealing with that in Python 2. 00:47:22.340 --> 00:47:22.760 Sure. 00:47:22.760 --> 00:47:26.100 That absolutely solves that whole, it's Unicode, it's not Unicode. 00:47:26.100 --> 00:47:26.640 What is it? 00:47:26.640 --> 00:47:26.920 Yeah. 00:47:26.920 --> 00:47:27.260 Yeah. 00:47:27.260 --> 00:47:31.060 Unfortunately, that has also been a major porting barrier. 00:47:31.060 --> 00:47:31.460 Yeah. 00:47:31.460 --> 00:47:35.760 I've heard that several times, especially people working on web frameworks or things that touch the network. 00:47:35.760 --> 00:47:50.580 The whole generation of Python 2 programmers grew up basically between 2000 and 2010 or so, who knew all the ins and outs of the compatibility and incompatibility between Unicode and bytes. 00:47:50.580 --> 00:48:08.020 And sometimes it interoperates with Unicode and sometimes it doesn't. 00:48:08.020 --> 00:48:17.200 But people got used to the ambiguity and actually exploited it in their APIs, which made their APIs difficult to port forward to Python 3. 00:48:17.200 --> 00:48:31.780 This portion of Talk Python is brought to you by Hired. 00:48:31.780 --> 00:48:34.820 Hired is the platform for top Python developer jobs. 00:48:34.820 --> 00:48:39.620 Create your profile and instantly get access to 3,500 companies who will work to compete with you. 00:48:39.940 --> 00:48:42.480 Take it from one of Hired's users who recently got a job and said, 00:48:42.480 --> 00:48:47.780 I had my first offer on Thursday after going live on Monday and I ended up getting eight offers in total. 00:48:47.780 --> 00:48:51.240 I've worked with recruiters in the past, but they've always been pretty hit and miss. 00:48:51.240 --> 00:48:54.080 I tried LinkedIn, but I found Hired to be the best. 00:48:54.080 --> 00:48:56.180 I really like knowing the salary up front. 00:48:56.180 --> 00:48:58.540 Privacy was also a huge seller for me. 00:48:58.540 --> 00:49:00.220 Sounds awesome, doesn't it? 00:49:00.220 --> 00:49:02.220 Well, wait until you hear about the sign-in bonus. 00:49:02.680 --> 00:49:05.660 Everyone who accepts a job from Hired gets $1,000 signing bonus. 00:49:05.660 --> 00:49:08.300 And as Talk Python listeners, it gets way sweeter. 00:49:08.300 --> 00:49:13.560 Use the link Hired.com slash Talk Python To Me and Hired will double the signing bonus to $2,000. 00:49:13.560 --> 00:49:15.340 Opportunity's knocking. 00:49:15.340 --> 00:49:19.100 Visit Hired.com slash Talk Python To Me and answer the door. 00:49:25.560 --> 00:49:31.220 So does it surprise you that new things are still happening in the CPython internals? 00:49:31.220 --> 00:49:35.960 Like dictionary got a major reworking in 3.6, things like that? 00:49:35.960 --> 00:49:37.600 It doesn't really surprise me. 00:49:37.600 --> 00:49:43.460 Dictionaries in particular are such an incredibly fundamental part of Python. 00:49:43.460 --> 00:49:49.440 It's used everywhere internally and it's used everywhere in applications. 00:49:50.440 --> 00:49:56.900 That it's totally par for the course that every, at least once every decade, maybe twice, 00:49:56.900 --> 00:49:59.880 some major innovation happens in that area. 00:49:59.880 --> 00:50:06.180 I remember when I first started Python, there was like, well, dictionaries were obviously something 00:50:06.180 --> 00:50:06.720 I needed. 00:50:06.720 --> 00:50:13.040 And in reaction to ABC, which used B trees, or at least some kind of, some form of balance 00:50:13.040 --> 00:50:17.500 trees, which I thought was very tedious and I had worked on that code forever in ABC and 00:50:17.500 --> 00:50:18.480 it was always buggy. 00:50:18.480 --> 00:50:21.980 So I thought, well, for Python, I'll do a hash table. 00:50:21.980 --> 00:50:22.980 That's a new thing. 00:50:22.980 --> 00:50:26.040 And I just sort of looked it up in Knuth volume three. 00:50:26.040 --> 00:50:28.200 What's the basic hashing algorithm? 00:50:28.200 --> 00:50:29.580 How do you hash a string? 00:50:29.580 --> 00:50:31.640 What, how do you implement a hash table? 00:50:31.640 --> 00:50:34.500 Well, open hashing or linked lists. 00:50:34.500 --> 00:50:38.460 And I made my decisions and implemented it. 00:50:38.460 --> 00:50:42.140 And then I had to move on to other things, module objects, functions. 00:50:42.140 --> 00:50:44.260 You'd write the whole thing, basically. 00:50:44.260 --> 00:50:47.940 I had to build a whole language, long integers, exceptions. 00:50:48.120 --> 00:50:49.000 Bytecode. 00:50:49.000 --> 00:50:56.420 So the first innovation, I think, happened when someone with an actual mathematical schooling 00:50:56.420 --> 00:51:01.920 in that area realized that I had copied an old algorithm from Knuth that was no longer 00:51:01.920 --> 00:51:02.860 state of the art. 00:51:02.860 --> 00:51:08.220 And there was something like, I think I, Knuth had discovered that it was good if you had 00:51:08.220 --> 00:51:12.180 hash table sizes that were primes or something, relative primes. 00:51:12.960 --> 00:51:19.780 And it turned out with changes in processor architecture, powers of two were suddenly better. 00:51:19.780 --> 00:51:20.340 Interesting. 00:51:20.340 --> 00:51:20.640 Yeah. 00:51:20.640 --> 00:51:22.060 I've always heard primes as well. 00:51:22.060 --> 00:51:23.200 So that's still in my mind. 00:51:23.200 --> 00:51:23.420 Yeah. 00:51:23.420 --> 00:51:25.980 That's no longer the state of the art. 00:51:26.120 --> 00:51:34.420 I mean, the caches in CPUs and the whole sort of L1, L2, L3 memory architecture has affected 00:51:34.420 --> 00:51:37.700 language implementations dramatically. 00:51:37.700 --> 00:51:41.400 And I don't even know all the ins and outs of that area. 00:51:41.560 --> 00:51:45.540 Unfortunately, we have other core developers who keep up with that and do that for me. 00:51:45.540 --> 00:51:46.260 Yeah, that's great. 00:51:46.260 --> 00:51:46.700 That's great. 00:51:46.700 --> 00:51:48.680 I'm sure we could dig into all sorts of those things. 00:51:48.680 --> 00:51:53.000 And there's so many other questions that I'd like to ask you. 00:51:53.000 --> 00:51:55.540 I want to be respectful of your time and not go too long. 00:51:55.620 --> 00:51:58.020 So let me ask you one final Python question. 00:51:58.020 --> 00:52:00.460 What do you see coming in Python 3.7? 00:52:00.460 --> 00:52:02.420 And what would you like to see there? 00:52:02.420 --> 00:52:07.000 Well, I mentioned earlier that I'm not the greatest visionary. 00:52:07.000 --> 00:52:10.900 Also, 3.6 only came out a month ago. 00:52:10.900 --> 00:52:17.080 Unlike some other languages, there's no secret cabal of people who are already planning what 00:52:17.080 --> 00:52:19.280 the language looks like five years from now. 00:52:19.980 --> 00:52:26.120 I'm sure somewhere there's a C++ committee that's anxiously designing C++ 19 or 20 or 00:52:26.120 --> 00:52:28.060 whatever the next version is going to be. 00:52:28.060 --> 00:52:30.740 Well, that's not how we do things in Python. 00:52:30.740 --> 00:52:39.520 And we just over time during the alpha stage of the next feature version, we tend to just 00:52:39.520 --> 00:52:47.700 collect ideas and peps and proposals and often a real life experience with the previous major 00:52:47.700 --> 00:52:52.980 version or the previous feature version, I should say, directs an evolution. 00:52:52.980 --> 00:53:01.700 For example, we added async.io in Python 3.4, which led Yuri Selvanov to come up with the 00:53:01.700 --> 00:53:06.940 idea of async, def and await and introduce that in 3.5. 00:53:06.940 --> 00:53:09.320 That's great because that really cleans up that API. 00:53:09.320 --> 00:53:11.860 That is an incredible improvement. 00:53:11.860 --> 00:53:18.980 And that is not something we could ever have done if we hadn't had the clunky version of 00:53:18.980 --> 00:53:23.420 async.io with yield from in Python 3.4. 00:53:23.420 --> 00:53:25.860 Because the generators actually are... 00:53:25.860 --> 00:53:27.320 I love the story of generators. 00:53:27.320 --> 00:53:34.240 I can talk for hours about that because it has been such a rich source of language improvements 00:53:34.240 --> 00:53:44.560 from the very early for loop to iterators and generators and coroutines and yield from and 00:53:44.560 --> 00:53:50.160 then async await and sort of small improvements to that in 3.6 even. 00:53:50.160 --> 00:53:53.260 That's been a very gratifying thing. 00:53:53.260 --> 00:53:55.500 But I don't know what's going to happen there next. 00:53:55.500 --> 00:53:56.600 Yeah, sure. 00:53:56.720 --> 00:54:02.860 I expect that static typing, optional static typing, what we're doing with mypy, that's 00:54:02.860 --> 00:54:04.740 my current project at Dropbox actually. 00:54:04.740 --> 00:54:09.740 I expect that will also make strides forward. 00:54:09.740 --> 00:54:13.780 But I don't know what those strides are going to be yet. 00:54:13.780 --> 00:54:18.620 Yeah, so it really highlights that the language is a journey and a living thing, right? 00:54:18.620 --> 00:54:21.300 You take a step and you look around. 00:54:21.860 --> 00:54:28.160 You see the landscape that you just ventured into from a different vantage point and you 00:54:28.160 --> 00:54:33.840 can pick a different destination or you can plan your next step or day trip or whatever 00:54:33.840 --> 00:54:36.280 metaphor you want to use. 00:54:36.280 --> 00:54:43.660 You can plan your next evolution based on where you are and what you see. 00:54:44.000 --> 00:54:51.040 And sometimes that just comes like, whoa, we see all these people using this feature that 00:54:51.040 --> 00:54:55.980 we just introduced two versions ago in a completely novel and interesting way. 00:54:55.980 --> 00:55:02.820 And now we suddenly realize, oh, there is a better syntax or a better API that's sort of 00:55:02.820 --> 00:55:04.260 waiting to come out. 00:55:04.260 --> 00:55:05.680 And that's very exciting. 00:55:05.880 --> 00:55:12.140 But I don't have sort of very specific plans for 3.7 yet, let alone beyond. 00:55:12.140 --> 00:55:19.700 The only thing that's sort of possibly way beyond would be the gillectomy, the removal of the 00:55:19.700 --> 00:55:20.020 gill. 00:55:20.020 --> 00:55:23.900 But that is also not at all certain. 00:55:23.900 --> 00:55:26.440 And it's a very complicated story. 00:55:26.440 --> 00:55:28.960 You could talk about the gillectomy for quite a while. 00:55:28.960 --> 00:55:32.780 And it's definitely an interesting area of focus. 00:55:33.480 --> 00:55:38.300 So does this sort of, you take one step, you climb one mountain, you see a new horizon of 00:55:38.300 --> 00:55:38.740 possibilities. 00:55:38.740 --> 00:55:44.060 Is that one of the things that keeps you interested in working on Python and maintaining and overseeing 00:55:44.060 --> 00:55:44.920 the language over time? 00:55:44.920 --> 00:55:50.100 The excitement of actually climbing a mountain and then seeing a whole new valley that you 00:55:50.100 --> 00:55:53.140 didn't know existed certainly motivates me. 00:55:53.140 --> 00:55:54.260 Yeah, that's excellent. 00:55:54.260 --> 00:55:56.000 So I guess we'll leave it there. 00:55:56.000 --> 00:55:59.220 Let me ask you two questions before we get out of here. 00:55:59.220 --> 00:56:03.000 First, if you're going to write some Python code, what editor do you open? 00:56:03.380 --> 00:56:03.700 Emacs. 00:56:03.700 --> 00:56:05.860 Well, actually, I don't open it. 00:56:05.860 --> 00:56:06.680 It's already open. 00:56:06.680 --> 00:56:08.120 It just stays open. 00:56:08.120 --> 00:56:10.980 Well, my shell also mostly runs in Emacs. 00:56:10.980 --> 00:56:11.440 Yeah. 00:56:11.440 --> 00:56:12.120 Oh, fantastic. 00:56:12.120 --> 00:56:12.560 Okay. 00:56:12.560 --> 00:56:17.880 I know you don't like to play favorites, but if there's a notable PyPI package that you really 00:56:17.880 --> 00:56:20.860 want to highlight that maybe people don't know about, but you'd like to say, hey, you guys 00:56:20.860 --> 00:56:23.620 really should check this out because it's a cool example of something. 00:56:23.620 --> 00:56:24.780 What comes to mind? 00:56:25.040 --> 00:56:31.400 Actually, the one thing I want to highlight here is mypy, which is a static type checker 00:56:31.400 --> 00:56:37.620 for Python that I didn't write, although I have now been contributing to it for well over 00:56:37.620 --> 00:56:37.980 a year. 00:56:38.420 --> 00:56:41.840 But it's still mostly Yuka Lettosolo's creation. 00:56:41.840 --> 00:56:44.580 And we're using this at Dropbox. 00:56:44.580 --> 00:56:51.760 And we have over 400,000 lines of annotated code in a code base that totals over 4 million 00:56:51.760 --> 00:56:52.480 lines of code. 00:56:52.600 --> 00:56:59.120 So we have a ways to go, but we also have more than 500 developers who are interacting 00:56:59.120 --> 00:57:08.120 with this tool and who, by and large, are very happy with how adding type annotations makes 00:57:08.120 --> 00:57:13.960 the code more understandable and more readable and sort of makes them more confident when they 00:57:13.960 --> 00:57:16.360 want to undertake refactorings and things like that. 00:57:17.160 --> 00:57:24.980 And mypy, the specific reason I want to highlight mypy as a PyPI package is that until a week 00:57:24.980 --> 00:57:31.980 ago, if you tried pip install mypy, you would get some completely unrelated package that is 00:57:31.980 --> 00:57:35.460 named mypy that has not been maintained for five years. 00:57:35.460 --> 00:57:41.540 And we finally convinced the leadership of PyPI to give us that project name. 00:57:41.540 --> 00:57:44.040 And now you can say pip install mypy. 00:57:44.040 --> 00:57:46.300 Actually, you have to do pip 3 install mypy. 00:57:46.560 --> 00:57:48.980 And it will actually do what you expect it to do. 00:57:48.980 --> 00:57:50.900 Okay, well, that sounds really excellent. 00:57:50.900 --> 00:57:55.560 It's great to see you guys using the type annotations and things like that at Dropbox. 00:57:55.560 --> 00:58:00.600 Guido, we have a final chance for you to give a call to action to the community. 00:58:00.600 --> 00:58:02.320 What would you like people to do? 00:58:02.320 --> 00:58:03.440 What's on your mind? 00:58:03.440 --> 00:58:07.880 Well, I really do want people to give mypy a try. 00:58:07.880 --> 00:58:10.200 That static type checker is pretty amazing. 00:58:10.200 --> 00:58:16.300 And I need to point out that there are still a lot of misunderstandings about what it can do 00:58:16.300 --> 00:58:17.420 and what it cannot do. 00:58:17.420 --> 00:58:23.860 Static types are completely orthogonal or almost completely orthogonal to unit testing, for example. 00:58:23.860 --> 00:58:26.840 It doesn't mean you have to write fewer tests. 00:58:26.840 --> 00:58:31.420 Occasionally, there are a few trivial tests you don't need to write because the static checker 00:58:31.420 --> 00:58:34.340 catches the same issues better. 00:58:34.340 --> 00:58:39.660 But by and large, a type checker just catches a very different category of errors than unit tests. 00:58:39.740 --> 00:58:41.620 So they complement each other nicely. 00:58:41.620 --> 00:58:44.100 And together, they give you more confidence. 00:58:44.100 --> 00:58:50.340 One of the misunderstandings about mypy has also been that when we started PEP 484, that 00:58:50.340 --> 00:58:53.400 the sort of the introduction of type checking was Python 3 only. 00:58:53.580 --> 00:58:55.480 But we've actually amended that. 00:58:55.480 --> 00:59:00.580 And for the past year, we've been using mypy successfully with Python 2 code base. 00:59:00.580 --> 00:59:01.120 Okay. 00:59:01.120 --> 00:59:07.040 So you can search PEP 484 for Python 2 support. 00:59:07.040 --> 00:59:09.440 And it's completely there. 00:59:09.440 --> 00:59:13.060 And mypy does as well on Python 2 as it does on Python 3. 00:59:13.060 --> 00:59:14.140 Oh, that's really nice. 00:59:14.140 --> 00:59:20.200 Yeah, the other misunderstanding about mypy or static typing in Python is that people have 00:59:20.200 --> 00:59:24.760 predicted that they will turn your Python code into Java, which is obviously nonsense. 00:59:24.760 --> 00:59:28.460 There's no interfaces all over the place. 00:59:28.460 --> 00:59:29.000 Yeah. 00:59:29.000 --> 00:59:35.400 So I think actually moving that type annotations to Python 2 is really important because it provides 00:59:35.400 --> 00:59:38.460 some foundation for when you do want to upgrade to Python 3. 00:59:38.460 --> 00:59:39.020 Correct. 00:59:39.020 --> 00:59:44.480 My secret plan at Dropbox is actually that once we have a large enough fraction of the 00:59:44.480 --> 00:59:51.120 code base annotated, we can start converting into Python 3 in a semi-automated fashion in 00:59:51.120 --> 00:59:55.040 a way that would not be possible without those annotations. 00:59:55.040 --> 00:59:58.220 We're not there yet, but that's my secret plan. 00:59:58.220 --> 00:59:59.220 That's fantastic. 00:59:59.220 --> 00:59:59.980 All right. 00:59:59.980 --> 01:00:02.380 Well, I want to say thank you again for the conversation. 01:00:02.380 --> 01:00:07.260 I really enjoyed talking with you and I'm sure everyone out there learned a lot. 01:00:07.260 --> 01:00:08.760 So thanks so much for being on the show. 01:00:08.760 --> 01:00:09.520 My pleasure. 01:00:09.520 --> 01:00:10.040 Yeah. 01:00:10.040 --> 01:00:10.780 Hope it's a good one. 01:00:10.780 --> 01:00:11.660 It's going to be a great one. 01:00:11.660 --> 01:00:12.000 Bye. 01:00:12.000 --> 01:00:12.440 Bye. 01:00:12.440 --> 01:00:16.460 This has been another episode of Talk Python To Me. 01:00:16.460 --> 01:00:21.020 Today's guest has been Guido Van Rossum and this episode has been sponsored by Rollbar 01:00:21.020 --> 01:00:21.700 and Hired. 01:00:21.700 --> 01:00:23.640 Thank you both for sponsoring the show. 01:00:23.640 --> 01:00:27.060 I want to say thank you again to Guido as well. 01:00:27.060 --> 01:00:29.100 It was really an honor to have him on the show. 01:00:29.100 --> 01:00:32.420 I know he's a very busy guy and there are tons of demands on his time. 01:00:32.420 --> 01:00:34.400 So thank you, Guido, for being on the show. 01:00:34.400 --> 01:00:37.260 I know everyone really enjoyed hearing your perspective. 01:00:38.220 --> 01:00:40.280 Rollbar takes the pain out of errors. 01:00:40.280 --> 01:00:45.040 They give you the context and insight you need to quickly locate and fix errors that 01:00:45.040 --> 01:00:47.980 might have gone unnoticed until your users complain, of course. 01:00:47.980 --> 01:00:52.800 As Talk Python To Me listeners, track a ridiculous number of errors for free at 01:00:52.800 --> 01:00:55.140 rollbar.com slash Talk Python To Me. 01:00:55.140 --> 01:00:58.380 Hired wants to help you find your next big thing. 01:00:58.380 --> 01:01:03.120 Visit Hired.com slash Talk Python To Me to get five or more offers with salary and equity 01:01:03.120 --> 01:01:06.880 presented right up front and a special listener signing bonus of $2,000. 01:01:06.880 --> 01:01:10.020 Are you or a colleague trying to learn Python? 01:01:10.020 --> 01:01:14.720 Have you tried books and videos that just left you bored by covering topics point by point? 01:01:14.720 --> 01:01:20.720 Well, check out my online course, Python Jumpstart, by building 10 apps at talkpython.fm/course 01:01:20.720 --> 01:01:23.320 to experience a more engaging way to learn Python. 01:01:23.940 --> 01:01:28.120 And if you're looking for something a little more advanced, try my WritePythonic code course 01:01:28.120 --> 01:01:30.660 at talkpython.fm/Pythonic. 01:01:30.660 --> 01:01:33.140 Be sure to subscribe to the show. 01:01:33.140 --> 01:01:35.340 Open your favorite podcatcher and search for Python. 01:01:35.340 --> 01:01:36.580 We should be right at the top. 01:01:36.960 --> 01:01:42.680 You can also find the iTunes feed at /itunes, Google Play feed at /play, and direct 01:01:42.680 --> 01:01:45.880 RSS feed at /rss on talkpython.fm. 01:01:45.880 --> 01:01:50.980 Our theme music is Developers, Developers, Developers by Corey Smith, who goes by Smix. 01:01:50.980 --> 01:01:55.680 Corey just recently started selling his tracks on iTunes, so I recommend you check it out at 01:01:55.680 --> 01:01:57.660 Talk Python.fm slash music. 01:01:57.660 --> 01:02:02.100 You can browse his tracks he has for sale on iTunes and listen to the full-length version 01:02:02.100 --> 01:02:03.020 of the theme song. 01:02:03.020 --> 01:02:05.100 This is your host, Michael Kennedy. 01:02:05.500 --> 01:02:06.380 Thanks so much for listening. 01:02:06.380 --> 01:02:07.580 I really appreciate it. 01:02:07.580 --> 01:02:09.720 Smix, let's get out of here. 01:02:09.720 --> 01:02:11.720 Thanks for listening. 01:02:11.720 --> 01:02:16.840 my voice, there's no norm that I can feel within, haven't been sleeping, I've been using lots of rest, 01:02:16.840 --> 01:02:19.660 I'll pass the mic back to who rocked it best. 01:02:19.660 --> 01:02:26.660 I'll see you next time. 01:02:26.660 --> 01:02:31.980 Developers, developers, developers, developers, developers, developers, developers, developers, developers, developers. 01:02:31.980 --> 01:02:32.000 Thank you. 01:02:32.000 --> 01:03:01.980 Thank you.