This is a transcript. For the video, see An interview with Josh Koenig.
[00:00:00] Michael Meyers: Hello. Welcome to Tag1 Team Talks, the blog and podcast of Tag1 Consulting. To commemorate the 20th anniversary of Drupal, Wwe're doing an interview series with community leaders to talk about their Drupal journeys, how Drupal has transformed their personal and professional lives, what Drupal means to them and their thoughts on the future of the platform and the community. I'm Michael Myers and managing director of Tag1, and I'm honored and excited to welcome Josh Koenig, the co-founder and chief strategy officer of Pantheon.
Welcome Josh. Thank you so much for joining me today.
[00:00:34] Josh Koenig: Hey, thanks for having me, Michael. I'm excited about this conversation.
[00:00:39] Michael Meyers: So before we get started, I love origin stories. I'm super curious because so many people that I know, including myself, studied one thing and then did another. You have a background, you have a BFA in drama from NYU.
I'm really curious. How did you go from that immediately into technology and did your background in [00:01:00] drama help you to this day?
Josh Koenig: That's a great question. And a decent story. And the short answer is yes, it has. But,the way it worked for me was I was, I was a computer kid literally as like a, a kid with my like Commodore Vic 20, loading up the basic prompt and like making like, little things happen with that.
But I was all like, self-taught, right. I- I had that Commodore, and then like, my mother was a graphic designer, so she had a Macintosh in the house and I would like, mess around with that and wanted to write video games, like I think most teenagers do. And, and, I'd gotten pretty good at technology, was lucky enough in my high school at the time in the late nineties to actually get introduced to a bunch of stuff around like early web.
[00:01:44] Josh Koenig: And we built a web version of the high school yearbook, at one point, which was great early experience. So like I had got an interest in technology, but very auto-didactical. And at the same time, I was also very interested in arts and theater and writing and other [00:02:00] stuff like that. And I had a, of a friend of my parents, pseudo mentor, who, he was, had the career as a shareware author.
This is like going way back in the world of community. There used to be this thing called shareware before there was the Internet where it was like, okay, it's fine to pirate or distribute copies of the software, but just, if you do, send me a check, please just send me 15 bucks. And he wrote like a couple like graphics programs and lived a comfortable life as a result of his shareware.
And, and what he said to me was, Josh, you're probably not going to like computer science school and a university. It's just not your style. Um, and, at the time, especially again, this is like 96, 97, this was, you would go and it would be like, let's go back and teach you like Fortran or COBOL or maybe like intro to like Java had only been just introduced and was not like standardizing the curriculum.
And he said, what I advise you to do is go to school for whatever you're most passionate about, but get a job, go to school for what your heart most desires, but find [00:03:00] work in technology. You won't have a problem finding a job. If you hustle a little bit, you're smart. You'll figure it out.
[00:03:05] Josh Koenig: In retrospect, that's really high risk advice. But it worked out well for me. I, I got into this conservatory program at NYU, which I didn't think, I was a little bit surprised. I surprised myself. Because it's a competitive program to study theater, which I was very much into.
And then I started working, started, at least it was a freshman. I started working in the computer lab just cause I was like first thing, but via people I met in the computer lab, I found other, slightly better paying, much more interesting work. And, essentially while I was getting my BFA in theater, like doing like literal singing, dancing, like movement classes, and, and Shakespeare and other fun stuff, I was essentially getting a journeyman apprentice course in how to build web applications, like working with Sun Solaris, to stick together a Postgres database and a Perl front end to run very rudimentary online applications like, a [00:04:00] calendar syncing application, a couple other things like that. And it was like, good technical experience.
[00:04:04] Josh Koenig: It was like the heady days of Silicon Valley. So I also got to see some business practices that were unorthodox, and learned a little bit from that too, I think. but, what kind of happened for me was that I graduated, and by the time I was in my final year of Theater school, I realized that I was not going to be able to pay my rent as an actor.
I've paid exactly two months rent with acting, which I am proud of, but like that, it's really hard. It's an incredibly difficult thing to do to make a career out of being an actor, because you have to just relentlessly audition for stuff that you're not that interested in and be super stoked when you get five seconds in that 7/11 commercial, because that means you can pay your rent for a few months and maybe you'll get an agent.
[00:04:45] Josh Koenig: And it just wasn't for me, I wasn't going to be able to. I just didn't have it in me. And I realized that. And so I kept doing art for fun and just doubled down on tech for work, mostly freelancing, early on, because I like, I [00:05:00] think my mom was like a little bit of a role model for that. She was a freelance graphic designer for a bunch of her career.
And, oh, so as a result of freelancing and working with a bunch of interesting people and people working with, you know, kind of following that self-taught path of okay, I'll take on a new job. That's going to stretch me a little bit, but I'll learn it as I go. And, by this time you could go online and use the web to learn about pretty much anything you want to do.So it all just worked for me over a few years to just kind of like, build up my technical chops.
[00:05:21] Michael Meyers: Well, business is a lot about storytelling. I know you give a lot of presentations, both to investors and audiences. And so I'm sure that drama background has helped you tremendously and in many aspects of your professional life, even if you didn't go into acting.
[00:05:44] Josh Koenig: Oh yeah. Sorry. I skipped over that part of the question. It's a huge, it's a huge, it's like a superpower, right? I like got trained in a really rigorous environment for several years to be absolutely comfortable walking up in front of hundreds of people, kinda knowing what I'm doing and just riffing.
And like the idea of, you're going to present at a conference in a big room, there's a, there's some nervousness that like anything, any human gets a little bit of butterflies in their stomach. But it doesn't ruffle me at all. I'm not worried about it. In fact, I really enjoy it.
[00:06:13] Josh Koenig: I like to get the opportunity to put together a really good presentation and deliver it really well and get people in the audience engaged. Like that's one of the, that's a very rewarding feeling. And, I, there's a part of me has, like wishes. I could actually do more of that in my career because it is personally really satisfying to deliver a great talk in the same way that it's very, it was very satisfying for me, to put on a great show, in an actual theater.
But yeah, it's like it's a little bit of a Superpower and, and it's definitely helped me. I'll- I wouldn't, I don't think I would be where I am today. If I didn't have that's
[00:06:44] Michael Meyers: That's an awesome way to look at it, Superpower it's mean, risky advice or not, it sounds like it is a great way to go.
I love it.
[00:06:51] Josh Koenig: Yeah. I just, I like to be cognizant of like, survivor bias and all these things. Like I, I have a lot of things going for me,just, coming out of the gate. and [00:07:00] I was lucky in a lot of ways and, I also worked very hard, so that's a huge part of it, but, but yeah, it, I don't know whether the path that worked for me is advisable for everyone else, but it did work for me.
[00:07:11] Michael Meyers: Well, you've been a member of the Drupal community now for just over 18 years. Your UID is 3,313. The, the early UIDs, there's a lot of blank space in there. So that makes you among the first hundred, a few hundred maybe thousand of people to discover, use and create a drupal.org account. I'm curious how you found it, how did you discover Drupal so early?
[00:07:38] Josh Koenig: Yeah, this is a good story. I like to say, I got into Drupal because, we were, I was part of, I was trying to take over the government. We, being a technology freelancer living in New York City in the early aughts. I, it was a really interesting and unique time to be a New Yorker for sure.
And, and I've always been politically active, since [00:08:00] my, since my youth and I just didn't want like the, I didn't want the war to happen. The Iraq war I was. It's a really big mistake. This is really bad idea. This is not going to end well. we're doing this for all, a bunch of wrong reasons.
[00:08:11] Josh Koenig: This is a transparently, a propaganda operation. This is not like it's not the way it should be. And so I got involved in the, all the protests, organizing that was going on in and around New York City. And they had New York City had some of the largest anti-war protests in 2002 and early 2003, anywhere in the world.
And that was a lot of work. I met a lot of interesting people, and a lot had a lot of interesting experiences, but ultimately those protests were complete failure. They did nothing that absolutely no effect. It just the President was like, I don't take foreign policy advice from a focus group, and just said, go fuck yourself.
And that made me rethink like, okay, this is - the stakes here are pretty high. This doesn't work. What might work well, you need to change who the President is. That means I need to start caring about the election cycle. And I got really into this, at the time, like a dark horse candidate named Howard Dean who was [00:09:00] using the Internet for the, he wasn't the first candidate to use the Internet, but he was the first national Presidential candidate in the U.S. to use the Internet effectively. And what he did was, he was a Governor of Vermont. No, no national name recognition was like at 1% in the polls, but he was one of two, people in the Democratic party that was actually willing to say, we should, this war is, and by the way, we should have universal healthcare and a few other things that like, we're still working on now, to be honest.
But he was out front and loud and proud about that and got, and at the time, I don't know if you recall, like at the time you couldn't get that message out in the traditional media, It was almost like, it was almost like forbidden to, to speak out against this stuff in any, on national news.
[00:09:40] Josh Koenig: And so he used the Internet really effectively to get that message out. And I received that message via the Internet, and I got really excited about it. And I had this thought process, which was like, I know how to make the Internet. Here's a candidate who's using the Internet. He's got the message that I believe in.
I, instead of doing [00:10:00] something really, um, you know, good volunteer work, but pedestrian or commodity, like gathering signatures or like making protest signs, like I've got skills that are hard, that are rare and valuable, that could be applied to this really important project.
I should get into this. And so I got really into this campaign and, this is how I met my, my business partner of almost 20 years now, Zach, because he did the same thing. Zack had set up a website. Yeah, it was called Hack4Dean with a four. The, and there was this like on his summer vacation from college and it was just like a mailing list with a bunch of, you know, total rando, wild, like people, joined this thing, like one of the original spec authors for XML was in there.
Like Cory Doctorow was like emailing in and it was like this whole thing of like, we got to do something and nobody knew what to do. and so we, what we arrived on after like about six weeks. Just incredibly intense, like mailing lists and IRC chatter was we should create campaign in a box [00:11:00] software for all of the affinity groups that are springing up around this campaign because we started to get traction, starting to move up in the polls, starting to raise money online effectively, which everybody then had to pay attention to.
And then you can spend more on advertising. You raise your name brand recognition. It was like there was a flywheel working for this campaign and it was a lot of grassroots stuff. It was a lot of like kind of bottoms up organizing. And so he said, all there's all these affinity groups. They all need websites because that's how we're getting the message out.
So we wanted to create campaign-in-a-box software for all of the volunteer grassroots affinity groups around the campaign. And the, the internal thing was, we obviously can't write this from scratch. We don't have the time. Let's look through the world of open source and see what could be a fit to start as a place to build.
And,the story goes, I actually wasn't here for that. But the story goes that, Neil Drumm, who was part of this project as well, went into the # PHP, IRC channel on freenode and said, [00:12:00] what's a good content management system. And three people said Drupal. And that was it we like picked Drupal but, and. We had evaluated like a couple others, like PHPnuke, PHPdb, WordPress was, I don't know. I think it might've still been called BBPress at the, at that point in time, and Drupal felt like the right fit. Cause it had the ability to handle multiple types of content was like a core thing in there. It had some, like, good inbuilt RSS support.
[00:12:23] Josh Koenig: We really were really high on like RSS is going to be a way to coordinate these. And we actually did. We created a, like a national syndicated, RSS network for events, select the campaign. The national campaign could propagate out events via RSS to all these other websites and have them show up and like basically create a coordinated global calendar.
So, yeah. We picked Drupal to, to build this thing. We rebranded it as DeanSpace, because Hack4Dean sounded scary to some people, and built this, like three months built this, like what, in retrospect you could squint at and look at and call like an installation profile or a Drupal distribution for the campaign.[00:13:00]
It ran hundreds of affinity groups: Pilots4Dean, Teachers4Dean, Farmers4Dean, whatever. And then the campaign itself used it for their state level or organizing things like that. Their like Iowa website, New Hampshire website, California website, whatever those were all using that software.
And, you know, Zack moved, actually went to join the campaign and moved to Burlington. And like some people actually joined up with the campaign. A bunch of other people remain like volunteers on the outside, like I did, but that was how I found Drupal. And we were, it's interesting.
We, we spent I think in that first three months in that burst of activity, I didn't get involved in that Drupal community really at all, because we were just like really on the outside focused on just building with this thing. And it was only when we got past that initial burst that we realized that there's a whole wealth of knowledge and information inside this community that we could start tapping into.
And so that was a, it was actually a self-interested thing. as it got to be like the end of 20, 2003 to go and start engaging, not just [00:14:00] reading doc pages, but actually engaging in the Drupal community. And then like, we had like always thought, we should give back some of the code we're writing.
So thinking about what do we have to do to actually release some of the stuff that we created and so forth. and, yeah, so that was my intro to the Drupal, the Drupal community, and that got followed up because the Dean Campaign was obviously not successful. What was it? It was actually unsuccessful and rather like spectacular fashion.
And which was heartbreaking to be honest, like we put a lot of, a lot of our passion into that and it just went from, could actually work, might actually take over the government to like, Nope, not gonna happen, in like a week. And, but a bunch of people kept going I worked for a nonprofit that was I found work with,a national nonprofit that was generally aligned with, it wasn't like a campaign, particular candidate,thing, but it was like trying to do online, organizing and activism from a progressive standpoint. And we kept using that technology stack.
[00:14:55] Josh Koenig: So I got more and more into Drupal, over the course of 2004, as I was like, helping with this Music for America project. And Zack started CivicSpace, which was basically let's take the concept of DeanSpace and make it non-partisan, totally open source. Like for all the NGOs in the world.
Cause at the time the nonprofit tech world was like, really dominated by some not so great legacy proprietary what you would call today, SAAS platforms, but they were just not awesome, right? It was, like very rudimentary tools, no ability to really like work on. Like what the actual digital experience was and like really expensive for these NGOs to run.
So that was the idea to create an open source alternative to all those things, which was pretty successful. Yeah, so that was like, through it, through that like kind of crucible, I ended up on the other side. Having not actually been in a position to start giving back to the Drupal community probably about a year after I first got into it.
[00:15:49] Michael Meyers: Well, DeanSpace.
I remember that campaign vividly. And DeanSpace revolutionized American politics online. I think you really undersold it. It really became the model by which every campaign was run, moving forward, it was transformative and it was also the first major success on the map for Drupal.
You know, I don't think I would have discovered Drupal if not for that, And that was that, that, when you go to use the platform, you say, who else has used it? What has been done with it? And, I was like, oh my gosh, that, Dean, so it was a major reason that I got into the community, ended up using it for my company.
So after DeanSpace, you know, Music from America, I think it's interesting that you spend a little bit of time at Trellon, and I just want to mention that I don't think Trellon is around anymore. Unfortunately, at least there was a 404 on their site when I checked it out this morning.
But so many amazing people at, at some point in time spent a stint at Trellon, especially a lot of like early people in the community, like more of a Sanford, the IRC chat bot and, Jim Gilliland who's on the security team. It's just, it blows my [00:17:00] mind, the number of amazing people that at some point,were touched by Trellon.
So then you went on, after, after all that and around 2006, if I recall, which is roughly a Drupal 4.5, you team back up was that. Zach Rosen, and Matt Cheney and you guys formed ChapterThree. what, what led you guys to come together and create a Drupal agency?
[00:17:23] Josh Koenig: Uh, that's a great question too.
So, let me see if I can roll. I can talk a little about Trellon. Cause I think there's actually a, there is a bridge point that's in there too. Okay. By the end of 2004, and like that the actual presidential election cycle, which was also heartbreaking, I got, I was just totally burned out.
Like I, early 2005, I was like, I need to, I need a break. I can't do politics right now. I need to do something else. I need to figure out a way to just hit the reset button on my life. So I, I started, I basically couch surfed for three or four months, in between San Francisco and New York City.
And then [00:18:00] I got together with the, my two high school best friends, and we pooled all of our money and did a three month cross-country road trip, which was actually. This is not about technology, but it was actually for me really. it was a lot of fun, first of all, but I like it at the point where I was like being so like burnout on, on, on politics and the election and the campaign, like spending time, just like traveling through the country and like meeting regular, totally regular old people in places like Texas and Alabama and Georgia, we spent time in all these like red states, I think very positive for me in an interesting, healing way. And like we put a travel blog up of the whole thing, obviously running on Drupal. And, I think that's still online if people want to,can search for it. http://www.vagabender.org/. And, and at the end of that, I came back to my home in Oregon, to my mom's place.
[00:18:52] Josh Koenig: And, we parted ways at the end of the road trip and I was basically broke. Like I was like, okay, mom, I need to move in with you for a little while. Cause I need to figure out what I'm doing next. And I started like, looking for a Drupal job because now I have these skills. And like in the intervening time, like the Drupal economy had started to really develop between 2004.
And in 2006, this was like, sorry, summer of 2005. So like, over that ,or that year and a half, this like Drupal economy had started to develop. And there were now these Drupal agencies that were out there and I emailed all of them. I think there were like six or seven and, and Mike Haggerty, he got back to me right away.
He was, “I will put you to work tomorrow.” And,I was like, great. That's what I need. And, and I think honestly like his, like I move fast. I like, hire people when they come in. I know how to spot talent. Like, I think that is a big part of why so many people I've spent time working. It did spend time working for that company.
And, and yeah, so I worked there for about eight or nine months and worked on some, some interesting projects. I met some, like some other really interesting, I had that connection to that also through, through Andrew Hoppen who had been [00:20:00] involved in the, the DeanSpace, CivicSpace stuff.
And, met some of the, my, my colleagues at the time were really great people, too. And, and it was with Mike and co that we went to Vancouver, British Columbia for the open source CMS summit, in I think it was like January or February of 2006. And that to me was like, it lit the spark again because, it was the first real, like, in-person, Drupal.
[00:20:29] Josh Koenig: It was actually the, I think the first in-person like big tech thing I had ever been to, that wasn't like a, yeah, that wasn't like, strolling around the halls at Mac world, which is like a consumer experience. This is here are the people. And we went to the Bright offices in Vancouver and Boris Mann gave a big speech and all these people were there.
And it wasn't just Drupal people. There's Matt Mullenweg was like hanging out and, And the power of that community and the people who were there, looking at. [00:21:00] shared opportunity and the chance to do something, it was this idea of okay, there were, there's potentially commercial interests here and people are starting businesses, but we're like, we're in a, like this idea of coop-ertition and we're building a commons that everybody can like leverage, was really, I, it really resonated with me.
And I was, and still am to this day, like a big believer that the Internet is going to be a, is, and will be a huge net positive for humanity and a, an important part of how we actually navigate the 21st century as a species. And being able to be a part of that community that's working on make, again, making better tools to build the Internet for people.
It, it rekindled that same, like passion, in me that I had, hadn't felt, for at least a few years. And, I got really excited and it got more involved in the community and, and that was sort of,you know, I just had a lot more ideas of my own and this is what I want to do.
[00:21:52] Josh Koenig: This is where I want to take things. And it wasn't possible to do all that while I was working full time, like doing client work. and so I had to [00:22:00] leave the, the job that I had without anything else lined up. Actually, I like what I did was I, I was talking to all my friends about like, my kind of dilemma.
And I feel like I'm a little bit stuck. And one of my friends that had been on this road trip with he lives like, in like remote Northern California. and he said, listen, man, why don't you come out here just for the summer? I have the Internet too, and your rent could be like 200 bucks a month.
And I said, that's not a bad idea. Let's try that. And, and so I went and lived in a town called Trinidad, which is way on the Northern coast of California. And I started going back down to San Francisco and then I reconnect with Zach and Matt and we had like very similar and we're going to be friends with Zach and through Zach, Matt, they had been at a UIUC together and we had a lot of the same ideas and it was what we landed on was you saw okay, so Mike did Trellon and, Jeff and Matt did Lullabot and there's the, you know, Bright did their thing.
So we could start out, we could start our own thing. We can be our own, we can be our own boss. we could call the shots. we could decide what we wanted it, [00:23:00] what it is we want to do. And that was where, like the idea for ChapterThree came from of just like we could try to chart our own course, in this emerging Drupal economy and be able to work on the things we wanted to work on.
And, yeah. And so that's that summer of 2006, we decided to go for it and throw in together to see if we could get this thing off the ground
[00:23:20] Michael Meyers: And get it off the ground, you did. I think where we first met is that Open Source CMS conference in Vancouver. I was racking my brain this morning before we chat.
I mean,
[00:23:31] Josh Koenig: Totally. Yeah. Cause you were. Were you working with, with Bright and Co or no, you are with, this was, sorry. I, your early career, from that, what was the name of that? the news site that you were like the CTO of?
[00:23:45] Michael Meyers: Yup, I started a company called NowPublic, which was the first venture backed based startup.
And Boris Mann is another key reason that I used Drupal. I sat down with him in Vancouver. He gave me like tons of his [00:24:00] time and he was like, this is why it's a good platform versus everything else that's out there. And he was a major reason. And then NowPublic ended up sharing and buying office space with Bright, like our two companies were co-located for gosh, the first two years of NowPublic. And then they merged with RainCity and kicked us out
[00:24:19] Michael Meyers: because they needed space. And, and we found
office down the street. They had a gorgeous office on the water in Vancouver. But, so ChapterThree, is still going strong today. you got, you ran it day to day for 15 years.
And then, I'm curious, cause I remember being in your office and talking to Strauss, before Pantheon was real. and I don't remember who all was at the table, but I remember being in the conference room and Strauss is always blowing my mind, with this amazing idea and vision.
And I think back to that conversation and so much of what. You guys were talking about in that room has come true today [00:25:00] with Pantheon. And so I'm curious, how did you go from, okay. We have a digital agency to, we want to create Pantheon.
[00:25:08] Josh Koenig: Oh, that's a good, that's a good, good story too. So I'd say this is a place where, some of this was, it was the experience we had at ChapterThree and like a little bit what we were seeing around us.
And it was the experience, both positive and in some cases, not so positive. We, so ChapterThree, started with three, three guys. and we grew it over like really four years from the three of us to about 30 people, that like boutique agency size, where like the next level of the game is figuring out how to scale, management and client services and other stuff that, that, and to be frank that at that point in our lives, none of us were passionate about those sorts of challenges.
[00:25:45] Josh Koenig: And, but we had done it all completely bootstrapped, right? just with our own grit and, not taking a paycheck, sometimes we could pay other people and it was a great experience, honestly. Yeah. I would say, we can talk later on some of the things that we were able to do in the community during that time, I think were [00:26:00] really positive.
But as it pertains to Pantheon, I think. What was happening for us was, we came up, with the agency and the agency grew and Drupal grew, and we started working with bigger and more exciting and interesting and challenging customers. And we started running into a lot of common patterns of problems, regardless of what the website we were building was, who was for what that design was.
Those are all very different Drupal development challenges that were unique, or more or less unique to the project. But there are these more fundamental kinds of meta, operational challenges that we kept running into. And the solutions we developed for those were like converging. You know, getting just incrementally better based on our experience and the challenges were around, how do we make a team of developers effective together?
[00:26:49] Josh Koenig: Especially if that team might include people from different organizations like you work on larger projects, maybe there's more than one agency involved. Maybe there's some subcontractors, maybe there's some developers or [00:27:00] IT people from the client side that are participating in the process. How do you make that work?
So you're not stepping on each other's toes, so that you can effectively collaborate, so that you can review, so that you can have transparency, so that you can integrate. and then like when you want to launch a website, how do you make it scale. And that, that was one of the big things that, uh, you know, David was really instrumental in with Pressflow, was really proving that you could scale Drupal.
In fact, actually going back to the, the campaign days, just as a quick aside, like you're totally right. I don't know. I would say it's a stretch to say that DeanSpace in particular laid down the, the track for digital organizing and presidential stuff going forward. But it was a part of that, but like quite literally the core digital team from the Dean campaign founded an agency called Blue State Group.
That then was the Obama tech operation, like the same people, the same core group. And I remember having a very late, and loud conversation with one of their principals in like, at Netroots nation 2007 maybe. And I was trying to tell them that they should use Drupal [00:28:00] for more stuff. And they were like, no, you Drupal can't scale.
[00:28:03] Josh Koenig: Got to do it like Movable Type does you gotta, you gotta write out the static files so that you can scale it because we have really big scale challenges. And at the time he, he was kind of right. it was, uh, you know, Drupal's, had some internal, like pretty good, honestly at the time, relative to other open-source projects.
So the internal capabilities to scale with like page caching and stuff. But the sort of stories of Drupal can survive a slash dotting, which was not true of every software, but it really wasn't ready to take on like out of the box. It wasn't ready to take on sustained large-scale traffic loads.
That was a kind of, you needed to do extra work to make that happen. And as we were working on these larger projects, right at ChapterThree, we had to do that extra work over and over again. So these twin operational challenges of how do we get teams of developers to work together effectively?
[00:28:52] Josh Koenig: How do we make websites that can scale? And we developed good expertise in those, through like the school of hard knocks to some extent. and, and then what happened was, I guess two things happened that started to spark the idea of - that eventually became Pantheon. One, we, you know, you go to the, now there was DrupalCon, And DrupalCon's amazing. and so DrupalCon, like you have your case study track where people tell their success stories, and then you have your like hallway track or your happy hour track where people sometimes talk about the not so success stories, which are interesting. And I remember talking to other peers in the digital agency space, and, you know, you can talk about like, projects that went off the rails or clients from hell or whatever, kind of stuff you commiserate about.
And a lot of the really, harrowing stories that other agency owners had to tell. Had this, these same features were like, yeah, we were getting a little bit behind. So we tried to add more developers that actually slowed things down and we got more behind. So we really had to like, really do a really big push to get to the launch date.
And then, the one - the website went live and then it crashed. And so then we had to put in like another, [00:30:00] like super, sprint just to get things stable. And they're stable now, but nobody wants to touch anything. And also the clients telling me they're not going to pay me the final installment because of all this, all the problems they want a discount.
And I still have to pay my people. And it was like, oh gosh, this puts peoples. So sorry. I think my Internet got unstable for a second. I'll pause.I'm back. Yep. so I can go back. So these, the, these are real, these challenges have real world consequences. Like it's websites, right?
We're not like nobody's going to live or die, if their website is up or down, but it really does affect people's like stress levels. And and if you can't, if your client doesn't pay you and you can't pay your staff, then your business will unravel. And, and so we thought, oh gosh, there's a real need for this stuff.
And we were doing pretty well at it. I, you know, not to toot our own horn, we were pretty good at this stuff, but it was like, expensive, right. It's like, okay, I have to convince the [00:31:00] client to hire me. And like my two most senior tech folks to do six weeks of work, up front before they'll see anything before they even see like, hello world with Drupal and like at our hourly rates, like that's expensive.
Like you're talking about people that can like, make like a high five low six figure investment upfront just in workflow and infrastructure that really makes it a, a luxury product that not everyone can afford. And so our realization was that, there was a real need to make this, it, there was a need to make this kind of operational capacity that we were good at building as a one-off much more available.
And so the first thing we did was we tried to open source it. We got together and this was still ChapterThree and David was still with Four Kitchens. We had worked together on a couple of these projects, so we were collaborating around this stuff. And I went out and I made a - these Amazon machine images like this early days of AWS, and you could make it EC2 machine image and make it publicly available.
[00:31:58] Josh Koenig: This is it was like, whoa, revelation. And, we made one for, to showcase that you can use Drupal with Varnish. So there could scale called Mercury. We made one with a Jenkins continuous integration setup called Vulcan. And then we made one that showcased that you could run Agar, like to do multi-site and.
The, they got some interest and in particular, like the Mercury one, people were just like, oh, cool. So where do I pay you for support for this? Like you're going to sell this to me, right. And maintain it for me. And I was like, no, I'm not. I'm relabeling all of this alpha, not for production use because I don't want as an open source project, I don't want that.
I can't take that kind of responsibility. and you know, but it demonstrated that again, there was this real need there. And partly, kudos to Zach for being like, ambitious and visionary enough to think we could do it. But we were like, Hey, we live in San Francisco.
Look at what's going on around us. Like we could maybe start uh, we could start up a new kind of company. You know, we, we like beat level one of the agency game and, but we don't need to say to play level two. We could play a different game and actually make a [00:33:00] product. We can build a platform.
And yeah, we need to work on that a lot to make it actually responsible to put anyone's hands, but like we could raise money and do that. And, and so we we like finagle our way at having dinner with the founders of Heroku. And they like these,you know, pretty successful at the time platforms and service.
[00:33:17] Josh Koenig: Like they're like Ruby guys. They're like from LA, they were very cool. They're like, they still are very cool. Um, you know, very stylish individuals and they were just like, you guys got to do this, and, we'd love to have dinner again, but not if you're not even decided you're going to do this, you gotta do it.
And so we decided to do it and it worked out. So again, more, more high-risk advice that went in our favor, firing ourselves from our jobs at the agency, stop taking a paycheck to really make it real and, and went out and raised a little bit of money to, to build a beta version of the platform and then use that to raise enough money, to build a real version of the platform and took it off from there.
[00:33:54] Michael Meyers: I got to ask when you guys started out, did you ever in your wildest dreams, think you were going to build a billion dollar company?
[00:34:02] Josh Koenig: I, in our wildest dreams, yes. Like that's the, and the thing is that that's part of the- like that's what those mechanics are what makes Silicon valley venture capital work like any, like you don't get, you can't raise money from people if you don't credibly have an ambition to build something that could be that successful. It's a weird, it's a weird thing. And again, I say like high-risk advice because it's a little bit like the music industry, it's a hit, driven, business, only one out of seven of these, of the VC backed projects will even make it to any kind of success.
And it's like one in 20 or less is a meaningful success. And it's like the one in a hundred that really make people's portfolios. So what, but what they're looking for is people who are going to swing for the fences and have a vision that, you could, again, you're thinking like, okay, I can squint and see how in 10 years, if these trends continue and these guys can execute or guys and girls now, thankfully, can execute it could be a really big deal. So like we had this idea like, Hey, we want to run a non-trivial [00:35:00] percentage of the Internet. and if we do that, yeah, that's a billion dollar business. and,and, we're, we got a long ways to go to be honest, but we have, I ran some stats the other day and we get over a billion unique visitors to some website on Pantheon every month, which is about 30% of the Internet browsing human population and, and yeah, and we've got a long ways to go still.
[00:35:25] Michael Meyers: That's amazing. That's mind blowing. I think you, you have to believe, if you're going to start a business, you have to believe, humorous or not. You have to believe that you can get there, or, you have no business starting a venture backed business for that matter.
Yeah. Sort of lifestyle business, but you're not going to get very far in, in the venture world. So we've talked a lot about your professional history in Drupal. a big part of Drupal is its open source nature. It, its successes, the community, a good segue, I know Pantheon does a lot for the community.
You know, you guys are the top sponsor [00:36:00] of every major event. You are a top sponsor of the Drupal Association, but I know you guys do so much more than that. And I want you to share so people know because I think it's really important that more companies get engaged and give back in more ways.
Some of the ways that you guys contribute that people probably don't know about.
[00:36:21] Josh Koenig: Yeah, I think it's an interesting question. because we don't, we don't have people who are like, full-time on staff working on Drupal Core. That's just not something that ever made sense for us to do, for a variety of reasons.
But the, we do make a bunch of technical contributions that are about that are actually about everything else around Drupal. So we've worked through a bunch of the things that are necessary in the Linux ecosystem to really support Drupal and like another, PHP, CMS is so that they can work well and they can work at scale.
Some of the contributions we made to, to Curl and systemd and other of these like like really like low level fundamental, [00:37:00] components that like, Drupal sits on top of a whole stack of other open source tech. and frankly from a technical perspective, we've given a lot more back to that underlying tech stack then to Drupal Core itself in particular.
We do make some other contributions to, in terms of the security team, and, and other things which are important. but yeah, it, I think it takes all kinds right. For us, as a, as an operations platform for the web, right? The frankly like the internal skills we have within our engineering organization and the place where we spend most of our mental energy and time is at that platform layer.
[00:37:32] Josh Koenig: And so that's where most of our contributions go back in. And, and I think it's right that like for companies that want to give back, like where you give back should be a combination of in your sweet spot of like where you legitimately have the skills and expertise to do, and it should be aligned with like your interests.
Like I actually, I think it's fine if people want give back as a form of charity, but it's not particularly sustainable, to look at giving back to open source as a [00:38:00] charitable activity. because as soon as push comes to shove, it'll always get like, dialed down, like giving back to open source should be strategic, in my opinion.
And so yeah, and like beyond that, like there, there's other places where we make contributions and we're actually building a team now internally, to do this. More, organized in sustainable fashion to look at third-party contribution third-party modules, third-party libraries and so forth.
That need, again, it's very often, for our, with our unique focus, not unique, but with our singular focus on helping people ensure like high-performance and scalability and all the sites, cause that's just fundamental table stakes for having a credible web presence. There's a lot of code that gets written to open source where the author does great work, but they're scratching their own itch in their particular use case.
And it's yeah, it works for me on my laptop or it works for me on my small website. And I'm very happy to give this back to the community and have other people use it, but it hasn't actually been - it isn't actually set up to handle higher performance or large-scale use [00:39:00] cases. And,but the good news is really often, it's not a huge change to get it to work in a scalable way.
And so if we can, the ability to give, to make things work, you know, make things like that,make those types of contributions so that there's less kind of foot guns out there in the open,in the contrib space where people like, oh, this looks like a great thing.
Let me turn this on. And something like your performance goes down, and you have to like frantically turn the thing off. We will be like to avoid those moments for people.
[00:39:25] Michael Meyers: Wow. I think contribution back should be driven by self-interest alignment of goals. I wholeheartedly believe that's the way that open source is going to be successful.
Your first contribution to Drupal, do you remember, I, you know, there are many ways to contribute to your ripple, as you said, code is just one of them, but code is at the core of Drupal and you've done hundreds of code commits to the platform. I'm curious. Do you remember one of your earliest commits and what that experience was like?
[00:39:59] Michael Meyers: Do you [00:40:00] remember what you committed and how it went.
[00:40:02] Josh Koenig: Yeah, I, I'm, I would be lying if I said for sure I remember the first line of code I contributed back, but I can, there's the first kind of things that I felt were meaningful. And I was, that was proud to do, one was still today. The only line of the only contrib that I have, that's in Drupal Core, and it's probably not even Drupal Core anymore, cause it's probably been rewritten, but there was a bug with the comment preview, functionality and Drupal six, oh no, Drupal five.
And I was like, oh, I like, I was annoyed by it with clients. And I was like, I'm going to figure out why this is happening. And it turned out it was a one line fix. It was just like- somebody missed like a class attribute. It was supposed to be applied to something. So I wrote it up as a patch, followed the instructions, send it in it, and it got committed.
And I actually- it felt really good. It was like, a small thing. but it felt very meaningful to have okay, I've got one contribution in there. and then from a in third-party contrib land, like the most meaningful thing I think I did there was, I wrote up the original version of the Varnish module, [00:41:00] which ended up being taken over by, by Dick Olson who did a much better job than I did at stewarding and managing, managing it going forward and handling contributions.
But I, I wrote the first,the original sort of instance of the Varnish contrib module to make, a module to help Drupal and Varnish and better yet Pressflow and Varnish work well together. Yeah. So I think those, I think of those as my two, like code, when you say code commits, those are the two things that come to mind.
I probably had some other stuff to the event module too, but I don't remember it for sure.
[00:41:28] Michael Meyers: You've made tremendous contributions to the community outside of code. What are you most proud of?
[00:41:35] Josh Koenig: Oh, I see the single thing that I think I credibly did to help Drupal, outside of the commercial stuff, like commercial stuff does help, like popularizing Drupal, building it into successful things.
Evangelizing Drupal. That's real. But the Drupal dojo was like my “Let's give back”. And it's an intentional thing where I wanted to give back, in a meaningful way. And it worked really, it works, much better than I thought it would. In a way that I think actually had, I can [00:42:00] see how it had really positive ripple effects.
So this was, like 2007, which we had gotten ChapterThree off the ground. I felt like I had been leaning on the community really hard while getting ChapterThree off the ground, just like, oh, we gotta do this for a client. I don't know how it works. Let me go into IRC and beg. Or I need to hire somebody real quick.
[00:42:19] Josh Koenig: Let me like, lean on the community. And I was like, okay, I gotta balance the scales here. Somehow. And we had,the groups.drupal.org had just gotten spun up and I thought let's use this. And what I'll do is I'm not saying that I know everything, but I like, again, like I have this particular skillset, which is not like, how do I build a Drupal Core, but it's how do I build a website for a use case with Drupal, right. And that was something that like, that was what all the agencies were doing, but there wasn't a ton of like, documentation around how to actually do that.
There's lots of if you want to write a module in abstract, here's how you go about doing it. And there wasn't like a clear thing of like, here's how the node hook system works. And if you're thinking about [00:43:00] customizing a publication workflow, this is a way to imagine how it's, how the pieces stick together.
Here's how you can put, put something in place. Here's how you debug it. And so what I landed on was I'm just going to do these things like, I'll do it every Sunday, from my like, my bedroom and in like remote Northern California and I'll just live stream. At the time we cobbled together this tech stack from like a bunch of bits and pieces, but now it's you just do it on Twitch.
But we were like, we built it ourselves. I'll just live stream myself trying to code, making mistakes, figuring it out, and I'll have the IRC channel open over here. So I can take questions from people and respond. And like it's literally exactly what Twitch is. And there are people that do live coding on Twitch now really successfully.
But we just assembled it ourselves, a while back. And what was amazing about it was, at first I thought, okay, I'll get like maybe 10 or 20 people that want to learn stuff and I'll mentor them up through this process. And then they'll, I'll have done something. And we got like over a hundred people by the third session.
[00:43:58] Josh Koenig: And then other people saying, I want to do this too. can I guest in? Can I do the, can I have a guest spot and do next Sunday? And so it just turned into this like really great community thing of people coming in and doing like presentations and showing their way of doing various things with Drupal and for a good couple of years, it was a very active self-propelling learning community that I know was helpful for a lot of people who are discovering Drupal at the time to get up to speed and to feel confident of using the platform.
[00:44:28] Michael Meyers: It had a major impact for sure. All so some, a tough question here, a little controversial, what is your favorite feature or aspect of Drupal and your least favorite?
[00:44:40] Josh Koenig: Oh my, I'm gonna, I'm gonna be, Let's answer this from a developer's perspective, because that's still how I, that's, how I feel Drupal is as code, less than as a, like an admin UI. And it's the double-edged sword. I love the way that Drupal has done certain things to abstract and make pluggable.
[00:44:57] Josh Koenig: And I hate the way Drupal has done certain things to abstract and make pluggable. and it's one of those, it's like a, it's like a, it's like a cliche engineering answer perspective of, you can just, if it's controversial because people put their, put a lot of work into doing these things, but I think you can take a step back if you look back at history where, where abstraction was a preemptive optimization, and ended up being a bunch of work that didn't yield much value and increase the complexity of the system in a way that didn't help anyone or it didn't help very many people, versus where the abstraction was like, oh no, that was the thing that we needed. And that like now enabled a whole ecosystem of new things to be built on top of it.
That totally is rad. So like if I had to to answer it in the controversial form, like I would still go back to like the core, the idea of node hooks, I think. Cause that was so it's just so powerful, so useful, like such a great way to get into like the decorator pattern of building things out and based on this is what you do when you're managing a [00:46:00] core, Content thing doesn't have to be a page. It doesn't have to be a blog. Doesn't have to be advanced. It's a note of some type and it goes through this series of processes and allows anything else to plug and play with it. Along that journey of being saved, edited the whole CRUD journey, That's actually really smart.
Like that's an internal pattern that isn't in on a lot of other content management systems and frankly everybody should copy. It's awesome. And there's other things that came after that with like entities and so forth that are similar, like smart, really solid content oriented thing. If I was going to point out something that I thought was not a great optimization and has every time I've had to deal with it has led to like kind of pain and suffering and be like the database abstraction layer.
Like we'd be way better off if it was just like Drupal only runs on mySQL. sorry. It's just. Because you would prevent confusion, you'd eliminate a bunch of code. You could lean into the things that mySQL does and only my, like in a certain way, and actually get value for the system versus having to settle for the more abstract, just generic ANSI SQL [00:47:00] feature set.
And I think that was like a missed opportunity early on. Again, like most of my like deep, like Drupal opinions or from not from the past few years, it's from like the time when I was like really involved in building sites, but I feel like that the, if I could go back in time and just say, let's not do database next generation, let's take the opposite path and say like Drupal's going to lean into mySQL.
I think it would have helped the project.
[00:47:24] Michael Meyers: For Drupal to be around for another 20 years. What has to happen? what do we need, what are we going to change?
[00:47:35] Josh Koenig: I think the, the biggest things that probably need to happen for Drupal to thrive for another 20 years. I think Drupal will be around for another 20 years. Pretty much, no matter what, but in that Dries sometimes had a set of keynotes, aren't worried. It's we, you want to be around like cockroaches are still around or we want to be around.
We're a real part of the, the story. Drupal has enough of a footprint and enough like places where it's not going, it's not going to disappear, [00:48:00] but for Drupal to really be thriving in another 20 years, I think that, you know, it's not one thing from a technical standpoint, but like Drupal has to attract a new generation of talent.
And, th- that's hard to do. That's actually pretty difficult to achieve. but frankly, that's what it would take. there's a, you know, there's a lot of people that are about your, and my age that were part of this community. And it's because we came up with Drupal and it was formative in our careers.
[00:48:25] Josh Koenig: And we learned it when we were, up an comers and, cut our teeth on it. And a lot of ways that doesn't happen. Now very much. and, and I think for Drupal to really be a thriving, vibrant thing, like the amount of like large-scale business and institutional investment in Drupal is going to keep it alive and healthy for quite some time.
But if it Drupal needs to find a new wave of talent, and I think there's a way that could actually happen in the sort of next generation, modern web development context, where,people want to build with,people are people who were building the [00:49:00] web now are increasingly, and for good reasons, starting with a user experience and building backwards from there.
And that's why everyone's interested in headless so forth because they build back to needing a content management system. And then I don't want, I've already built the user experience. I don't want the CMS to give me pages. I just want it to give me content. And, there's a real gap in the market right now.
[00:49:19] Josh Koenig: There is no solid, widely supported, understood open source, headless content management system. It doesn't exist. and there are a bunch of proprietary SAAS, of headless CMS is that are really good. and I, it's a stretch to think that like Drupal could fill that gap because it's a pretty big gap to fill, but there's a, there's definitely an opening there for a CMS that can support, the modern user experience development,and,and do it in a way that's open source, which I think is, would be a really positive and awesome thing for Drupal to do because people need to own their, you want to own your content model.
You want to own your editorial workflow. You want to own your data. Those are all [00:50:00] like, really good, pragmatic reasons to want open source for your content. but you know, people also want a really great user experience and they'll give up a lot to deliver a better user experience, sometimes.
So I don't know if that's actually feasible or if it's even, or if it's even the right thing, there might be another path for Drupal too, but that's one that I see that could be interesting.
[00:50:19] Michael Meyers: This is a tough question to ask, because I'm sure there are countless people that have helped you or influenced you along the way. and Drupal is a community and one person is not a community. But you know, if you had to single out one, or maybe two people who were really influential in your Drupal journey as mentors, who would you think.
[00:50:44] Josh Koenig: Um, so yeah, I, I have like, uh, the super old timer answer. so it would be, ChX and DWW like old heads know. But it's like, I, it was, you know, at the time that I was trying to build a [00:51:00] mental model in my head of how Drupal worked then, so that I could feel confident creating stuff with it. They were the two people who spent the most time.
Typing back and forth with me in IRC until it started to click. And that really, that really, Karoly and Derek, I really owe them that debt of gratitude, for being willing to just, volunteer to, to teach me, about this stuff. And, yeah, so I'd say those are probably the two people that were the most like mentors for me in, in, in Drupal.
There are a lot of people that I found very inspiring, but like in terms of like where I actually got the hands-on help, it was the two of them.
[00:51:37] Michael Meyers: Yeah, both amazing people and brains, Karoly, ChX was very influential in my career. Yeah, he was my first employee.at NowPublic, worked with him for seven, eight years.
And every day he managed to blow my mind. He was, you know,controversial, but amazing. And, I learned so much from the guy. I miss him a lot.
Alright one [00:52:00] last question you've been overly generous with your time. Who should I interview next? When you think about, people in the community that have been a major part of its history and success?
[00:52:11] Josh Koenig: That's a great question. I, so I would say again, just sticking with like people that I feel like. I had some kind of closeness to, and that aren't like, you know, that might not have art. I'm going to try to get again, another good deep cut.
The hips, hipster, Drupal, I think you should try to track down Dimitri Gaskin.
[00:52:29] Michael Meyers: Oh, wow.
[00:52:31] Josh Koenig: DimitriG01. Yeah. he, who, who like he, I remember for him coming and he was like the child prodigy of, of Drupal. I remember him coming into think it was like DrupalCon DC or something like that.
And he got up on a step stool to be above the podium to give a talk to 300 like adults. And he owned [00:53:00] like he wrote a bunch of good code too, but I would just like, a really, and I he's in the Bay Area. So I met him a few times, as he was growing up and he's gone on to, he went to Stanford, he's doing a startup now.
He's doing great. but he was just like great, really fantastic individual. And, and did a bunch of, I met him through the Drupal dojo, and he did a bunch of really cool stuff and would be an interesting person to talk to.
[00:53:19] Michael Meyers: That's an amazing recommendation.
I remember that presentation. So, I mean, it's just the image of him putting that stool in place and getting up and that you said owning the room. It was amazing. He is a brilliant guy and I will definitely reach out to him, dude. I wish we could keep going. There's so many things that I would love to talk to you about, but like I said, you've been overly generous in your time.
We're over time. so thank you so much for joining us. I really appreciate it. I know you're a really busy guy.
Yeah, no, it's my pleasure. and thank you for, thanks for doing this project. I think it's a really wonderful thing to do to you know, kind of at the 20 year mark to collect these stories.
[00:54:00] And, thanks for hosting this and giving me a chance to like, reminisce about the. some of like the funnest times of my career, to be honest. And yeah, the community has been a huge part of, of our success. And, and it's also just been a part of who we are as people and so it's good to recognize that and nice to remember, all those things too.
[00:54:19] Michael Meyers: Yeah, I'm going to try and do one a week between now and the end of the year. I, every time I do one, I love it. Like I said, reminiscing, catching up, we lead busy lives with so much going on. it's great to see you, to our listeners.
Please make sure you check out the other interviews in the series. Tag1.com/20 that's two zero. if you liked this talk, please remember to upvote, subscribe and share it out. You can check out our other Tag1 Team Talks at Tag1.com/talks, as always we'd love your input, your feedback, on this show on people we should interview other topics we should cover.
You can reach us at talks@tag1.com. That's tag the number one.com. [00:55:00] Josh again, huge. Thank you for joining us to everyone who tuned in. We really appreciate it. Thank you.
[00:55:07] Josh Koenig: Thanks for having me, Michael.