# 50 Years of Microsoft and Developer Tools with Scott Guthrie

An episode of [The Pragmatic Engineer](https://invisiblenetwork.ai/podcast/1769051199), 4 Jun 2025, 1 h 04 min.

Audio: https://api.substack.com/feed/podcast/165051944/c04887ff8c22dbd64d76c77e228c61e6.mp3

Supported by Our Partners • Statsig — The unified platform for flags, analytics, experiments, and more. • Sinch — Connect with customers at every step of their journey. • Modal — The cloud platform for building AI applications. — How has Microsoft changed since its founding in 1975, especially in how it builds tools for developers? In this episode of The Pragmatic Engineer, I sit down with Scott Guthrie, Executive Vice President of Cloud and AI at Microsoft. Scott has been with the company for 28 years. He built the first prototype of ASP.NET, led the Windows Phone team, led up Azure, and helped shape many of Microsoft’s most important developer platforms. We talk about Microsoft’s journey from building early dev tools to becoming a top cloud provider—and how it actively worked to win back and grow its developer base. In this episode, we cover: • Microsoft’s early years building developer tools • Why Visual Basic faced resistance from devs back in the day: even though it simplified development at the time • How .NET helped bring a new generation of server-side developers into Microsoft’s ecosystem • Why Windows Phone didn’t succeed • The 90s Microsoft dev stack: docs, debuggers, and more • How Microsoft Azure went from being the #7 cloud provider to the #2 spot today • Why Microsoft created VS Code • How VS Code and open source led to the acquisition of GitHub • What Scott’s excited about in the future of developer tools and AI • And much more! — Timestamps (00:00) Intro (02:25) Microsoft’s early years building developer tools (06:15) How Microsoft’s developer tools helped Windows succeed (08:00) Microsoft’s first tools were built to allow less technically savvy people to build things (11:00) A case for embracing the technology that’s coming (14:11) Why Microsoft built Visual Studio and .NET (19:54) Steve Ballmer’s speech about .NET (22:04) The origins of C# and Anders Hejlsberg’s impact on Microsoft (25:29) The 90’s Microsoft stack, including documentation, debuggers, and more (30:17) How productivity has changed over the past 10 years (32:50) Why Gergely was a fan of Windows Phone—and Scott’s thoughts on why it didn’t last (36:43) Lessons from working on (and fixing) Azure under Satya Nadella (42:50) Codeplex and the acquisition of GitHub (48:52) 2014: Three bold projects to win the hearts of developers (55:40) What Scott’s excited about in new developer tools and cloud computing (59:50) Why Scott thinks AI will enhance productivity but create more engineering jobs — The Pragmatic Engineer deepdives relevant for this episode: • Microsoft is dogfooding AI dev tools’ future • Microsoft’s developer tools roots • Why are Cloud Development Environments spiking in popularity, now? • Engineering career paths at Big Tech and scaleups • How Linux is built with Greg Kroah-Hartman — See the transcript and other references from the episode at https://newsletter.pragmaticengineer.com/podcast — Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email podcast@pragmaticengineer.com. Get full access to The Pragmatic Engineer at newsletter.pragmaticengineer.com/subscribe

## Transcript

[0:00] One of the things that we did in the early days of 2014 was
[0:02] kind of just looked
[0:04] SPEAKER_01: around and said,
[0:04] SPEAKER_01: you know,
[0:05] I think we have an opportunity to make a couple of choices that
[0:08] will be bold and
[0:09] aggressive that give us a shot to rewind relevance with
[0:12] developers of the world.
[0:13] And if we don't,
[0:14] we're going to be on an iceberg that's going to slowly melt.
[0:16] SPEAKER_01: And at some point we're going to be swimming.
[0:18] And so we kind of had a set of meetings in spring of 2014 and
[0:21] kind of wrote on the
[0:22] SPEAKER_01: whiteboard,
[0:23] SPEAKER_01: let's be bold and came up with three big things that we said,
[0:26] SPEAKER_01: okay,
[0:26] SPEAKER_01: what can we do to become more relevant?
[0:28] the first one was we actually introduced a community edition of
[0:31] visual studio
[0:32] decision number two is let's open source.net and let's make it
[0:34] cross-platform we
[0:35] want the community to contribute code and do it under the right
[0:38] license and put it
[0:39] on github and then decision number three was as much as we love
[0:43] visual studio the
[0:44] ide let's also recognize increasingly web developers and those
[0:48] that are not using a
[0:48] compiled language are looking for something that's much more of
[0:51] a lightweight code
[0:52] SPEAKER_01: optimized editor
[0:53] SPEAKER_01: And there was a great project that had been started.
[0:56] SPEAKER_01: It was basically a web-based editor written in Node,
[0:59] SPEAKER_01: written in TypeScript,
[1:00] SPEAKER_01: but it ran in a VS Online.
[1:02] SPEAKER_01: It was great technology,
[1:03] SPEAKER_01: but at the time,
[1:04] SPEAKER_01: people weren't really looking to write code in a browser.
[1:07] SPEAKER_01: And the three different decisions we made,
[1:09] SPEAKER_01: we kind of made all those decisions,
[1:10] SPEAKER_01: I think,
[1:11] SPEAKER_01: in about an hour and a half.
[1:12] SPEAKER_01: Of the three decisions, they were all very big.
[1:14] The ones that we remember the most would be VS Code and then
[1:16] the open sourcing of .NET.
[1:18] SPEAKER_00: Microsoft was founded 50 years ago in 1975.
[1:21] So how has the company changed and how it builds developer
[1:23] tools across these five decades?
[1:25] SPEAKER_00: Today,
[1:25] SPEAKER_00: I sat down with Scott Guthrie,
[1:27] SPEAKER_00: who has been with Microsoft for 28 years,
[1:29] SPEAKER_00: created the first prototype version of ASP.NET,
[1:32] SPEAKER_00: and is currently the executive vice president for cloud and AI.
[1:35] SPEAKER_00: In this episode,
[1:36] we talk about two decades of Microsoft and how Windows became a
[1:40] success in part
[1:40] thanks to shipping programming tools like QuickSee or MFC to
[1:43] help devs build
[1:44] SPEAKER_00: programs on top of the OS.
[1:46] SPEAKER_00: how and why Visual Basic,
[1:48] SPEAKER_00: C Sharp,
[1:48] SPEAKER_00: ASP.NET,
[1:49] SPEAKER_00: and Visual Studio were created,
[1:51] the time when developers paid thousands of dollars per year to
[1:54] access quality
[1:55] SPEAKER_00: documentation over MSDN,
[1:57] how Microsoft decided to embrace open source and create VS Code
[2:01] as an open source
[2:01] SPEAKER_00: project,
[2:02] SPEAKER_00: and many more topics,
[2:03] SPEAKER_00: including what Scott is excited about looking ahead.
[2:06] SPEAKER_00: This is a rare conversation with Scott,
[2:08] SPEAKER_00: who has been with Microsoft for over more than half his lifetime,
[2:11] SPEAKER_00: and I hope you enjoy the stories and details that he shares with us.
[2:14] SPEAKER_00: So Scott, welcome to the podcast.
[2:16] SPEAKER_00: It's great to be here.
[2:17] SPEAKER_00: Thanks for having me.
[2:18] SPEAKER_00: It is so nice to meet again,
[2:19] SPEAKER_00: especially that we've met a long time ago,
[2:22] SPEAKER_00: 10 years ago,
[2:23] SPEAKER_00: and then I also worked at Microsoft.
[2:24] SPEAKER_00: But today, I'd like to talk, just go back to way to the beginning.
[2:28] SPEAKER_00: 50 years ago, it's crazy to say that, that's when Microsoft was founded.
[2:32] SPEAKER_00: And when I think about Microsoft, my first memory is Windows.
[2:36] SPEAKER_00: But actually, it started with developer tools, right?
[2:38] SPEAKER_00: Right.
[2:38] SPEAKER_01: Yeah.
[2:39] SPEAKER_01: The very first product that Microsoft built and,
[2:42] SPEAKER_01: and really,
[2:42] SPEAKER_01: you know,
[2:42] SPEAKER_01: not just the first product,
[2:43] SPEAKER_01: but for,
[2:44] SPEAKER_01: for many years,
[2:45] SPEAKER_01: we were just a developer tools company.
[2:47] SPEAKER_01: And so it was Microsoft basic for the Altair and,
[2:50] SPEAKER_01: uh,
[2:51] SPEAKER_01: Bill and Paul Allen built it while Bill was still in Harvard.
[2:55] SPEAKER_01: And,
[2:55] SPEAKER_01: uh,
[2:55] SPEAKER_01: you know,
[2:56] SPEAKER_01: the legend is they flew out to Albuquerque,
[2:59] SPEAKER_01: New Mexico,
[2:59] SPEAKER_01: which is where,
[3:00] SPEAKER_01: um,
[3:02] SPEAKER_01: the,
[3:02] SPEAKER_01: the company that built the Altair was,
[3:04] SPEAKER_01: it's a computer no longer exists,
[3:06] SPEAKER_01: obviously.
[3:07] SPEAKER_01: And, um,
[3:08] SPEAKER_01: And they brought a tape of Microsoft Basic,
[3:12] SPEAKER_01: a paper tape,
[3:14] SPEAKER_01: and they actually plugged it in and loaded it,
[3:17] SPEAKER_01: and it worked.
[3:17] And it was actually,
[3:18] they built the app Basic entirely without access to the
[3:23] computer.
[3:24] SPEAKER_01: And it was kind of the hit product for that
[3:27] that particular computer and then they expanded to other
[3:29] platforms and then
[3:30] SPEAKER_00: eventually operating systems.
[3:32] But the first dollars they made by selling basically a
[3:34] developer tool,
[3:35] a compiler,
[3:35] right?
[3:35] So like you could write basic code and it would compile to
[3:38] machine code,
[3:38] right?
[3:39] SPEAKER_00: That was the product, right?
[3:40] SPEAKER_01: I think it was an interpreter at that time.
[3:41] SPEAKER_00: The interpreter, yeah.
[3:43] But basically,
[3:43] yeah,
[3:44] I mean,
[3:44] it was literally,
[3:45] and back then that was how you kind of,
[3:47] SPEAKER_01: used computers is often you were a programmer and you actually,
[3:50] SPEAKER_01: you know,
[3:50] SPEAKER_01: I guess today would be modern equivalent be scripting.
[3:53] SPEAKER_01: And you, you know, you built your own logic and your own applications.
[3:57] SPEAKER_01: And so,
[3:57] SPEAKER_01: yeah,
[3:57] SPEAKER_01: it's,
[3:58] SPEAKER_01: it's,
[3:58] SPEAKER_01: you know,
[3:58] that's one of the things that's been cool is that the company
[4:00] started as a
[4:01] SPEAKER_01: developer tools company.
[4:02] SPEAKER_01: And when you look at our,
[4:04] SPEAKER_01: what we built today,
[4:05] SPEAKER_01: whether it's VS code,
[4:06] SPEAKER_01: whether it's get out,
[4:07] SPEAKER_01: whether it's Azure,
[4:08] SPEAKER_01: we remain a developer tools company.
[4:09] SPEAKER_00: If you want to build a great product, you have to ship quickly.
[4:13] SPEAKER_00: But how do you know what works?
[4:14] SPEAKER_00: More importantly, how do you avoid shipping things that don't work?
[4:19] SPEAKER_00: The answer, Statsig.
[4:21] SPEAKER_00: Statsig is a unified platform for flags,
[4:23] SPEAKER_00: analytics,
[4:24] SPEAKER_00: experiments,
[4:25] SPEAKER_00: and more,
[4:25] combining five plus products into a single platform with a
[4:28] unified set of data.
[4:30] SPEAKER_00: Here's how it works.
[4:32] SPEAKER_00: First, StatSick helps you ship a feature via FeatureFly or Config.
[4:36] SPEAKER_00: Then it measures how it's working,
[4:38] SPEAKER_00: from alerts and errors,
[4:39] SPEAKER_00: to replays of people using that feature,
[4:42] SPEAKER_00: to measurement of top-line impact.
[4:44] SPEAKER_00: Then you get your analytics,
[4:45] SPEAKER_00: user account metrics,
[4:46] SPEAKER_00: and dashboards to track your progress over time,
[4:49] SPEAKER_00: all linked to the stuff you ship.
[4:51] SPEAKER_00: Even better,
[4:51] SPEAKER_00: Statsik is incredibly affordable with a super generous free tier,
[4:55] SPEAKER_00: a starter program with $50,000 of free credits,
[4:58] SPEAKER_00: and custom plans to help you consolidate your existing spend on flags,
[5:01] SPEAKER_00: analytics,
[5:01] SPEAKER_00: or A-B testing tools.
[5:03] SPEAKER_00: To get started, go to statsik.com slash pragmatic.
[5:07] SPEAKER_00: That is S-T-A-T-S-I-G dot com slash pragmatic.
[5:10] SPEAKER_00: Happy building!
[5:12] SPEAKER_00: This episode is brought to you by Cinch,
[5:14] the customer communications cloud trusted by thousands of
[5:17] engineering teams around
[5:18] SPEAKER_00: the world.
[5:19] If you've ever added messaging,
[5:20] voice,
[5:21] or email into a product,
[5:22] you know the pain.
[5:23] SPEAKER_00: Flaky delivery and platform stack with middlemen.
[5:26] SPEAKER_00: Cinch is different.
[5:28] They run their own network with direct carrier connections in
[5:31] over 60 countries.
[5:32] That means faster delivery,
[5:34] higher reliability,
[5:35] and scale that just works.
[5:37] SPEAKER_00: Developers love Cinch for its single API that covers 12 channels,
[5:41] SPEAKER_00: including SMS,
[5:42] SPEAKER_00: WhatsApp,
[5:42] SPEAKER_00: and RCS.
[5:44] SPEAKER_00: Now is the time to pay attention to RCS, rich communication services.
[5:48] SPEAKER_00: It's like SMS, but smarter.
[5:50] Your brand name,
[5:51] logo,
[5:51] and verified checkmark all inside the native messaging app.
[5:55] SPEAKER_00: Built by Google,
[5:56] SPEAKER_00: now rolling out with Apple and major carriers,
[5:58] SPEAKER_00: RCS is becoming the messaging standard.
[6:01] SPEAKER_00: Cinch is helping teams go live globally.
[6:04] SPEAKER_00: Learn more at cinch.com slash pragmatic.
[6:07] SPEAKER_00: That is S-I-N-C-H.com slash pragmatic.
[6:12] SPEAKER_00: And just stepping forward from basics,
[6:15] SPEAKER_00: this was 1975,
[6:16] SPEAKER_00: but in the 90s or actually like mid 80s,
[6:18] SPEAKER_00: Microsoft started to become big because of Windows.
[6:20] SPEAKER_00: And as I looked into the history,
[6:22] I was pretty surprised to see that as Microsoft developed
[6:25] Windows 1.0 and then 3.0
[6:26] SPEAKER_00: and 3.1 was a big,
[6:27] SPEAKER_00: big breakthrough.
[6:31] SPEAKER_00: They always developed programs to build on top of Windows.
[6:36] SPEAKER_00: So they did Quick Basic,
[6:39] SPEAKER_00: Microsoft C,
[6:40] SPEAKER_00: and Quick C.
[6:41] SPEAKER_00: Do you think,
[6:42] SPEAKER_00: you know,
[6:42] SPEAKER_00: having talked with early Microsoft employees and just knowing this year,
[6:46] SPEAKER_00: was this an intentional strategy?
[6:48] SPEAKER_00: Or might have this thing actually helped Windows become as big as it did?
[6:51] SPEAKER_01: Well, I think it absolutely helped Windows become as big as it did.
[6:54] SPEAKER_01: I mean,
[6:54] SPEAKER_01: at the end of the day,
[6:55] if you're building an operating system or you're building a
[6:58] cloud platform,
[6:59] You know,
[6:59] your success is very much,
[7:01] you know,
[7:01] no one buys a platform by itself.
[7:04] SPEAKER_01: They buy it for the applications that run on top.
[7:06] SPEAKER_01: And so if you don't have developers,
[7:09] SPEAKER_01: you know,
[7:10] SPEAKER_01: building those applications,
[7:11] SPEAKER_01: you don't,
[7:11] SPEAKER_01: you don't have a business.
[7:12] SPEAKER_01: And so I think definitely in the early days with windows,
[7:15] SPEAKER_01: you know,
[7:16] SPEAKER_01: part of the whole thesis was,
[7:17] SPEAKER_01: you know,
[7:18] SPEAKER_01: build great tools,
[7:19] SPEAKER_01: make it easy to build applications that are great,
[7:21] SPEAKER_01: help developers be successful.
[7:24] SPEAKER_01: And if their apps are successful, we'll sell more windows.
[7:27] SPEAKER_01: And it's similar today with Azure in terms of,
[7:29] SPEAKER_01: you know,
[7:30] SPEAKER_01: one of the things I covered in my keynote was ChatGPT,
[7:33] SPEAKER_01: which runs,
[7:34] SPEAKER_01: you know,
[7:34] SPEAKER_01: entirely on Azure.
[7:35] SPEAKER_01: And,
[7:36] SPEAKER_01: you know,
[7:36] SPEAKER_01: it's another great example of,
[7:38] SPEAKER_01: you know,
[7:38] SPEAKER_01: how do we make,
[7:39] SPEAKER_01: in this case,
[7:40] SPEAKER_01: a company OpenAI successful and enable them to build an amazing app.
[7:44] And if they are successful and they're running on our platform,
[7:47] we're successful too.
[7:49] And so it's a model that I think is very much core to the DNA
[7:52] of the company.
[7:53] SPEAKER_00: Yeah,
[7:54] SPEAKER_00: and then in the 90s,
[7:56] SPEAKER_00: as Windows became more popular and there were these tools,
[7:59] SPEAKER_00: Microsoft did something really,
[8:00] SPEAKER_00: really interesting.
[8:01] They built some tools that allowed maybe not as technical
[8:05] developers to build stuff.
[8:07] SPEAKER_00: Visual Basic, Microsoft Access, later Microsoft Front Page.
[8:12] SPEAKER_00: And I'm now having a little deja vu of what we're seeing today,
[8:15] SPEAKER_00: but just reflecting on that time and what you've seen,
[8:18] SPEAKER_00: Visual Basic seems to still have a massive impact.
[8:21] SPEAKER_00: When I worked at a bank at J.P.
[8:23] SPEAKER_00: Morgan,
[8:24] SPEAKER_00: building software that was used by professional traders,
[8:28] SPEAKER_00: some of the traders loved using Visual Basic that they,
[8:31] SPEAKER_00: as non-technical people,
[8:32] SPEAKER_00: could write their own programs.
[8:35] And Visual Basic at the time was absolutely revolutionary
[8:38] because I remember I was
[8:41] SPEAKER_01: in high school and I was a developer on the Mac.
[8:45] SPEAKER_01: I mean, System 7 had just come out.
[8:47] It was like a long time ago on the Mac,
[8:48] you know,
[8:49] pre,
[8:50] you know,
[8:51] OS 10 and everything.
[8:52] SPEAKER_01: And, you know, building a GUI application was a lot of work.
[8:55] You know,
[8:55] if you wanted a menu,
[8:56] if you wanted buttons,
[8:57] if you want a dialog.
[8:59] I mean,
[8:59] you were writing a lot of code and it was,
[9:02] it was often very error prone code.
[9:03] SPEAKER_01: Cause if you got anything wrong, it would crash.
[9:06] SPEAKER_01: And,
[9:06] SPEAKER_01: um,
[9:07] SPEAKER_01: there were no visual designers,
[9:08] SPEAKER_01: you know,
[9:08] SPEAKER_01: it was pretty much,
[9:09] SPEAKER_01: you kind of had to do everything from scratch and,
[9:11] SPEAKER_01: and visual basic came along and it was sort of,
[9:13] SPEAKER_01: you could drag and drop buttons,
[9:15] SPEAKER_01: you know,
[9:16] SPEAKER_01: visually lay them out,
[9:17] SPEAKER_01: double click,
[9:17] SPEAKER_01: write a couple lines of code,
[9:18] SPEAKER_01: hit F5 and it ran.
[9:20] SPEAKER_01: And it really transformed development.
[9:23] And,
[9:23] um,
[9:24] you know,
[9:24] we ultimately took that same paradigm and used it with access
[9:28] for databases.
[9:29] SPEAKER_01: We also had a product, which is one of the ones I used.
[9:31] I think it was the first thing I ever got paid for was a
[9:33] product called Visual Fox
[9:35] SPEAKER_01: Pro,
[9:35] SPEAKER_01: which was another database product that was graphical.
[9:38] Um,
[9:39] and then,
[9:39] uh,
[9:40] and then extended it to things like visual J plus plus.
[9:43] SPEAKER_01: And then,
[9:44] SPEAKER_01: uh,
[9:45] SPEAKER_01: yeah,
[9:45] SPEAKER_01: ultimately the visual studio family,
[9:47] SPEAKER_01: including C sharp C plus plus,
[9:50] SPEAKER_01: uh,
[9:50] SPEAKER_01: and others.
[9:51] SPEAKER_01: And so,
[9:51] SPEAKER_01: you know,
[9:51] SPEAKER_01: it's that,
[9:52] SPEAKER_01: that notion of,
[9:53] SPEAKER_01: you know,
[9:53] SPEAKER_01: again,
[9:54] SPEAKER_01: it was court of like,
[9:55] how do we take the friction out of what developers needed to do
[9:59] and provide,
[10:00] SPEAKER_01: um,
[10:01] great developer tools to do it and as much as you know the drag
[10:04] and drop aspect of
[10:05] virtual basic was revolutionary the other thing that was
[10:07] revolutionary was just the
[10:08] speed with which you could run and uh it had a pioneered a
[10:12] feature called edit and
[10:12] continue so you could even have the application running modify
[10:17] a couple lines of
[10:18] code then push the button again and without having to kind of
[10:20] rebuild or recompile
[10:22] or relaunch you could actually see the edits immediately take
[10:25] effect and that was
[10:26] SPEAKER_01: also kind of transformative like you know today we
[10:30] SPEAKER_01: With JavaScript,
[10:31] SPEAKER_01: we're sort of used to that kind of paradigm of make a change,
[10:34] SPEAKER_01: hit save,
[10:34] SPEAKER_01: and it just works.
[10:35] SPEAKER_01: But in the 90s, that was radical in a big way.
[10:41] SPEAKER_00: Can we just rewind back time?
[10:42] SPEAKER_00: Because I feel there's a bit of a similarity.
[10:45] Back then,
[10:46] in the 90s,
[10:46] before Visual Basic and Access and Foxtrot came about,
[10:50] there was no way for non-super-technical people who knew you
[10:55] learned the syntax to
[10:56] SPEAKER_00: write programs.
[10:57] SPEAKER_00: And these things just allowed,
[10:58] SPEAKER_00: honestly,
[10:59] SPEAKER_00: everyday people,
[10:59] SPEAKER_00: like my dad,
[11:00] SPEAKER_00: who was not a developer,
[11:02] SPEAKER_00: he put together a bunch of stuff with Access,
[11:05] SPEAKER_00: later FoxPro.
[11:06] SPEAKER_00: And today we're in a similar thing,
[11:10] SPEAKER_00: but I'd like to know,
[11:11] SPEAKER_00: do you remember what the mood was like where developers were saying,
[11:14] SPEAKER_00: oh no,
[11:14] SPEAKER_00: this is the end of development,
[11:15] SPEAKER_00: we're not going to have...
[11:18] SPEAKER_00: Was there any of this?
[11:20] SPEAKER_00: And then what happened as a result of this?
[11:23] Because now we have like 20 or 30 years of looking back of how
[11:26] it transformed
[11:27] SPEAKER_00: software engineering and the industry.
[11:28] SPEAKER_01: There probably was some,
[11:30] SPEAKER_01: you know,
[11:30] SPEAKER_01: questioning in terms of like,
[11:32] SPEAKER_01: oh,
[11:32] SPEAKER_01: is this going to,
[11:33] SPEAKER_01: you know,
[11:34] SPEAKER_01: mean anyone can be a developer?
[11:35] SPEAKER_01: I think it's often the refrain,
[11:37] SPEAKER_01: whether it's,
[11:38] SPEAKER_01: you know,
[11:38] SPEAKER_01: that era,
[11:39] SPEAKER_01: whether it's web development,
[11:40] SPEAKER_01: whether it's,
[11:41] SPEAKER_01: you know,
[11:42] SPEAKER_01: co-pilots and AI assistants now.
[11:44] You know,
[11:44] I remember that same pushback when Java came out and .NET came
[11:47] out.
[11:48] It was sort of like,
[11:49] ah,
[11:49] you know,
[11:50] no one,
[11:50] no real developer would ever use garbage collection.
[11:53] SPEAKER_01: Oh yeah.
[11:54] SPEAKER_01: It's really should be, you know, Malik and free.
[11:56] SPEAKER_01: Um,
[11:56] SPEAKER_01: and,
[11:57] SPEAKER_01: uh,
[11:57] SPEAKER_01: you know,
[11:57] I think the history of development and the history of computing
[12:01] is really around
[12:02] SPEAKER_01: how do we continue to optimize the productivity?
[12:05] And I think what makes a developer successful or have impact is
[12:11] not syntax and it's
[12:13] SPEAKER_01: not unnecessary writing of code that can be automated.
[12:17] SPEAKER_01: It's often the logic,
[12:19] SPEAKER_01: it's identifying a problem,
[12:20] SPEAKER_01: it's the creativity,
[12:21] SPEAKER_01: and it's often the problem solving to make it work.
[12:25] SPEAKER_01: And I think developers that embrace the new technology
[12:29] SPEAKER_01: and build on top of it,
[12:31] SPEAKER_01: end up being so much more productive,
[12:34] SPEAKER_01: which ultimately means that they have better careers,
[12:35] SPEAKER_01: they're more impactful.
[12:37] SPEAKER_01: And so I always encourage people to kind of,
[12:40] SPEAKER_01: don't assume your value is the syntax that you're familiar with.
[12:44] SPEAKER_01: Assume the value is a higher level thing that you can leverage.
[12:47] SPEAKER_01: And I think it's true both for developer tools and languages,
[12:52] SPEAKER_01: especially in the context of AI now.
[12:54] SPEAKER_01: But I think it's also true in terms of cloud platforms.
[12:58] SPEAKER_01: I talked to my keynote this morning at Build about
[13:02] SPEAKER_01: ChatGPT.
[13:03] SPEAKER_01: And if ChatGPT had been built 10 years ago,
[13:07] SPEAKER_01: or certainly 20 years ago,
[13:10] SPEAKER_01: a company would need to build their own data centers.
[13:13] SPEAKER_01: They would need to build their own operating system.
[13:15] SPEAKER_01: They would need to build their own deployment orchestrator.
[13:17] SPEAKER_01: They need to build their own database.
[13:18] SPEAKER_00: That's what Google had to do like 20 or 30 years ago.
[13:19] SPEAKER_01: That's exactly what Google had to do.
[13:21] And the fact that they were able to leverage things like Cosmos
[13:24] DB and our Azure
[13:25] SPEAKER_01: Kubernetes service,
[13:27] SPEAKER_01: they now have an app that's the fastest growing in history.
[13:31] SPEAKER_01: with 500 million weekly users.
[13:34] And one of the stats I put in the keynote was they have a total
[13:37] of 12 people on
[13:38] SPEAKER_01: their infrastructure team that manages all their Kubernetes and compute
[13:41] SPEAKER_01: infrastructure.
[13:42] SPEAKER_01: And that type of productivity is stunning.
[13:45] SPEAKER_01: And so I do think,
[13:48] SPEAKER_01: I would encourage if you're a developer,
[13:49] SPEAKER_01: embrace the technology and the productivity that's coming.
[13:53] SPEAKER_01: And it only makes you more successful and have more impact.
[13:57] SPEAKER_01: There's more than enough problems that are still left to solve.
[14:00] After Visual Basic,
[14:02] one of the big years was 1997,
[14:05] I guess for two reasons.
[14:06] SPEAKER_00: Was that the year that I remember that you joined Microsoft?
[14:09] SPEAKER_01: I joined Microsoft full-time in 1997.
[14:11] SPEAKER_01: I was actually an intern in 1996, but I joined out of college.
[14:15] SPEAKER_01: I think I graduated in May of 1997, and I joined Microsoft on June 16th.
[14:23] SPEAKER_00: And obviously,
[14:24] SPEAKER_00: in history books,
[14:26] SPEAKER_00: we don't look back on you joining specifically,
[14:28] SPEAKER_00: but it was Visual Studio released.
[14:30] And of course,
[14:30] later,
[14:31] Visual Studio became this really,
[14:32] really powerful developer of tools.
[14:34] SPEAKER_00: I started my career working on Microsoft Technologies and Visual Studio.
[14:40] Back then,
[14:41] you were at Microsoft as an intern before this came out,
[14:45] Visual Studio came out.
[14:46] SPEAKER_00: What was the story behind it?
[14:49] SPEAKER_00: Why did Microsoft decide to double down?
[14:51] SPEAKER_00: There were developer tools.
[14:52] SPEAKER_00: I think Microsoft had their QuickSee.
[14:55] SPEAKER_00: There were things that kind of,
[14:58] SPEAKER_00: compared to Visual Studio,
[14:59] SPEAKER_00: it looked like very early versions,
[15:00] SPEAKER_00: but there were things like that.
[15:02] SPEAKER_00: How did this whole development come about?
[15:05] SPEAKER_00: And then why did Microsoft decide to just double and triple down on it?
[15:08] SPEAKER_01: Well,
[15:08] SPEAKER_01: I think one of the big things that happened,
[15:10] SPEAKER_01: you know,
[15:11] SPEAKER_01: really,
[15:11] SPEAKER_01: and we kind of,
[15:13] SPEAKER_01: uh,
[15:13] really started the project in 98 was both the real emergence of
[15:21] Visual Studio as a
[15:22] SPEAKER_01: tool and then also .NET.
[15:24] SPEAKER_01: And they kind of,
[15:25] SPEAKER_01: you know,
[15:26] SPEAKER_01: in,
[15:26] SPEAKER_01: in 2000,
[15:26] we unveiled kind of .NET and Visual Studio .NET together at the
[15:31] PDC conference in
[15:32] SPEAKER_01: July of 2000,
[15:32] SPEAKER_01: if I remember.
[15:34] SPEAKER_01: And, um,
[15:35] SPEAKER_01: You know,
[15:35] SPEAKER_01: the impetus behind that was,
[15:36] SPEAKER_01: you know,
[15:37] SPEAKER_01: Visual Basic had been hugely successful in the mid to late 90s.
[15:45] But there was still this gap in between where you had Visual
[15:47] Basic or Visual Fox
[15:48] SPEAKER_01: Pro,
[15:49] SPEAKER_01: and then you had on the right hand side,
[15:51] SPEAKER_01: in the more advanced category,
[15:52] SPEAKER_01: C++.
[15:53] SPEAKER_01: And C++ was, and still is, a key programming language.
[15:59] SPEAKER_00: And there was MFC, right?
[16:00] SPEAKER_00: The MFC library.
[16:00] SPEAKER_00: There's MFC libraries.
[16:01] SPEAKER_00: Which I guess you can't talk about Windows development.
[16:05] SPEAKER_00: It wasn't just C++.
[16:06] SPEAKER_00: MFC made it so productive.
[16:08] SPEAKER_01: Yeah, and there was also a thing called ATL, which is another library.
[16:12] SPEAKER_01: I haven't thought of that in a long time, but yeah.
[16:14] SPEAKER_01: So we had a couple different framework libraries,
[16:16] SPEAKER_01: but it's still,
[16:17] SPEAKER_01: if you use C++,
[16:20] SPEAKER_01: MFC made a huge difference,
[16:23] SPEAKER_01: but it wasn't as productive as it could be.
[16:27] SPEAKER_01: And there was a sort of gap in the between.
[16:29] SPEAKER_01: And part of why we created .NET was to sort of say,
[16:31] SPEAKER_01: okay,
[16:32] SPEAKER_01: can we create this common language runtime that could handle VB
[16:37] could handle a C++ like language and could handle a bunch of
[16:40] languages in between.
[16:41] SPEAKER_01: And then,
[16:42] SPEAKER_01: you know,
[16:43] could we also avoid each language having to have its own
[16:46] programming framework?
[16:48] SPEAKER_01: And so like MFC only worked with C++.
[16:50] SPEAKER_01: It didn't work with VB.
[16:51] SPEAKER_01: You know, the VB designer didn't work with C++.
[16:53] SPEAKER_01: And, you know, instead, could we build a common set of developer tools
[16:57] SPEAKER_01: whether it was debugging,
[16:59] SPEAKER_01: whether it was around visual design,
[17:01] SPEAKER_01: whether it was around code optimization,
[17:03] SPEAKER_01: profiling,
[17:04] SPEAKER_01: et cetera,
[17:05] SPEAKER_01: IntelliSense,
[17:06] SPEAKER_01: and leverage it.
[17:08] SPEAKER_01: That's where really.NET and Visual Studio really took root.
[17:14] SPEAKER_01: Basically, we came up with the Common Language Runtime,
[17:17] SPEAKER_01: I started a project called ASP.NET with Mark Anders,
[17:21] SPEAKER_01: the two of us,
[17:24] SPEAKER_01: and that was a web framework.
[17:27] SPEAKER_00: When did that start compared to .NET?
[17:29] SPEAKER_00: Did .NET come first or was ASP.NET parallel to all of this?
[17:33] SPEAKER_01: Parallel,
[17:33] SPEAKER_01: I mean,
[17:34] SPEAKER_01: basically,
[17:35] SPEAKER_01: ASP.NET started,
[17:36] SPEAKER_01: I mean,
[17:36] SPEAKER_01: I kind of wrote the original prototype over Christmas 97 to 98.
[17:41] SPEAKER_01: Wow.
[17:42] And the prototype I wrote was,
[17:45] you know,
[17:46] I used some C++,
[17:47] some JavaScript,
[17:48] and some Java.
[17:49] SPEAKER_01: It was kind of a,
[17:50] SPEAKER_01: you know,
[17:50] SPEAKER_01: it was more the idea as opposed to,
[17:53] SPEAKER_01: there was no code that was actually reused.
[17:55] But it was,
[17:55] you know,
[17:55] it was this idea of like,
[17:56] okay,
[17:57] could you use classes?
[17:58] SPEAKER_01: Could you use objects?
[18:00] And could you have language productivity that allowed you to
[18:03] kind of work very quickly?
[18:05] SPEAKER_01: And so I started,
[18:06] SPEAKER_01: you know,
[18:07] SPEAKER_01: we started,
[18:07] Mark and I started showing it to a lot of people internally and
[18:10] got a lot of
[18:11] SPEAKER_01: excitement around it.
[18:11] Sort of in parallel at the same timeframe,
[18:14] the common language runtime got started.
[18:16] SPEAKER_01: It wasn't called .NET, but it was, it was called Core.
[18:21] SPEAKER_01: It was the original, I think, code name.
[18:25] SPEAKER_01: But they were kind of building a runtime that could do languages.
[18:28] SPEAKER_01: They didn't have libraries, but they had languages.
[18:31] SPEAKER_01: And then Visual Studio was trying to figure out,
[18:33] SPEAKER_01: okay,
[18:33] SPEAKER_01: we have the VB IDE,
[18:34] SPEAKER_01: we have the Visual C++ IDE,
[18:38] SPEAKER_01: we have this Java IDE.
[18:39] SPEAKER_01: How do we merge that?
[18:40] SPEAKER_01: Yeah, J Sharp, right?
[18:41] SPEAKER_01: J Sharp.
[18:42] SPEAKER_01: And then it all kind of came together.
[18:45] SPEAKER_01: And so in the process of 98, these three teams sort of found each other.
[18:50] SPEAKER_01: And we started working together and,
[18:53] SPEAKER_01: um,
[18:54] SPEAKER_01: 99,
[18:54] SPEAKER_01: we built a whole bunch of stuff,
[18:55] SPEAKER_01: including windows libraries and,
[18:58] SPEAKER_01: uh,
[18:58] SPEAKER_01: for GUI and for other things.
[18:59] SPEAKER_01: And then ultimately, um, we were supposed to release it to the world.
[19:03] I think it was like the equivalent of build that we called the
[19:05] PDC back then.
[19:06] SPEAKER_01: I think it was supposed to be February or March of 2000 and we were late.
[19:11] Um,
[19:11] and so we slipped it to July and,
[19:13] uh,
[19:13] July 2000 was when we kind of unveiled it at,
[19:17] SPEAKER_01: a big event,
[19:18] SPEAKER_01: much like the build event we're doing today,
[19:20] SPEAKER_01: and showed the world languages,
[19:23] SPEAKER_01: frameworks,
[19:24] SPEAKER_01: and tools all working together.
[19:25] SPEAKER_01: And that was kind of really the unveiling of .NET and tools.
[19:29] SPEAKER_01: And it was pretty critical for our success in the 2000s.
[19:34] SPEAKER_01: In the same way that VB really helped drive Windows client,
[19:37] .NET really helped Windows Server and SQL Server and really
[19:42] introduced Microsoft to
[19:43] SPEAKER_01: a generation of server programmers.
[19:46] SPEAKER_00: Yeah,
[19:46] and I think one of the most iconic videos and now memes is
[19:51] Steve Ballmer yelling,
[19:52] SPEAKER_00: developers,
[19:52] SPEAKER_00: developers,
[19:53] SPEAKER_00: developers,
[19:54] SPEAKER_00: sweating in the t-shirt.
[19:55] SPEAKER_00: And usually, that's the only part that gets quoted.
[19:58] SPEAKER_00: And people think, oh, Microsoft is all about developers.
[20:01] SPEAKER_00: But when you watch the whole thing, which I did,
[20:03] the whole thing that he said,
[20:04] he starts at,
[20:05] okay,
[20:06] what's a $64,000 question?
[20:08] SPEAKER_00: What the hell are we supposed to do with .NET, Steve?
[20:11] SPEAKER_00: And then he goes developer.
[20:12] SPEAKER_00: So as I understand,
[20:14] this was about before the .NET release about how to reach
[20:19] developers with .NET.
[20:21] SPEAKER_00: Can you bring a little bit more behind the meme?
[20:24] SPEAKER_00: Because I feel there's way more depth in his statement.
[20:29] SPEAKER_01: Yeah, I wasn't at that event, so I don't have all the cuttings.
[20:32] SPEAKER_01: But I think
[20:33] SPEAKER_01: I think the main point that Steve was trying to get across and,
[20:36] SPEAKER_01: you know,
[20:37] you see it if you ever watched the video is just the passion
[20:41] you had around
[20:42] SPEAKER_01: developers.
[20:43] SPEAKER_01: And his main point was just developers really matter.
[20:47] SPEAKER_01: And so when people say, why are you doing all this?
[20:48] SPEAKER_01: It's because of developers and developers, developers, developers.
[20:51] SPEAKER_01: And I think that that is kind of the,
[20:54] SPEAKER_01: just a critical nature,
[20:55] SPEAKER_01: which is,
[20:56] SPEAKER_01: again,
[20:56] SPEAKER_01: if you want to build a platform,
[20:57] SPEAKER_01: if you want to have an ecosystem that,
[21:00] SPEAKER_01: You have to have developers.
[21:01] SPEAKER_01: And,
[21:01] SPEAKER_01: you know,
[21:02] ultimately developers are the ones that both build the most
[21:05] interesting solutions
[21:06] SPEAKER_01: and they also push the platform and apps the hardest.
[21:10] SPEAKER_01: And,
[21:10] SPEAKER_01: you know,
[21:11] I think that that was his main goal was just to get across to
[21:13] the audience just how
[21:14] SPEAKER_01: passionate he was around that.
[21:16] SPEAKER_01: It wasn't about money.
[21:17] SPEAKER_01: It wasn't about press.
[21:18] SPEAKER_01: It's really around, are you winning the hearts and minds of developers?
[21:21] And I think that's a big part of why Microsoft Build is special
[21:24] is even take today's keynotes.
[21:28] SPEAKER_01: It's tons of demos, lots of hands-on, live demos, hands-on, lots of labs.
[21:34] SPEAKER_01: It really is an opportunity for developers to get together.
[21:37] SPEAKER_01: And it's not about
[21:38] SPEAKER_01: sort of chest pumping.
[21:39] SPEAKER_01: It's more around,
[21:41] SPEAKER_01: here's what you can do,
[21:42] and how do we have a good conversation and dialogue around what
[21:47] can we do better at
[21:47] SPEAKER_01: Microsoft,
[21:48] SPEAKER_01: but also how can we make you successful?
[21:51] SPEAKER_00: And so .NET was huge when it launched, with Visual Studio hand-in-hand.
[21:55] But a third part of why I think it was really successful at
[21:57] this time was C Sharp itself.
[22:00] SPEAKER_00: Where did C Sharp come from?
[22:01] SPEAKER_00: Well, it must have come from inside Microsoft, but was it
[22:05] SPEAKER_00: Before .NET, was it during .NET?
[22:07] Because I do remember that C Sharp kept evolving as well with
[22:10] new features.
[22:12] SPEAKER_00: Well,
[22:12] SPEAKER_00: hand-in-hand with .NET,
[22:14] SPEAKER_00: Link,
[22:14] SPEAKER_00: for example,
[22:15] SPEAKER_00: was a good example where it's both a language feature,
[22:17] SPEAKER_00: but you needed framework support as well.
[22:20] SPEAKER_01: Well,
[22:20] SPEAKER_01: you know,
[22:21] SPEAKER_01: the real genius behind C-Sharp is Anders Heilsberg,
[22:26] SPEAKER_01: especially in the early days.
[22:27] SPEAKER_01: And, you know, Anders is still at Microsoft.
[22:29] SPEAKER_01: He's still building languages.
[22:30] SPEAKER_01: I mean, he also is responsible for TypeScript.
[22:33] SPEAKER_01: And before Microsoft,
[22:36] SPEAKER_01: Anders worked at Borland,
[22:39] SPEAKER_01: which is a name that most people,
[22:41] SPEAKER_01: if you're not my age,
[22:43] SPEAKER_01: don't remember.
[22:44] SPEAKER_01: But it was an iconic developer tools company in the...
[22:50] early 90s,
[22:51] late 80s,
[22:52] and built some amazing tools,
[22:54] one of which was called Turbo Pascal.
[22:56] SPEAKER_01: Yep.
[22:57] SPEAKER_01: And Anders was the guy who wrote Turbo Pascal.
[22:59] SPEAKER_01: And he wrote it originally from Denmark.
[23:02] I think he wrote it when he was in Denmark and sold it or
[23:05] licensed it to Borland.
[23:06] SPEAKER_01: And part of what made Turbo Pascal revolutionary,
[23:09] SPEAKER_01: this even came out before,
[23:10] SPEAKER_01: I think,
[23:10] SPEAKER_01: VB,
[23:11] SPEAKER_01: Visual Basic,
[23:12] SPEAKER_01: was it was just lightning fast.
[23:14] SPEAKER_01: And so you could literally, on a PC with 256K of RAM,
[23:22] SPEAKER_01: You know,
[23:22] SPEAKER_01: it had an editor,
[23:23] SPEAKER_01: it had a debugger,
[23:24] SPEAKER_01: and if you hit run,
[23:25] SPEAKER_01: you know,
[23:26] SPEAKER_01: in a few seconds,
[23:26] SPEAKER_01: your Pascal app would work.
[23:28] And,
[23:29] you know,
[23:29] he added good language features into Pascal and really built
[23:32] that.
[23:32] SPEAKER_01: And, you know, we were very fortunate.
[23:34] SPEAKER_01: He joined Microsoft in the mid-90s,
[23:37] SPEAKER_01: along with a bunch of Borland employees,
[23:39] SPEAKER_01: and really brought both that developer ethos,
[23:42] SPEAKER_01: he kind of really helped rejuvenate the developer ethos at the time,
[23:45] SPEAKER_01: and then also just a language sensibility.
[23:48] SPEAKER_01: And, you know, I've worked with Anders now for 25 plus years.
[23:52] SPEAKER_01: And he's just,
[23:53] he's absolute genius in terms of understanding both what to add
[23:58] into a language and
[23:59] SPEAKER_01: what not to.
[24:01] And there's a real aesthetic where it's easy to kind of just
[24:05] throw in the kitchen
[24:05] SPEAKER_01: sink into a language,
[24:07] SPEAKER_01: you know,
[24:07] SPEAKER_01: but how do you make these things make sense?
[24:09] How do you,
[24:10] you know,
[24:10] you mentioned Link,
[24:11] which was a language query technology.
[24:15] SPEAKER_01: And
[24:16] SPEAKER_01: You know,
[24:16] I think part of what made the elegance of Link at the time so
[24:20] great was it built on
[24:22] SPEAKER_01: generics,
[24:23] SPEAKER_01: which was also built into the language,
[24:25] SPEAKER_01: which then composed into the runtime.
[24:27] SPEAKER_00: Yeah, and generics in C Sharp was very powerful.
[24:29] SPEAKER_01: Yeah,
[24:30] SPEAKER_01: and it was,
[24:30] SPEAKER_01: you know,
[24:31] SPEAKER_01: at the time it was a big differentiator versus Java,
[24:33] SPEAKER_01: which didn't have generics when C Sharp first introduced it.
[24:37] SPEAKER_01: And just the way that Anders kind of saw over multiple generations,
[24:42] SPEAKER_01: okay,
[24:42] SPEAKER_01: we're going to add generics into the CLR,
[24:44] SPEAKER_01: We're going to add it into the language.
[24:46] SPEAKER_01: And then the next version, we're going to come up with link.
[24:49] SPEAKER_01: And, you know, it was just, it's, it's been sort of a mastery to see.
[24:53] SPEAKER_01: And I think versus other languages,
[24:55] SPEAKER_01: there's great continuity throughout it versus,
[24:58] SPEAKER_01: you know,
[24:59] SPEAKER_01: different languages.
[24:59] SPEAKER_01: I think sometimes have taken kind of a left turn or right turn.
[25:04] SPEAKER_01: And not always a kind of linear progression.
[25:07] SPEAKER_01: And they've taken detours.
[25:09] SPEAKER_01: And I think it's been great to see the way Anders is,
[25:13] SPEAKER_01: both with C Sharp and also with TypeScript,
[25:15] has kind of had a vision that he's built on top of that really
[25:20] has a bunch of
[25:20] SPEAKER_01: consistency and sort of common direction over time.
[25:25] SPEAKER_00: And then in the early and mid and late 2000s,
[25:30] the Microsoft ecosystem was really special in the sense when I
[25:32] worked at one of my
[25:33] SPEAKER_00: first companies,
[25:34] SPEAKER_00: we were paying $1,000,
[25:37] SPEAKER_00: $2,000,
[25:43] SPEAKER_00: ASP.NET,
[25:44] getting access to the different software like SQL Server and
[25:48] IIS so we could
[25:48] SPEAKER_00: develop and use them because licenses were very expensive otherwise.
[25:51] $3,000 per developer to access Visual Studio together with C Sharp,
[25:52] SPEAKER_00: And to MSDN library.
[25:54] SPEAKER_00: Now,
[25:54] I just want to pause on the MSDN library because I've never
[25:56] really seen a company
[25:57] do this before or after where this was before Stack Overflow
[26:02] and even before having
[26:03] SPEAKER_00: good things on the internet.
[26:04] SPEAKER_00: We would pay to access really, really good documentation.
[26:09] SPEAKER_00: Early on, it was sent out on CDs because they were so big.
[26:13] SPEAKER_00: Do you remember,
[26:13] SPEAKER_00: like,
[26:14] SPEAKER_00: how or why did Microsoft get this idea of,
[26:16] SPEAKER_00: like,
[26:16] SPEAKER_00: let's just invest in documentation,
[26:18] SPEAKER_00: especially because as a developer,
[26:19] SPEAKER_00: I'm going to be honest,
[26:20] SPEAKER_00: you know,
[26:20] SPEAKER_00: like,
[26:20] SPEAKER_00: it's usually one of the last things I come around to.
[26:24] SPEAKER_01: Yeah,
[26:24] SPEAKER_01: it's,
[26:25] SPEAKER_01: you know,
[26:25] SPEAKER_01: I feel like I'm dating myself on this podcast,
[26:27] SPEAKER_01: but,
[26:28] SPEAKER_01: you know,
[26:29] SPEAKER_01: I think for a lot of listeners,
[26:30] SPEAKER_01: you know,
[26:31] SPEAKER_01: the idea of,
[26:31] SPEAKER_01: like,
[26:32] SPEAKER_01: buying documentation sounds weird.
[26:34] SPEAKER_01: These days, yes, but back then, we did it.
[26:36] SPEAKER_01: Back then, it was revolutionary.
[26:37] And part of it was,
[26:38] you know,
[26:38] back,
[26:39] like,
[26:39] when I joined Microsoft or was an intern,
[26:42] SPEAKER_01: you know,
[26:42] SPEAKER_01: the internet was still very,
[26:44] very new and there wasn't like a search engine out there that
[26:49] was very good.
[26:50] SPEAKER_01: And it was all pre-Google, pre-Bang.
[26:52] SPEAKER_01: And, you know, HTML was still pretty rudimentary.
[26:57] SPEAKER_01: You know,
[26:59] SPEAKER_01: I remember up until about 2002,
[27:01] SPEAKER_01: 2003,
[27:01] SPEAKER_01: you know,
[27:03] SPEAKER_01: a lot of sites did not even rely on JavaScript or
[27:07] most used very,
[27:08] very minimal JavaScript because some browsers didn't support
[27:11] it.
[27:11] SPEAKER_01: You know, this was pre-CSS.
[27:13] SPEAKER_01: And so as a result,
[27:14] SPEAKER_01: you know,
[27:14] the idea of searching for documentation and reading it in HTML
[27:20] in a browser was
[27:21] SPEAKER_01: just not done.
[27:22] SPEAKER_01: And so,
[27:22] SPEAKER_01: you know,
[27:23] SPEAKER_01: the original impetus,
[27:24] SPEAKER_01: I think,
[27:24] SPEAKER_01: between MSDN in the 90s was this subscription service.
[27:28] SPEAKER_01: And you got
[27:29] SPEAKER_01: Every quarter, every month, you got the CD.
[27:32] SPEAKER_01: Yeah, you got a lot of CDs.
[27:33] SPEAKER_01: I remember it was, you know, 50 CDs sometimes.
[27:36] SPEAKER_01: Yeah.
[27:36] SPEAKER_01: You know, it would have all the updated applications.
[27:39] So it would have all the Microsoft operating systems,
[27:41] databases,
[27:43] developer tools...
[27:44] SPEAKER_01: I think even office at one point was included.
[27:46] SPEAKER_00: Yeah.
[27:47] SPEAKER_00: More and more things came into it.
[27:48] SPEAKER_01: And then you had like tons and tons of CDs with the documentation.
[27:50] SPEAKER_01: And so you could kind of install it and then you could search and,
[27:54] SPEAKER_01: um,
[27:54] SPEAKER_01: you know,
[27:55] SPEAKER_01: that coupled with IntelliSense,
[27:57] SPEAKER_01: which is,
[27:58] SPEAKER_01: or,
[27:58] SPEAKER_01: or,
[27:58] SPEAKER_01: you know,
[27:58] SPEAKER_01: statement completion,
[28:00] SPEAKER_01: which is another thing that I think a lot of people take for granted now.
[28:03] And then the other thing that I think we also take for granted
[28:05] is,
[28:05] is debuggers.
[28:07] SPEAKER_01: Um,
[28:08] SPEAKER_01: I remember when I was in university,
[28:10] the debugging experience was often print F statements or a
[28:16] command line debugger
[28:17] SPEAKER_01: that was very rudimentary.
[28:19] You could kind of dump your symbols or registers,
[28:25] but it wasn't like debugging today.
[28:27] SPEAKER_01: And so as a result, the MSDN was for a different time, I guess.
[28:31] SPEAKER_01: It's a little quaint, but it evolved over time.
[28:34] SPEAKER_01: And part of it also became this notion of a subscription service where
[28:37] SPEAKER_01: your apps were always up to date.
[28:39] And so I think at some point the documentation just turned on
[28:42] to be on the
[28:43] SPEAKER_01: internet,
[28:44] SPEAKER_01: but the notion of the apps was still.
[28:46] SPEAKER_00: But I think one of my takeaways is that when I think back,
[28:49] SPEAKER_00: like I was at a startup,
[28:51] SPEAKER_00: like a small company,
[28:52] SPEAKER_00: 50% company,
[28:53] SPEAKER_00: it was actually in Hungary.
[28:55] SPEAKER_00: We didn't have that much.
[28:57] SPEAKER_00: like revenue or income compared to the US.
[28:59] And my employer still paid that $1,000 or $2,000 per developer
[29:04] per year because the
[29:06] SPEAKER_00: developers were so much more productive using this whole Microsoft stack.
[29:11] SPEAKER_00: And I know things have changed,
[29:12] SPEAKER_00: but to me,
[29:13] SPEAKER_00: that point in time is a bit of a reminder that there is a big premium
[29:17] on just being so much more productive than anything else
[29:19] because back then you know
[29:20] everything was already together i think we take these days it's
[29:23] it's so like
[29:24] everything that we're amazed about having documentation of
[29:26] working debugger you
[29:28] know like software that you can use for free like for database
[29:31] software like that
[29:32] SPEAKER_00: was sql server back then
[29:34] SPEAKER_00: But it was all there.
[29:34] And I wonder if we'll move this notion that a company that
[29:38] everything's so much
[29:39] SPEAKER_00: better and built an ecosystem that just worked,
[29:42] SPEAKER_00: it was really valuable.
[29:43] SPEAKER_00: It probably applies today as well.
[29:45] Maybe not in a multi-thousand dollar subscription,
[29:48] but something is there.
[29:49] SPEAKER_00: I just vividly remember how no one forced anyone to use Microsoft.
[29:54] SPEAKER_00: And it was not about Microsoft.
[29:55] SPEAKER_00: It was just about there was this thing.
[29:58] SPEAKER_00: We're working faster,
[29:59] SPEAKER_00: especially when we were building either the first websites,
[30:02] SPEAKER_00: because ASP.NET was still
[30:03] There was a point where 25% of all sites were ASP.NET based on
[30:07] various things.
[30:09] SPEAKER_00: After launch, it just went up on this thing.
[30:11] SPEAKER_00: Do you have theories on why that was?
[30:14] SPEAKER_00: I mean, ASP.NET was your baby a little bit.
[30:17] SPEAKER_01: Well,
[30:17] SPEAKER_01: I think the general thing,
[30:18] SPEAKER_01: I think it's true for pretty much every field,
[30:22] SPEAKER_01: in fact,
[30:23] SPEAKER_01: when it comes to technology,
[30:24] SPEAKER_01: is people like things that make them more productive and
[30:29] SPEAKER_01: let them do more faster, cheaper.
[30:32] SPEAKER_01: Yeah.
[30:32] SPEAKER_01: I think it's,
[30:33] SPEAKER_01: it's always worth reflecting because like on just,
[30:37] SPEAKER_01: you know,
[30:38] SPEAKER_01: imagine 10 or 15 years ago,
[30:39] SPEAKER_01: what the world looked like.
[30:41] SPEAKER_01: You know,
[30:41] SPEAKER_01: if you were going to an event like this,
[30:42] SPEAKER_01: you would either rent a car or,
[30:44] SPEAKER_01: you know,
[30:46] SPEAKER_01: you'd stand in line for a long taxi and whether it gets you there,
[30:49] SPEAKER_01: whether you get ripped off,
[30:50] SPEAKER_01: you know,
[30:50] SPEAKER_01: now you have Uber or Lyft,
[30:52] SPEAKER_01: you know,
[30:53] SPEAKER_01: similarly coffee.
[30:54] SPEAKER_01: Um,
[30:57] SPEAKER_01: or a coffee maker,
[30:58] SPEAKER_01: you know,
[30:58] SPEAKER_01: you'd go and find a store and try eight of them and you'd search a job.
[31:02] You know,
[31:02] now you can order on Amazon and it'll show up same day or next
[31:05] day delivery.
[31:06] SPEAKER_01: And so,
[31:06] SPEAKER_01: you know,
[31:06] I do think we kind of overlook the productivity that's changed
[31:11] in just 10 years on
[31:12] SPEAKER_01: a regular basis.
[31:14] SPEAKER_00: It's not baseline, right?
[31:15] SPEAKER_01: Yeah.
[31:15] SPEAKER_01: When you step back and think about what the world looked like back then,
[31:18] SPEAKER_01: even five or 10 years ago,
[31:19] SPEAKER_01: it's completely different.
[31:20] SPEAKER_01: And I think that's going to be very much true for AI.
[31:23] SPEAKER_01: And when we look at
[31:26] the kind of productivity that something like GitHub Copilot has
[31:29] or other AI systems
[31:32] SPEAKER_01: tools provide,
[31:34] it's going to be like that printf and the debugger kind of
[31:38] story I told earlier,
[31:39] SPEAKER_01: or it's going to be also kind of a quantum leap in terms of productivity.
[31:44] And at the end of the day,
[31:45] if you're getting more productive and saving time,
[31:49] SPEAKER_01: you know, companies and developers are willing to pay for it.
[31:52] SPEAKER_01: If the developer can make more money because they're using those tools.
[31:55] SPEAKER_00: We see from the past, like very hard money.
[31:58] This episode is brought to you by Modal,
[32:00] the cloud platform that makes AI development simple.
[32:03] SPEAKER_00: Need GPUs without the headache?
[32:05] SPEAKER_00: With Modal,
[32:06] SPEAKER_00: just add one line of code to any Python function and boom,
[32:09] SPEAKER_00: it's running in the cloud on your choice of CPU or GPU.
[32:12] SPEAKER_00: And the best part, you only pay for what you use.
[32:15] SPEAKER_00: With sub-second container start and instant scaling to thousands of GPUs,
[32:19] SPEAKER_00: it's no wonder companies like Suno,
[32:21] SPEAKER_00: Ramp,
[32:21] SPEAKER_00: and Substack already trust Modal for their AI applications.
[32:25] SPEAKER_00: Getting an H100 is just a pip install away.
[32:28] SPEAKER_00: Go to modal.com slash pragmatic to get $30 in free credits every month.
[32:32] SPEAKER_00: That is modal.com slash pragmatic.
[32:36] SPEAKER_00: I just wanted to touch briefly on Windows Phone.
[32:40] SPEAKER_00: I was a Windows Phone developer.
[32:43] SPEAKER_00: In fact,
[32:43] I think that's where we connected on one of the events when
[32:45] Microsoft was about to
[32:47] SPEAKER_00: launch Windows Phone.
[32:47] SPEAKER_00: There was a pre-launch event.
[32:49] SPEAKER_00: And to this date,
[32:50] SPEAKER_00: when people ask me what I think about developing on iOS,
[32:53] SPEAKER_00: developing on Android,
[32:53] SPEAKER_00: I tell them,
[32:55] SPEAKER_00: the best development experience I've had was on Windows Phone.
[32:58] SPEAKER_00: So Windows Phone,
[33:01] SPEAKER_00: many listeners will have not used it for obvious reasons,
[33:04] SPEAKER_00: but I was a huge fan,
[33:06] SPEAKER_00: again,
[33:06] SPEAKER_00: not because of Microsoft or anything,
[33:08] SPEAKER_00: but it had some really forward-looking features like Live Tiles,
[33:11] SPEAKER_00: which...
[33:12] SPEAKER_00: are now kind of a given everywhere, but it was years ahead of it.
[33:15] SPEAKER_00: And the development experience was so smooth.
[33:17] SPEAKER_00: We had that first-class debugger, we had the simulator that just worked.
[33:22] SPEAKER_00: And in the end,
[33:22] SPEAKER_00: Windows Phone is now history,
[33:23] SPEAKER_00: but you were there when Windows Phone was born,
[33:27] SPEAKER_00: throughout its rise,
[33:28] SPEAKER_00: and then in the end when Microsoft decided to discontinue it.
[33:31] What are some learnings that you have on building platforms and
[33:35] delighting
[33:35] SPEAKER_00: developers and what works and what doesn't?
[33:38] SPEAKER_01: Yeah, there was some great tools.
[33:41] I ran the Windows Phone development tool team at one point,
[33:45] amongst other things.
[33:48] SPEAKER_01: So I was involved in the developer tool side for Windows Phone.
[33:50] I didn't work on the Windows Phone team itself,
[33:52] but we kind of took .NET and C Sharp and
[33:56] SPEAKER_01: Silverlight technology and XAML and Visual Studio.
[34:00] SPEAKER_01: And I think,
[34:01] SPEAKER_01: to your point,
[34:02] SPEAKER_01: I think we had some really great tools that were pretty differentiated.
[34:04] SPEAKER_00: The tools were really good.
[34:06] I think that if there were two or three lessons from Windows
[34:10] Phone for me,
[34:11] SPEAKER_01: at least from a development perspective,
[34:12] SPEAKER_01: I think one was,
[34:13] SPEAKER_01: which is true,
[34:14] SPEAKER_01: I think,
[34:15] SPEAKER_01: in technology in general,
[34:17] SPEAKER_01: which is
[34:19] SPEAKER_01: If you're not number one,
[34:20] you've got to be number two because it's really hard to be
[34:25] number three and number
[34:26] SPEAKER_01: four in the market and ever catch up.
[34:29] SPEAKER_01: And so in some ways,
[34:30] the Windows Phone project started after the iPhone was released
[34:35] or the modern
[34:36] SPEAKER_01: Windows Phone that you like.
[34:38] And I think Windows Phone would have had a shot other than
[34:41] Android came out.
[34:42] SPEAKER_01: and came out beforehand a short amount of time.
[34:45] And yeah,
[34:46] I think it was really a question of whether Android or Windows
[34:49] Phone was gonna be the
[34:52] SPEAKER_01: the number two in the market.
[34:53] And I think if we'd been a year earlier,
[34:56] maybe it would have turned out differently.
[34:58] I think the other thing is the need to kind of connect with
[35:03] developers broadly.
[35:06] And that means we supported Windows Phone Windows as a
[35:10] development experience,
[35:12] SPEAKER_01: but it didn't work on the Mac.
[35:14] SPEAKER_01: And if you think about to 2009,
[35:16] SPEAKER_01: 2010,
[35:16] SPEAKER_01: if you went to a developer conference,
[35:19] SPEAKER_01: a huge number of people were using Macs.
[35:22] SPEAKER_01: They're using open source.
[35:23] SPEAKER_01: And so basically saying,
[35:26] SPEAKER_01: okay,
[35:26] you got to install Windows on your Mac and use it in order to
[35:30] build an app was a
[35:31] SPEAKER_01: huge issue.
[35:33] SPEAKER_01: And then I think also at that time,
[35:35] SPEAKER_01: you know,
[35:36] SPEAKER_01: designers were exclusively Mac and to some extent today,
[35:39] SPEAKER_00: they still are,
[35:40] SPEAKER_01: but,
[35:41] SPEAKER_01: um,
[35:41] SPEAKER_01: and so,
[35:41] SPEAKER_01: you know,
[35:41] SPEAKER_01: that,
[35:42] SPEAKER_01: that combination was also headwind.
[35:45] SPEAKER_01: Um,
[35:45] SPEAKER_01: you know,
[35:46] SPEAKER_01: and,
[35:46] SPEAKER_01: and,
[35:46] SPEAKER_01: and that combination I think was,
[35:48] SPEAKER_01: was the reason why ultimately it didn't get the escape velocity.
[35:51] SPEAKER_01: And that's, that's why I think speed matters in platform shifts.
[35:57] SPEAKER_01: And that was true for windows.
[35:58] SPEAKER_01: That was true.
[35:59] SPEAKER_01: Um, you know, even the context of Azure, when we,
[36:02] SPEAKER_01: first launched Azure,
[36:03] SPEAKER_01: I think we were number seven in the market on cloud providers,
[36:06] SPEAKER_01: and thankfully we were able to become number two.
[36:08] SPEAKER_00: Yeah,
[36:08] SPEAKER_00: and let's talk about Azure,
[36:09] SPEAKER_00: because after Windows Phone,
[36:12] SPEAKER_00: maybe even during Windows Phone,
[36:15] SPEAKER_00: as I understand,
[36:15] SPEAKER_00: you were a big part of the Azure,
[36:18] SPEAKER_00: even creating Azure.
[36:19] SPEAKER_00: How did Azure start?
[36:20] And how,
[36:20] especially,
[36:21] I think we just need to remind ourselves that this was back
[36:23] then where
[36:24] SPEAKER_00: Windows was everywhere.
[36:26] SPEAKER_00: The internet was just, I guess, starting up.
[36:28] SPEAKER_00: Some visionaries might have seen it,
[36:29] SPEAKER_00: but it was not that obvious that it will be as big as it would be.
[36:32] Microsoft was making so much money from Windows licenses and
[36:35] selling developer tools.
[36:37] SPEAKER_00: CDs were arriving on the mail.
[36:39] SPEAKER_00: And then there was some big investment,
[36:42] SPEAKER_00: lots of people starting to work on this thing called Azure,
[36:44] SPEAKER_00: which now we know it's big,
[36:46] SPEAKER_00: but back then it wasn't.
[36:47] What was your conviction and how did this whole project get off
[36:50] the ground?
[36:51] SPEAKER_00: Especially as you said, you were number seven when you started.
[36:54] Well,
[36:54] you know,
[36:55] I think we introduced Azure to the world in the 2008 PDC.
[36:59] SPEAKER_01: So it's kind of the equivalent of build.
[37:01] SPEAKER_01: So in 2008, and I think it went general availability in 2010.
[37:06] SPEAKER_01: So the 2008 was kind of a preview.
[37:08] SPEAKER_01: And, you know, back then in 2010, cloud was still very new.
[37:12] SPEAKER_01: I mean, Amazon was the leader.
[37:15] But there were lots of,
[37:16] I call it hosting companies,
[37:17] you know,
[37:18] Rackspace and
[37:20] and a whole bunch of companies that probably are less familiar
[37:24] name wise today that
[37:27] SPEAKER_01: different companies had cloud solutions,
[37:30] SPEAKER_01: but they're really kind of more hosting solutions.
[37:34] SPEAKER_01: And Azure did have,
[37:36] SPEAKER_01: when it came out,
[37:37] SPEAKER_01: it pioneered this idea of platform as a service,
[37:40] SPEAKER_01: meaning kind of some higher level services that made it easier
[37:43] SPEAKER_01: for developers to build solutions.
[37:46] SPEAKER_01: But,
[37:46] SPEAKER_01: you know,
[37:47] SPEAKER_01: I would say the platform is,
[37:48] SPEAKER_01: you know,
[37:49] SPEAKER_01: it had some usability issues in that timeframe.
[37:53] SPEAKER_01: And it also did not support Linux or open source at all.
[37:58] SPEAKER_01: And the tooling was not great.
[38:01] SPEAKER_01: And,
[38:01] SPEAKER_01: you know,
[38:02] SPEAKER_01: I'd say by 2010,
[38:02] SPEAKER_01: when it went general availability,
[38:04] SPEAKER_01: it wasn't doing super well in the market.
[38:07] SPEAKER_01: And I think we were like number seven or number eight or number six.
[38:10] SPEAKER_01: I can't remember what it was.
[38:11] SPEAKER_01: But beginning in 2011, I think Satya...
[38:13] SPEAKER_01: took over what was in the server and tools business where Azure lived.
[38:17] SPEAKER_01: And I think it was about three or four weeks later,
[38:20] SPEAKER_01: two or three weeks later,
[38:21] SPEAKER_01: you kind of wandered into my office and said,
[38:24] SPEAKER_01: hey,
[38:25] SPEAKER_01: how would you like to work on Azure?
[38:27] SPEAKER_01: And at the time I was in the developer division
[38:30] SPEAKER_01: For some reason, I said yes.
[38:34] At the time,
[38:35] a lot of people emailed me and they're like,
[38:36] what are you doing?
[38:37] SPEAKER_01: This is career suicide.
[38:39] SPEAKER_01: This is not going to go anywhere.
[38:42] After we announced it,
[38:44] I got lots of mail from people telling me I'd made a huge
[38:46] mistake.
[38:47] There was a brief moment I was like,
[38:48] well,
[38:48] maybe I made a mistake,
[38:49] I don't know.
[38:50] But this idea of having kind of this cloud-based computer that
[38:54] you could build
[38:55] SPEAKER_01: platform,
[38:56] build solutions on top of and run them at scale and
[38:59] dramatically change the curve
[39:00] SPEAKER_01: of productivity
[39:03] SPEAKER_01: as well as success for startups and small companies and big companies,
[39:09] SPEAKER_01: you know,
[39:10] SPEAKER_01: it was enticing.
[39:11] SPEAKER_01: And so, yeah, I think I started in 2011.
[39:14] SPEAKER_01: You know,
[39:15] one of the first things we did that's kind of gone down in lore
[39:19] was I kind of
[39:20] SPEAKER_01: worked on the project for maybe 60 days,
[39:22] SPEAKER_01: played with the product a lot,
[39:23] SPEAKER_01: and said,
[39:24] SPEAKER_01: gosh,
[39:24] SPEAKER_01: we have a lot of things to fix.
[39:27] And so we had sort of this off-site where we brought together
[39:31] all the senior
[39:31] SPEAKER_01: leaders and architects into a room.
[39:34] And I went to Safeway and bought these sort of Visa credit
[39:39] cards that were like
[39:40] SPEAKER_01: prepaid and gave one to each table.
[39:43] SPEAKER_01: And we mixed the teams up and we said, you're going to build an app.
[39:46] SPEAKER_01: And we have two days to build an app together, starting with signing up.
[39:49] SPEAKER_01: And, you know, half the people couldn't figure out how to sign up.
[39:53] SPEAKER_01: And
[39:53] SPEAKER_01: You know, half the people struggled to get the tools installed.
[39:56] SPEAKER_01: The documentation was out of date.
[39:58] SPEAKER_01: It didn't work.
[39:59] And,
[39:59] you know,
[40:00] the idea was just,
[40:01] you know,
[40:01] can you build a Hello World app?
[40:02] SPEAKER_01: And it was an eye opener,
[40:04] SPEAKER_01: I think,
[40:05] SPEAKER_01: to a lot of people of like,
[40:06] SPEAKER_01: oh,
[40:07] SPEAKER_01: I thought my part was great,
[40:08] SPEAKER_01: but no one can use it.
[40:10] SPEAKER_01: And, you know, we used that offsite at the end.
[40:13] SPEAKER_01: I kind of went to the whiteboard and said, okay.
[40:15] SPEAKER_01: Let's list all the things we need to fix over the next year.
[40:17] SPEAKER_01: And we kind of created a punch list.
[40:20] SPEAKER_01: Just starting as a new user who wants to.
[40:22] SPEAKER_01: Starting with a website and sign up, start with documentation.
[40:25] SPEAKER_01: Let's start with the developer tools.
[40:27] SPEAKER_01: You know, hey, we probably need to support open source.
[40:30] Yeah,
[40:30] we created this punch list and 12 months later,
[40:33] we relaunched Azure in 2012.
[40:37] SPEAKER_01: Started to get some good traction.
[40:38] SPEAKER_01: And as part of that, we also supported Linux.
[40:40] SPEAKER_01: We supported VMs.
[40:43] SPEAKER_00: That was a huge change.
[40:44] SPEAKER_00: I still remember.
[40:44] SPEAKER_01: Which we didn't do.
[40:45] SPEAKER_01: And then in 2014, we renamed Azure to be Microsoft Azure.
[40:50] SPEAKER_01: Back then, before that, it was Windows Azure.
[40:52] SPEAKER_01: And then as part of that,
[40:53] we also very much focused on businesses because we realized
[40:59] Amazon really owned the
[41:01] SPEAKER_01: consumer startup space,
[41:03] SPEAKER_01: which was the biggest market in cloud at that time.
[41:07] And I think one of the things that was part of my lessons from
[41:10] watching the Windows
[41:11] Phone project was if you simply try to do exactly what your
[41:14] competitor does and
[41:15] SPEAKER_01: they're ahead of you,
[41:15] SPEAKER_01: it's hard to catch up.
[41:17] Pick something,
[41:18] pick a beachhead that is small enough to win,
[41:22] big enough to matter.
[41:22] SPEAKER_01: And we sort of said, let's be the cloud for modern business.
[41:25] SPEAKER_01: It was kind of the tagline.
[41:26] SPEAKER_01: It was a hybrid,
[41:28] SPEAKER_01: take advantage of cloud,
[41:29] SPEAKER_01: use cloud,
[41:30] SPEAKER_01: connect to the existing enterprise you already have.
[41:33] SPEAKER_01: And we were able to kind of build differentiation in Azure
[41:36] that went after a segment that was still pretty small at the
[41:40] time and Amazon wasn't great at.
[41:42] And that was the key thing that kind of took us from like
[41:45] number seven to number
[41:46] SPEAKER_01: two was people said,
[41:47] SPEAKER_01: okay,
[41:49] SPEAKER_01: I get why I might use Amazon and Azure.
[41:52] And it helped us get some of that escape velocity and put us in
[41:56] second place.
[41:57] SPEAKER_01: And then we've kind of grown every year since then.
[41:59] SPEAKER_01: And now we have done,
[42:00] SPEAKER_01: you know,
[42:01] SPEAKER_01: it's ChatGPT and we have lots of other- Yeah,
[42:03] SPEAKER_00: now we're going head on hatch.
[42:03] But it helped us in 2014 to 2016 or 17 that we were kind of
[42:05] great in hybrid cloud
[42:11] SPEAKER_01: And Amazon kind of didn't want to go after that space.
[42:14] SPEAKER_01: And as a result, that really helped us accelerate and become relevant.
[42:18] SPEAKER_00: And this is probably a good lesson for even like startups,
[42:20] SPEAKER_00: small startups who are in a very crowded space,
[42:22] SPEAKER_00: let's say with AI,
[42:23] SPEAKER_00: like,
[42:24] SPEAKER_00: you know,
[42:24] SPEAKER_00: look at the places that might be a bit more opportunity.
[42:27] If you can find an underserved market where people are
[42:31] desperate for a solution,
[42:33] That's a great place whether you're a startup or you're an
[42:35] established company to
[42:36] SPEAKER_01: kind of build a new product.
[42:38] If all you're doing is building the same thing that your
[42:41] competitor's doing,
[42:42] SPEAKER_01: but you're coming out later without as much share,
[42:44] SPEAKER_01: It's really, really hard, the history of computing, to catch up.
[42:47] So you mentioned open source,
[42:48] and a big shift in the 2010s was open source.
[42:51] SPEAKER_00: Microsoft did have a bit of a,
[42:53] SPEAKER_00: felt to me,
[42:53] SPEAKER_00: a half-hearted attempt with something called CodePlex.
[42:56] SPEAKER_00: A lot of people will not remember it.
[42:58] SPEAKER_00: It was basically like, hey, host your stuff on CodePlex.
[43:01] SPEAKER_00: You can choose your license, or you can opt for your license.
[43:03] SPEAKER_00: But most projects were like non-permissive licenses.
[43:06] SPEAKER_00: You could see the source code, but the license was...
[43:09] SPEAKER_00: It's just not there.
[43:10] SPEAKER_00: And then something happened.
[43:11] SPEAKER_00: So the weird thing was that Microsoft did have this solution,
[43:13] SPEAKER_00: CodePlex,
[43:14] SPEAKER_00: and it was popular with Microsoft.
[43:16] SPEAKER_00: One of my old companies used it.
[43:18] SPEAKER_00: And then Microsoft started to just like first embrace GitHub.
[43:20] SPEAKER_00: Obviously, there was a acquisition of GitHub.
[43:22] But even before,
[43:23] I think Microsoft already started to do a lot more permissive
[43:27] open source.
[43:27] I think the Ajax toolkit might have been one of the first ones
[43:30] to go out there.
[43:32] SPEAKER_00: What happened then?
[43:33] SPEAKER_00: Was this coming internally from below?
[43:35] SPEAKER_00: Was it leaders like yourself saying,
[43:37] SPEAKER_00: all right,
[43:38] SPEAKER_00: we should just be a lot more serious about open source,
[43:42] SPEAKER_00: like proper open source?
[43:44] Yeah,
[43:44] I think it was very much a cultural shift in the company and
[43:50] also a business model shift.
[43:52] SPEAKER_01: I mean,
[43:52] SPEAKER_01: I think ultimately there was philosophical things about open source,
[43:55] but I think a lot of Microsoft's early hesitation around open
[43:59] source was really
[44:00] SPEAKER_01: around business model.
[44:02] And I think it's true today even if you look at any company and
[44:05] what they're doing,
[44:07] SPEAKER_01: you know,
[44:07] SPEAKER_01: start with what is their business model.
[44:09] SPEAKER_01: And often what they do is driven by the business model.
[44:12] SPEAKER_01: And so if you are social media,
[44:16] SPEAKER_01: you care a lot about advertising because that is your business model.
[44:19] And so when you think about privacy or you think about data
[44:23] protection,
[44:24] you know,
[44:26] SPEAKER_01: your position on that matters a lot as to whether you could monetize and,
[44:30] SPEAKER_01: you know,
[44:31] SPEAKER_01: pay the rent.
[44:32] SPEAKER_01: And,
[44:33] SPEAKER_01: you know,
[44:33] SPEAKER_01: similarly,
[44:33] SPEAKER_01: I think if you are a hardware company,
[44:37] SPEAKER_01: you care a lot about the hardware and your,
[44:39] SPEAKER_01: your business model is really around selling the hardware.
[44:41] SPEAKER_01: And so you care about, you know, a different set of things.
[44:44] SPEAKER_01: And,
[44:44] SPEAKER_01: you know,
[44:44] we were at the time in the early nineties or the late nineties
[44:47] and in early two
[44:48] SPEAKER_01: thousands,
[44:49] SPEAKER_01: you know,
[44:49] SPEAKER_01: most of our money came from commercial software licenses,
[44:53] SPEAKER_01: um,
[44:53] SPEAKER_01: And so the idea that like you could buy one copy and give it to everyone,
[44:56] SPEAKER_01: you know,
[44:57] SPEAKER_01: for free,
[44:58] SPEAKER_01: for free was kind of a scary thing at that time.
[45:01] SPEAKER_01: And so,
[45:01] SPEAKER_01: you know,
[45:01] SPEAKER_01: I think,
[45:02] SPEAKER_01: so part of it,
[45:02] SPEAKER_01: part of our kind of concern around open source was really grounded,
[45:06] SPEAKER_01: frankly,
[45:06] SPEAKER_01: in business model.
[45:07] And then I think there was also kind of a misunderstanding of
[45:09] what is open source.
[45:11] SPEAKER_01: And,
[45:11] SPEAKER_01: you know,
[45:12] SPEAKER_01: back then there was sort of GPL and,
[45:14] SPEAKER_01: you know,
[45:15] SPEAKER_01: if you checked in one line of code,
[45:16] SPEAKER_01: would it,
[45:17] SPEAKER_01: you know,
[45:17] SPEAKER_01: the license mean that like you just gave away all your IP?
[45:21] And so there was still in those early days,
[45:23] there was a little bit of misunderstanding.
[45:26] SPEAKER_01: It was before,
[45:27] SPEAKER_01: you know,
[45:27] GPL kind of modified the license a little bit and clarified a
[45:31] little bit what it
[45:31] SPEAKER_01: meant.
[45:32] SPEAKER_01: But I think there was also some fear, uncertainty and doubt.
[45:35] SPEAKER_01: that a lot of companies, not just Microsoft, kind of had around it.
[45:38] SPEAKER_01: And so, you know, that was the 90s.
[45:40] SPEAKER_01: I think what changed in the late 90s,
[45:43] SPEAKER_01: and I probably played a big role in this,
[45:45] SPEAKER_01: was just recognizing,
[45:47] SPEAKER_01: but if you're a new developer trying to say you need to pay us money for
[45:51] SPEAKER_01: everything,
[45:52] SPEAKER_01: whereas something else,
[45:53] SPEAKER_01: and that you can't see the code,
[45:54] SPEAKER_01: you can't help contribute to the code,
[45:56] SPEAKER_01: you can't participate in the development,
[45:58] SPEAKER_01: you know, you're swimming against the tide.
[46:00] SPEAKER_01: And,
[46:01] SPEAKER_01: you know,
[46:02] SPEAKER_01: at the end of the day,
[46:02] SPEAKER_01: if you are a new or an old developer in that timeframe,
[46:07] SPEAKER_01: you kind of want to be able to see the code.
[46:08] SPEAKER_01: You want to be able to contribute.
[46:11] SPEAKER_01: And so,
[46:12] SPEAKER_01: you know,
[46:12] SPEAKER_01: we took some early steps and it was a little bit,
[46:15] SPEAKER_01: you know,
[46:17] SPEAKER_01: as you mentioned,
[46:18] SPEAKER_01: the Ajax toolkit,
[46:19] SPEAKER_01: I think was the first thing we open source.
[46:21] SPEAKER_00: I think it was.
[46:22] SPEAKER_00: I talked to some folks at Microsoft, so.
[46:24] SPEAKER_01: And I remember another big step we did was we added jQuery,
[46:27] SPEAKER_01: going back memory lane,
[46:30] SPEAKER_01: which was at the time one of the most popular.
[46:33] I think actually prototype was the most popular and jQuery was
[46:35] the up and coming
[46:36] SPEAKER_01: because it was really streamlined.
[46:37] SPEAKER_00: Yeah, it overtook it in the old time.
[46:38] SPEAKER_00: It overtook it.
[46:39] SPEAKER_01: But we kind of added it into the ASP.NET template inside Visual Studio.
[46:44] SPEAKER_01: And that was also considered groundbreaking at the time.
[46:45] SPEAKER_01: I mean,
[46:46] SPEAKER_01: in hindsight,
[46:46] SPEAKER_01: this all sounds trite,
[46:48] SPEAKER_01: I guess,
[46:48] but at the time it was a big thing because suddenly we were
[46:50] taking licenses,
[46:51] SPEAKER_01: I think it was both like MIT or BSD license and incorporating it into our
[46:55] SPEAKER_01: commercial product.
[46:56] And it was,
[46:57] you know,
[46:57] it was a bit of a,
[46:58] we crossed a bridge or a Rubicon.
[47:00] And people said,
[47:00] okay,
[47:01] I guess the sky didn't fall in and maybe we could do more.
[47:05] And so,
[47:05] you know,
[47:06] it was a bit of a journey to kind of take the company through
[47:09] it.
[47:09] SPEAKER_01: But probably my team in particular helped a lot with that.
[47:13] And then the other thing that really changed in the kind of
[47:16] early 2000s or the
[47:16] SPEAKER_01: 2010,
[47:16] SPEAKER_01: 2011,
[47:18] 2012 was just recognizing that also...
[47:22] SPEAKER_01: In the cloud world,
[47:23] you could be very successful if all your customers only use
[47:26] Linux and open source
[47:27] SPEAKER_01: because the business model did also change.
[47:30] SPEAKER_00: And now people are paying for compute,
[47:31] SPEAKER_00: for resources,
[47:32] SPEAKER_00: for services,
[47:33] SPEAKER_00: for all the abstractions if you needed.
[47:37] And we also then recognized at the time that if you didn't
[47:39] really have developers'
[47:43] SPEAKER_01: hearts and minds...
[47:44] They weren't going to pay any attention to the cloud platform
[47:47] you put out or the
[47:48] SPEAKER_01: tools you put out or the databases you put out.
[47:50] And so we kind of went through this shift over a couple of
[47:53] years where I think both
[47:54] SPEAKER_01: culturally,
[47:55] SPEAKER_01: we kind of brought the company along of like,
[47:57] SPEAKER_01: hey,
[47:57] SPEAKER_01: this is not scary.
[47:58] SPEAKER_01: This is good.
[47:59] And then also business model wise,
[48:00] we had permission to kind of experiment and take risks.
[48:04] where you could say,
[48:04] okay,
[48:05] let's make it free or let's make it open source.
[48:07] SPEAKER_00: Which leads us into Visual Studio Code.
[48:10] SPEAKER_00: In 2015,
[48:10] SPEAKER_00: if you would have asked me what is unlikely to happen,
[48:13] SPEAKER_00: I would have said that people use more of Microsoft's IDE,
[48:17] SPEAKER_00: which was Visual Studio.
[48:18] SPEAKER_00: Again, you still had to pay a bunch of money.
[48:19] SPEAKER_00: It was amazing.
[48:20] SPEAKER_00: I came from that world.
[48:21] But when I moved to new companies,
[48:22] they were like,
[48:23] let's use open source stuff.
[48:24] SPEAKER_00: Let's use free stuff.
[48:24] SPEAKER_00: We use Atom or Sublime,
[48:27] SPEAKER_00: or maybe we might have paid for JetBrains,
[48:30] SPEAKER_00: but people usually like to just...
[48:32] SPEAKER_00: you know, use a worse experience, but it was open source hackable.
[48:35] SPEAKER_00: And then out of seemingly nowhere,
[48:37] SPEAKER_00: VS Code came out,
[48:38] SPEAKER_00: which is,
[48:39] SPEAKER_00: it's free to use,
[48:40] SPEAKER_00: you know,
[48:40] SPEAKER_00: there's no like tricky things around it.
[48:43] SPEAKER_00: And it just felt,
[48:44] SPEAKER_00: again,
[48:45] SPEAKER_00: a bit incredible to me,
[48:46] because just knowing how Microsoft used to base their all
[48:50] developer business on pay
[48:51] SPEAKER_00: for quality tools,
[48:53] SPEAKER_00: it just seemed impossible.
[48:54] SPEAKER_00: How did that happen inside?
[48:55] SPEAKER_00: And what was the thinking?
[48:56] SPEAKER_00: And now, of course, most developers use VS Code or a fork of VS Code.
[49:00] SPEAKER_01: Yeah, so I'd been out of the developer division when I took over Azure.
[49:03] SPEAKER_01: So from 2011 to say 2014, I hadn't been in the developer division.
[49:10] SPEAKER_01: I'd been focused on Azure.
[49:11] And then when Satya took over as CEO in 2014,
[49:15] the same day I kind of took over his old job.
[49:18] SPEAKER_01: And that included DevDiv.
[49:19] SPEAKER_01: Yeah.
[49:21] SPEAKER_01: And so that kind of came back into my world.
[49:25] And I think one of the things that we did in the early days of
[49:29] 2014 was I kind of
[49:31] SPEAKER_01: just looked around and said,
[49:33] I think we have an opportunity to make a couple of choices that
[49:37] will be bold and
[49:38] SPEAKER_01: aggressive.
[49:40] SPEAKER_01: that give us a shot to rewind relevance with developers of the world.
[49:44] And if we don't,
[49:45] we're going to be on an iceberg that's going to slowly melt.
[49:48] SPEAKER_01: And at some point we're going to be swimming.
[49:51] SPEAKER_01: And so,
[49:51] SPEAKER_01: you know,
[49:52] SPEAKER_01: we kind of had a set of meetings in,
[49:55] SPEAKER_01: you know,
[49:55] SPEAKER_01: in spring of 2014 and kind of wrote on the whiteboard,
[49:59] SPEAKER_01: let's,
[49:59] SPEAKER_01: let's be bold and came up with kind of three big things that we said,
[50:03] SPEAKER_01: okay,
[50:04] SPEAKER_01: let's,
[50:06] SPEAKER_01: What can we do to become more relevant?
[50:09] The first one was we actually introduced a community edition of
[50:13] Visual Studio.
[50:14] SPEAKER_01: So previously,
[50:15] if you wanted to use Visual Studio and take advantage of the
[50:19] features that
[50:20] SPEAKER_01: developers loved,
[50:21] SPEAKER_01: You had to pay a minimum of about $1,000 to do it.
[50:24] SPEAKER_01: And we said, okay, let's make it free.
[50:26] SPEAKER_01: And it was scary for a lot of people at the time.
[50:30] SPEAKER_01: But we said,
[50:30] SPEAKER_01: look,
[50:31] SPEAKER_01: we can make it for small projects,
[50:32] SPEAKER_01: for startups,
[50:34] SPEAKER_01: for independent developers that want to build something.
[50:38] SPEAKER_01: And let's make the full feature set available.
[50:41] SPEAKER_01: And so that was decision number one.
[50:45] SPEAKER_01: And decision number two is let's open source.net.
[50:50] SPEAKER_01: And let's make it cross platform.
[50:51] SPEAKER_01: With Mono.
[50:52] SPEAKER_01: And it was less with Mono.
[50:54] SPEAKER_01: We actually took our base library.
[50:57] SPEAKER_00: Oh, sorry.
[50:57] SPEAKER_00: Mono was before.
[50:58] SPEAKER_00: Yeah.
[50:58] SPEAKER_00: Mono was before.
[50:59] SPEAKER_00: Yeah.
[51:00] SPEAKER_00: Because there's no alternative.
[51:01] SPEAKER_00: And now there's no need for Mono.
[51:03] SPEAKER_01: And then we said, let's open source it and do it right.
[51:06] SPEAKER_01: Not meaning just open source where we can contribute code.
[51:10] SPEAKER_01: But like, no, true open source.
[51:11] SPEAKER_01: We want the community to contribute code.
[51:14] SPEAKER_01: And do it under the right license and put it on GitHub.
[51:18] SPEAKER_01: And that was big decision number two.
[51:21] SPEAKER_01: And have a great Linux and Mac port and make them first class.
[51:27] SPEAKER_01: That was kind of decision number two.
[51:29] SPEAKER_01: And then decision number three was...
[51:31] SPEAKER_01: As much as we love Visual Studio,
[51:33] SPEAKER_01: the IDE,
[51:34] let's also recognize increasingly web developers and those that
[51:38] are not using a
[51:39] compiled language are looking for something that's much more of
[51:41] a lightweight code
[51:42] SPEAKER_01: optimized editor.
[51:44] SPEAKER_01: And there was a great project that had been started.
[51:49] It was basically a web-based editor written in Node,
[51:53] written in TypeScript.
[51:56] SPEAKER_01: But you ran it in a VS Online, I think is what we called it.
[52:00] SPEAKER_01: Oh, yeah, VS Online, yes.
[52:01] SPEAKER_01: And it was great technology, but...
[52:05] SPEAKER_01: you know,
[52:05] at the time people weren't really looking to write code in a
[52:08] browser and people
[52:09] SPEAKER_01: would say,
[52:09] SPEAKER_01: well,
[52:10] SPEAKER_01: it's a great,
[52:10] SPEAKER_01: it's great if I want to edit five or six lines of code,
[52:12] SPEAKER_01: but like,
[52:13] SPEAKER_01: it's not a,
[52:13] SPEAKER_01: you know,
[52:14] SPEAKER_01: it doesn't have a debugger.
[52:15] SPEAKER_01: It doesn't have Intel sense.
[52:16] SPEAKER_01: It doesn't, you know, it wouldn't scale for a large project.
[52:19] SPEAKER_01: And,
[52:19] SPEAKER_01: you know,
[52:20] SPEAKER_01: we said,
[52:20] SPEAKER_01: what,
[52:21] SPEAKER_01: what if,
[52:21] SPEAKER_01: you know,
[52:21] SPEAKER_01: I think I kind of said,
[52:22] SPEAKER_01: why don't you take this?
[52:23] SPEAKER_01: And is there a way that you could package it up in a,
[52:26] SPEAKER_01: you know,
[52:27] SPEAKER_01: Mac and windows and Linux version?
[52:30] SPEAKER_01: you know, shell and add file system support.
[52:33] And why don't we take this open source debugger we're about to
[52:35] do in .NET and port
[52:37] SPEAKER_01: it and make it work with it and really,
[52:40] SPEAKER_01: you know,
[52:41] don't have a project system was kind of one of the mantras and
[52:44] really focus on
[52:46] SPEAKER_01: streamlining editing and let's make it open source.
[52:51] SPEAKER_01: And the three different decisions we made,
[52:54] SPEAKER_01: and we kind of made all those decisions,
[52:56] SPEAKER_01: I think in about an hour and a half.
[52:58] SPEAKER_01: So it was a good meeting.
[52:59] SPEAKER_01: That was a good one.
[53:01] And honestly,
[53:02] if you asked like,
[53:03] what was the risk probability of each one succeeding?
[53:06] SPEAKER_01: I thought the first one,
[53:07] SPEAKER_01: which is making the tools free,
[53:09] SPEAKER_01: was gonna definitely drive more developer usage.
[53:12] SPEAKER_01: And whether it was gonna destroy the Visual Studio revenue, I don't know.
[53:14] SPEAKER_01: But I said like, let's take that risk.
[53:16] SPEAKER_01: I think the .NET being open source, I thought it would help.
[53:19] SPEAKER_01: And it certainly did.
[53:20] SPEAKER_01: And then the VS Code one was the most,
[53:24] SPEAKER_01: I would say,
[53:26] SPEAKER_01: speculative,
[53:27] SPEAKER_01: where I thought,
[53:27] SPEAKER_01: like,
[53:28] SPEAKER_01: I think this could help,
[53:30] SPEAKER_01: but I don't know.
[53:31] SPEAKER_01: There's a lot of other editors out there.
[53:32] There was Atom,
[53:33] there was Sublime,
[53:34] there was,
[53:36] gosh,
[53:36] a couple other ones that were only Mac.
[53:38] SPEAKER_00: And then later new ones are like NeoVim, etc. ?
[53:40] SPEAKER_01: Yeah,
[53:40] SPEAKER_01: and so,
[53:41] SPEAKER_01: you know,
[53:41] SPEAKER_01: it was,
[53:41] SPEAKER_01: but it was,
[53:42] SPEAKER_01: you know,
[53:43] SPEAKER_01: of the three decisions,
[53:44] SPEAKER_01: they were all very big.
[53:45] I think the ones that we remember the most would be VS Code and
[53:48] then the open sourcing of .NET.
[53:49] SPEAKER_01: And ironically,
[53:50] the one that I thought would have the biggest impact was
[53:52] probably the one that had
[53:52] SPEAKER_01: the least impact,
[53:53] SPEAKER_01: although it helped a tremendous amount.
[53:55] SPEAKER_00: Yeah, it comes to tell you, you can never know even from the inside.
[53:59] But,
[53:59] you know,
[54:00] basically,
[54:01] I think later in that year,
[54:02] we kind of launched,
[54:03] announced all three.
[54:03] SPEAKER_01: And,
[54:04] SPEAKER_01: you know,
[54:05] SPEAKER_01: that really,
[54:05] SPEAKER_01: I think,
[54:05] SPEAKER_01: helped us with our next kind of,
[54:07] SPEAKER_01: I'll call it rejuvenation with the developer community.
[54:10] SPEAKER_01: And in some ways also the VS Code,
[54:13] SPEAKER_01: I remember doing events in 2015,
[54:17] SPEAKER_01: 2016 with a lot of the GitHub team,
[54:21] SPEAKER_01: which was an independent company.
[54:23] SPEAKER_01: And VS Code also was that bridge that I think earned us the credibility
[54:28] SPEAKER_01: to talk to GitHub and to be sort of in the developer community at large.
[54:35] SPEAKER_01: And,
[54:36] SPEAKER_01: um,
[54:36] SPEAKER_01: you know,
[54:36] SPEAKER_01: it was,
[54:37] SPEAKER_01: was,
[54:37] was the bridge that kind of started the conversation around
[54:40] ultimately GitHub
[54:41] SPEAKER_01: becoming part of Microsoft,
[54:43] SPEAKER_01: um,
[54:43] SPEAKER_01: which,
[54:43] SPEAKER_01: you know,
[54:43] SPEAKER_01: if you,
[54:44] SPEAKER_01: if you went back to 2010,
[54:45] SPEAKER_00: it was impossible.
[54:48] SPEAKER_01: It would be the most crazy solution suggestion ever.
[54:51] SPEAKER_01: And then frankly,
[54:52] I think it beyond the fact that the GitHub team wouldn't want
[54:54] to have been bought.
[54:56] SPEAKER_01: Um,
[54:56] or be part of Microsoft,
[54:57] I think the developer community would have left immediately.
[55:00] And,
[55:01] and yet we,
[55:02] you know,
[55:02] fast forward to 2017,
[55:04] I think was when we did the acquisition.
[55:05] SPEAKER_01: People still had some concern,
[55:07] SPEAKER_01: but they said,
[55:07] SPEAKER_01: no,
[55:07] SPEAKER_01: let's tell you what,
[55:08] SPEAKER_01: you've earned the right,
[55:09] SPEAKER_01: we're going to give you a chance.
[55:10] SPEAKER_00: Yeah, there was a lot of skepticism as well.
[55:13] SPEAKER_00: I agree.
[55:14] SPEAKER_00: But yeah,
[55:15] SPEAKER_00: and then now it brings us today actually to build where Copilot,
[55:19] SPEAKER_00: built by the GitHub team,
[55:20] we know the AI system with a bunch of advanced capabilities is
[55:23] now open source as
[55:25] SPEAKER_00: well,
[55:25] SPEAKER_00: together with VS Code.
[55:28] SPEAKER_00: So now that we're here,
[55:30] SPEAKER_00: just looking ahead,
[55:31] SPEAKER_00: we are where we are with lots of exciting new tools,
[55:36] SPEAKER_00: lots of new ways to work,
[55:37] SPEAKER_00: and I think we're going to figure out.
[55:39] SPEAKER_00: Agents are the hot thing.
[55:41] Personally,
[55:41] what are you excited about in terms of developer tools and also
[55:45] cloud?
[55:47] SPEAKER_00: When you look at,
[55:48] SPEAKER_00: may these be projects,
[55:49] SPEAKER_00: may these be directions or things that need to be figured out?
[55:52] SPEAKER_01: I think the thing I get excited most about is,
[55:55] SPEAKER_01: certainly on the developer tool side,
[55:56] SPEAKER_01: I do think this notion of having an agent that works with us,
[56:00] SPEAKER_01: and that's some of the demos we showed today and yesterday of,
[56:03] SPEAKER_01: how can you assign an issue in GitHub to your co-pilot?
[56:09] SPEAKER_01: I think a lot of us have used,
[56:11] SPEAKER_01: whether it's GitHub co-pilot,
[56:12] SPEAKER_01: whether it's ChatGPT or N35 co-pilot or cursor,
[56:17] SPEAKER_01: people are used to kind of a request response
[56:20] model where you type something,
[56:22] you get a response immediately,
[56:23] and that's super powerful.
[56:24] SPEAKER_01: So that's not going away.
[56:26] But this notion as the models get richer that you can just
[56:29] assign a task to
[56:31] SPEAKER_01: effectively a coworker that is AI that can do something over maybe 10,
[56:37] SPEAKER_01: 15,
[56:38] SPEAKER_01: and then assign it back to you,
[56:39] SPEAKER_01: I think that is super profound.
[56:41] 20 minutes,
[56:41] And so the ability to say,
[56:42] hey,
[56:43] take this Figma or take this screenshot that I've sketched out,
[56:48] SPEAKER_01: you know,
[56:49] SPEAKER_01: turn it into a nice HTML with CSS or,
[56:52] SPEAKER_01: you know,
[56:53] SPEAKER_01: create for me,
[56:55] SPEAKER_01: you know,
[56:55] a Kubernetes deployment file and a basic microservice
[57:00] architecture that has these
[57:01] five dimensions and is connected to these three resources and
[57:04] make it scalable and
[57:06] SPEAKER_01: secure,
[57:07] SPEAKER_01: like assign it to the co-pilot,
[57:09] SPEAKER_01: grab a coffee,
[57:10] SPEAKER_01: come back and get something back.
[57:12] SPEAKER_01: Um,
[57:13] SPEAKER_01: Yeah, that I think is really profound.
[57:15] SPEAKER_01: And then how do you extend that so that in addition to development,
[57:19] SPEAKER_01: a lot of kind of what you might think of as the operation tasks or SRE,
[57:24] SPEAKER_01: things like,
[57:24] SPEAKER_01: hey,
[57:25] SPEAKER_01: is there anomaly detection in my logs?
[57:27] SPEAKER_01: You know, is the performance dropping?
[57:29] SPEAKER_01: Why?
[57:31] SPEAKER_01: You know, can you ask it, what changed in the environment?
[57:33] SPEAKER_01: What changed in the code base?
[57:35] SPEAKER_01: What's changed in terms of the user behavior?
[57:37] You know,
[57:37] how do you have an AI agent that can kind of assist you in
[57:39] that?
[57:40] SPEAKER_01: I think that's really about kind of giving every developer superpowers.
[57:46] SPEAKER_01: And, you know, it's kind of like the Marvel Iron Man.
[57:49] You know,
[57:49] suddenly you have a suit that gives you kind of these amazing
[57:53] superpowers.
[57:54] You know,
[57:54] I think to some extent,
[57:55] these types of copilots are going to do that.
[57:56] SPEAKER_01: And then when you combine it with...
[57:59] SPEAKER_01: a cloud like Azure,
[58:02] where if you've got a great idea and you're going into business
[58:07] as a startup or
[58:08] SPEAKER_01: whether you're a big company,
[58:10] SPEAKER_01: the ability to run it in 70 regions around the world.
[58:15] SPEAKER_01: Certainly right now,
[58:15] SPEAKER_01: when you think about whether it's tariffs or whether it's geopolitical,
[58:20] SPEAKER_01: just like,
[58:21] SPEAKER_00: okay,
[58:21] SPEAKER_01: how do I meet every country's local residency requirements on data?
[58:26] SPEAKER_01: Cloud is the way to do that.
[58:28] SPEAKER_01: Similarly,
[58:28] SPEAKER_01: if you think about how do you target the growing markets in the world,
[58:35] SPEAKER_01: cloud is the way you do that.
[58:36] SPEAKER_01: And then when you start thinking about how do I actually build my own AI
[58:40] SPEAKER_01: application and take advantage of the latest models?
[58:43] SPEAKER_01: How do I use diffusion or fine-tune my own model that's based on,
[58:50] SPEAKER_01: say,
[58:50] SPEAKER_01: my own data or do post-training?
[58:52] SPEAKER_01: You know, the cloud's going to be the way you do that.
[58:54] SPEAKER_01: And so I get excited in terms of when you take these sort of tools,
[58:58] SPEAKER_01: you know,
[58:58] SPEAKER_01: tying it back to the beginning of the conversation,
[59:00] SPEAKER_01: going back to MFC and Visual Basic.
[59:03] SPEAKER_01: You know,
[59:03] SPEAKER_01: I think this is the same level of kind of exponential jump,
[59:07] but probably even more profound in terms of the impact because
[59:11] it's not just
[59:11] SPEAKER_01: development,
[59:12] SPEAKER_01: it's also runtime.
[59:13] SPEAKER_01: And then it's that feedback loop of, okay, now I get something out there.
[59:16] SPEAKER_01: I'm learning from my user behavior.
[59:18] SPEAKER_01: Let me improve it.
[59:19] SPEAKER_01: And,
[59:20] SPEAKER_01: um,
[59:20] SPEAKER_01: that's the thing that always motivates me is,
[59:22] SPEAKER_01: is can you ultimately make someone more successful?
[59:27] SPEAKER_01: And,
[59:27] SPEAKER_01: um,
[59:28] if you can allow them to take an idea and run with it faster
[59:33] and do it faster,
[59:35] SPEAKER_01: better,
[59:35] SPEAKER_01: cheaper,
[59:35] SPEAKER_01: you know,
[59:36] SPEAKER_01: you can change their life and you can bring to life this great idea.
[59:39] SPEAKER_01: And if you can do that at scale, it's, it's a fun journey.
[59:45] SPEAKER_00: hearing about like AI colleagues can be a little scary for as engineers.
[59:51] SPEAKER_00: I think it's a similarly big shift as back,
[59:54] SPEAKER_00: if you think back in the nineties,
[59:55] SPEAKER_00: oh,
[59:55] anyone can program with like dragging and dropping and doing
[59:58] similar stuff that I'm
[59:59] SPEAKER_00: doing right now.
[1:00:00] SPEAKER_00: What would your advice be for
[1:00:02] software engineer today like you know mid-level or experienced
[1:00:04] engineers who want
[1:00:05] to be the you know standout engineers of tomorrow on on how to
[1:00:10] approach this and
[1:00:10] how to think about the fact that now they can a lot of the work
[1:00:13] that they have been
[1:00:14] valuable for until now can potentially be offloaded it's a
[1:00:17] little bit you see what
[1:00:18] i mean it kind of messes with your mind a bit it does i mean i
[1:00:21] think the thing i'd
[1:00:21] probably say to encourage is like i'm a big student of history
[1:00:25] like technology
[1:00:26] SPEAKER_01: history i've been part of and i was studying sometimes but
[1:00:28] I also like to look back in the last 500 years of history and
[1:00:32] read lots of biographies.
[1:00:33] SPEAKER_01: History is a way of repeating itself.
[1:00:38] SPEAKER_01: Just going back to the development community for a while,
[1:00:41] SPEAKER_01: I remember in the early 2000s,
[1:00:46] people would come to developer conferences and they would
[1:00:48] literally have t-shirts
[1:00:49] SPEAKER_01: that said,
[1:00:50] SPEAKER_01: IntelliSense rots the mind.
[1:00:54] SPEAKER_00: Oh.
[1:00:54] SPEAKER_01: Because like no real developer.
[1:00:56] SPEAKER_00: And it sounds like it was a really good autocomplete.
[1:00:59] Yeah,
[1:00:59] like no real developer would use autocomplete was the statement
[1:01:02] at the time.
[1:01:04] SPEAKER_01: Real developers knew the name of the method and the parameters.
[1:01:07] SPEAKER_00: And the classes, yeah.
[1:01:08] SPEAKER_01: And similarly, I think type safety, when .NET was introduced,
[1:01:13] SPEAKER_01: Siebel's host developers were like, type safety is for wimps.
[1:01:17] SPEAKER_01: All you need is a void star pointer.
[1:01:20] SPEAKER_01: And if you got it wrong, you'd crash.
[1:01:23] SPEAKER_01: But who needs type safety?
[1:01:25] SPEAKER_01: And similarly, I think even when JavaScript emerged in HTML5,
[1:01:31] SPEAKER_01: you know,
[1:01:31] SPEAKER_01: people would say like,
[1:01:32] SPEAKER_01: well,
[1:01:32] SPEAKER_01: you can't really have a real app with HTML and JavaScript.
[1:01:35] SPEAKER_00: We had the thing, it's not a real language JavaScript.
[1:01:37] SPEAKER_00: I remember this one.
[1:01:38] SPEAKER_01: And so,
[1:01:38] SPEAKER_01: you know,
[1:01:38] SPEAKER_01: all these parallels,
[1:01:40] SPEAKER_01: I kind of,
[1:01:40] SPEAKER_01: and then going back to,
[1:01:41] SPEAKER_01: you know,
[1:01:42] SPEAKER_01: the days of debuggers,
[1:01:43] SPEAKER_01: you know,
[1:01:44] SPEAKER_01: real developers don't use debuggers or profilers.
[1:01:48] SPEAKER_01: And so it's,
[1:01:49] SPEAKER_01: you know,
[1:01:49] SPEAKER_01: I've heard these things before where people say,
[1:01:52] SPEAKER_01: you know,
[1:01:53] SPEAKER_01: you can't,
[1:01:54] SPEAKER_01: and certainly with cloud,
[1:01:55] SPEAKER_01: I've certainly heard lots of people say,
[1:01:56] SPEAKER_01: well,
[1:01:57] SPEAKER_01: I can run my private cloud cheaper and faster and better than you can.
[1:02:00] It's like,
[1:02:01] okay,
[1:02:02] most of those companies are not in business anymore,
[1:02:04] but sure,
[1:02:06] go for it.
[1:02:07] And so I do think that would be my only kind of history lesson
[1:02:12] I'd pass on,
[1:02:12] which is generally these things are frightening sometimes
[1:02:16] because they do make it
[1:02:17] SPEAKER_01: look like,
[1:02:18] SPEAKER_01: wow,
[1:02:19] SPEAKER_01: could this automate me?
[1:02:20] SPEAKER_01: Could this replace my job?
[1:02:21] SPEAKER_01: But ultimately, if you embrace the productivity,
[1:02:25] you suddenly discover you can do a lot more and you can deliver
[1:02:29] a lot more.
[1:02:30] SPEAKER_01: And that is really what makes you valuable.
[1:02:33] It's not whether or not you know the name of the method and are
[1:02:36] proud that you
[1:02:37] SPEAKER_01: remember the 18 parameters to it.
[1:02:39] SPEAKER_01: It's, you know, what does the method do?
[1:02:42] SPEAKER_01: And does the application calling the method accomplish some value?
[1:02:46] And so to the extent that we can leverage AI to kind of make
[1:02:49] developers more
[1:02:49] SPEAKER_01: successful,
[1:02:51] SPEAKER_01: I think we,
[1:02:52] SPEAKER_01: in the limit,
[1:02:53] SPEAKER_01: it's going to create more jobs,
[1:02:54] SPEAKER_01: not less.
[1:02:55] And I think ultimately it means those jobs are going to be
[1:02:57] higher paying because
[1:02:58] the business value you're delivering to the company or the
[1:03:02] startup or your app that
[1:03:05] SPEAKER_01: you're building becomes even greater.
[1:03:07] SPEAKER_01: So it's not always a perfectly linear progression,
[1:03:12] SPEAKER_01: but I think that's,
[1:03:13] SPEAKER_01: if I look again back in history,
[1:03:16] SPEAKER_01: garbage collection,
[1:03:17] SPEAKER_01: tele-sense,
[1:03:17] SPEAKER_01: debuggers,
[1:03:19] SPEAKER_01: type safety,
[1:03:20] SPEAKER_01: even open source,
[1:03:22] SPEAKER_01: those are all things that were proven right
[1:03:25] SPEAKER_01: And the naysayers were proven wrong over time.
[1:03:28] SPEAKER_00: Yeah,
[1:03:28] SPEAKER_00: and I guess it's just good to remind that open source,
[1:03:30] SPEAKER_00: I think from the outside,
[1:03:31] SPEAKER_00: we would not think it's scary.
[1:03:32] SPEAKER_00: But from the inside, like Microsoft, it was.
[1:03:34] SPEAKER_00: And it turned out pretty good.
[1:03:35] SPEAKER_00: It worked out pretty well.
[1:03:36] SPEAKER_00: So thank you very much, Scott.
[1:03:38] SPEAKER_00: This was so much fun.
[1:03:39] SPEAKER_00: This was a lot of fun.
[1:03:40] SPEAKER_00: Thanks so much for having me.
[1:03:41] Thanks to Scott for this trip down memory lane and for his
[1:03:44] advice on how software
[1:03:45] engineers can keep growing in an industry that changes as fast
[1:03:48] as software
[1:03:49] SPEAKER_00: engineering does.
[1:03:50] SPEAKER_00: For more details about Microsoft's developer tools evolution,
[1:03:53] check out our deep dive into Pragmatic Engineer linked in the
[1:03:55] show notes below.
[1:03:56] SPEAKER_00: If you've enjoyed this podcast,
[1:03:57] SPEAKER_00: please do subscribe on your favorite podcast platform and on YouTube.
[1:04:01] SPEAKER_00: A special thank you if you also leave a rating on the show.
[1:04:04] SPEAKER_00: Thanks and see you in the next one.

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.