text stringlengths 0 1.49k |
|---|
**Carlisia Thompson:** And what city are you in? |
**Brian Ketelsen:** Milan. |
**Carlisia Thompson:** Milan, nice. |
**Brian Ketelsen:** It's beautiful. Right next to my hotel is this building that they have all lit up with pretty lights, and stuff... So I'm standing outside, looking at this building, trying to figure out what it is. I walked around the block, and there's a sign out in front in Italian and English that says "This use... |
**Erik St. Martin:** Wow. |
**Brian Ketelsen:** We have such a short-term view of things in the United States, where the oldest buildings that we see are just a couple hundred years old, and that's only in the rarest cases... Generally, everything's less than 100 years old. It's just crazy. |
**Erik St. Martin:** So Jeff, do you wanna give maybe a little bit of background about yourself and some of the things you're working on? |
**Jeff Lindsay:** Yes... It can be difficult to do that, but... |
**Brian Ketelsen:** Uh-oh... He could tell us, but then he'd have to kill us. |
**Jeff Lindsay:** Yeah, especially -- so I'm doing a lot of open source stuff... Actually, for the past few years I've been doing a lot of work in kind of [Docker](https://www.docker.com/) or distributed systems and platform, ecosystem stuff and [Kubernetes](https://kubernetes.io/), and I'm really kind of pushing out a... |
**Carlisia Thompson:** Explain, please... \[unintelligible 00:03:20.10\] We'll have to go back to that. |
**Brian Ketelsen:** So do you wanna talk about some of the things you've done in the past, rather than expose this secret thing that you're working on? Like [Flynn](https://flynn.io/) or [Dokku](https://github.com/dokku/dokku), or my personal favorite, [Envy](https://github.com/progrium/envy)... |
**Jeff Lindsay:** Which one? |
**Brian Ketelsen:** Envy. |
**Jeff Lindsay:** Oh, you saw that...? |
**Brian Ketelsen:** I saw it. I had the most fun with that application. I actually used it as a training platform. |
**Jeff Lindsay:** \[03:51\] Oh, cool. Yeah, that one was interesting. I like trying to do kind of mind-bending things, and that was kind of mind-bending... But then people were also really interested in it. I didn't really take it all the way -- it kind of comes back into some of my bigger plans, but basically, everyth... |
So I worked on Flynn a little bit in the beginning, Dokku - that was basically the first killer app for Docker; it's like a Heroku implementation in 100 lines of Bash. It's now more than 100 lines, but it's very cool, because it's very hackable. There's tons of stuff in between, stuff like Envy, that people don't know ... |
I worked on [Docker](https://github.com/docker) at the beginning, and... What did I do before Docker? I guess I worked at Twilio, and I mostly learned -- it was great working at [Twilio](https://github.com/twilio), being employee ten at Twilio, but it was my least productive in terms of open source output, so that was ... |
Then, let's see... Before that I've worked on a lot of weird projects. Webhooks is a weird one; it doesn't really -- it's kind of a weird pattern... It's a weird thing to call a project. Some people say I invented webhooks, but I don't think you can invent webhooks. I popularized it, I coined it, I really pushed it to ... |
I did a startup way back when I was 22 or something called DevjaVu, and it was \[unintelligible 00:06:16.27\] this was right when GitHub started, and they were doing such a good job. |
**Erik St. Martin:** Track... I haven't seen that in ages... |
**Jeff Lindsay:** One of those things... Yeah. I don't use Track, but I learned a lot of really cool tricks from it and I actually bring some of those up today, architecturally. But I made a product called DevjaVu around Track, and then shut it down because I was like "Nah, GitHub's killing it. I don't wanna waste my t... |
A lot of people, like [Solomon Hykes](https://twitter.com/solomonstre) at Docker - he was a user of DevjaVu back in the day; it's kind of how we knew each other. A lot of people started using DevjaVu before they were like "Now, GitHub!" |
Before then, I did [SuperHappyDevHouse](https://en.wikipedia.org/wiki/SuperHappyDevHouse), which was one of the first major hackathons in the Silicon Valley, and did that for five years. That was incredible... I can't even capture how amazing those events were. Then Facebook copied SuperHappyDevHouse... |
But most of this time I've been freelancing or just building kind of open stuff, or weird ventures... I don't know. A bunch of other stuff. |
**Erik St. Martin:** So you've only done "just a couple things." \[laughs\] |
**Jeff Lindsay:** I made a thing called localtunnel... |
**Brian Ketelsen:** The precursor to [ngrok](https://github.com/inconshreveable/ngrok). |
**Jeff Lindsay:** Yeah. And localtunnel was kind of the inspiration for trying to find a sustainable model for web apps... The open source equivalent for services, because there is -- if you wanna build something and run it on the internet, there's really no way to do that sustainably without building a business, and I... |
\[08:08\] It also isn't ideal for people that like to just create things and put them out there, because now you have to dedicate your life to making a business work around it. There's plenty of things that have value that you can't necessarily build a business around. So I've been kind of working on infrastructure to ... |
[Commando.io](https://commando.io/) is a recent project that's not out of this kind of private alpha stage, but that's sort of a recent project... I bring it up because I was gonna try and validate this idea of "Can we build a service that's actually -- it looks like a startup, but it's actually a sustainable self-runn... |
So many great things that have been built, like Parse and some other stuff... It's like, the business aspect of software can be really harsh sometimes, so I spend a lot of time thinking about that. |
I also think a lot about making -- just in general, I think this is me trying to encourage everybody to be more of a hacker/builder type. I really wanna make more of the world, and even just the computer systems that we have, more programmable, more extensible, more scriptable, more customizable. They've become really ... |
And even as a professional programmer, there's this idea -- this is kind of ideal, that we have this great kind of imagination compiler, and you kind of come up with an analogy like Legos; you can just kind of like compose things together and throw them together, and that's not really how it works, most of the time, un... |
So that's really unfortunate, and I spend a lot of time thinking about that and trying to make better building blocks. That's a common theme of a lot of my work. |
**Carlisia Thompson:** Do you ever think of using artificial intelligence? |
**Jeff Lindsay:** You know, a lot of other people are thinking about artificial intelligence, and there's so much value you can get out of not using artificial intelligence... Just basic automation techniques. It's like, there's so many simple things that it's like oh, you could imagine modeling that in a few lines of ... |
**Brian Ketelsen:** \[12:14\] That's a good way to put it... More of a tool than a full solution. I've been thinking about the idea of the composable software recently. I found - and I promise this isn't an advertisement - Microsoft Flow, and I had never seen it before, or anything really like it. It's just drag-and-dr... |
Every time you save an Excel spreadsheet, it can read the value from column seven and send a tweet about it. It's really cool automation, and I'm kind of surprised that it hasn't caught on more. I had never heard about it until I started at Microsoft. That's pretty easy stuff for your average data worker to just go use... |
**Jeff Lindsay:** Yeah. There's some great -- I kind of think of them as walled garden experiments... I love any kind of experiment in things to facilitate end user programming, things that make it easier for not necessarily programmers to easily program things, either visually or whatever. But the problem I have with ... |
Some friends and I were working on this project that we're calling [WebPipes] (https://github.com/webpipes); it's kind of like formalizing some webhook stuff to try and come up with an open source spec for building "if this, then that" type systems. And it was layered in such a way that you're not just creating these... |
**Brian Ketelsen:** Nice. |
**Jeff Lindsay:** A lot of the stuff that I've worked on - Docker, things like this - were pieces to something else that I wanted to build. |
The way I think of things is -- and I have frustrations with this, because this is the conventional way of thinking in software, and actually most fields is very relative to the current state of things. So it's like you're trying to solve immediate problems, when it actually goes a really long way to step back and say ... |
I have all these ideals of a world where everybody is empowered to build things very easily, and tap the full potential of computing and automation and all of this, and it requires a lot of infrastructure to get there, especially if it's something that is more than just a walled garden; it needs to be built up from oth... |
\[16:12\] When I first found [AppEngine](https://en.wikipedia.org/wiki/Google_App_Engine), I was in love. I was like, "This is amazing. I can build applications and hit a button and not worry about the operations in some ways", so that lead to this sort of AppEngine and [Heroku](https://en.wikipedia.org/wiki/Heroku). W... |
So then that building block was out there, but then you had so many other things you needed to develop for building these kinds of distributed systems, or a system that you could build a good Heroku or AppEngine on top of, which a lot of people started, you know, the floodgates opened and everybody started building stu... |
Actually, for a long time I had all the components, like how I would build Kubernetes, but I really hesitated from building them because I knew so many other people were building them, and it would have been -- I'm not getting paid to do it, so it's like I don't wanna waste this precious commodity, when I could be buil... |
It's frustrating though to know that you want something in a particular way, and then it doesn't happen exactly as you want... But as long as you're kind of focusing on the bigger picture, it helps. |
**Erik St. Martin:** So what's your opinion on where we're going with, say, Kubernetes? Because you're interested in distributed systems stuff and solving things at a higher level, on abstractions and building blocks... How we're getting to the point where we're trying to make the running of Kubernetes, or getting it r... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.