The Drupal Association’s Chief Technology Officer, Tim Lehnen, talks with Blake about the future of Drupal. Plus, Brie and Blake talk about how an upstart candidate for president led to Drupal’s explosive growth in the late 2000s.
Blake Newman [00:00:06] From Washington, D.C., I’m Blake Newman.
Brie Ripley [00:00:09] I’m Brie Ripley.
Blake [00:00:10] And this is the Agile Podcast where we explore the stories and people behind agile web solutions.
Brie [00:00:15] And on this season of the show, we’re talking all about Drupal 7, which is reaching its end of life very soon. But you should think about moving out of Drupal as soon as … Blake?
Blake [00:00:27] As soon as possible. And today, on the show, we have an interview with Tim Lehnen, the chief technology officer of the Drupal Foundation.
Brie [00:00:34] Which is a nonprofit that supports the open source community of Drupal.
Blake [00:00:38] And we’re talking to Tim about the future of Drupal, where Drupal is going and how this move to Symfony in D8 and D9 fits into that.
Brie [00:00:45] But first, we’re going to talk about Drupal’s past. We have a little bit of a history lesson.
Blake [00:00:51] So maybe you’ve heard this from me a few times. But Drupal’s big moment was powering whitehouse.gov, which took it from relative obscurity to a household name.
Brie [00:00:59] Household name?
Blake [00:01:00] Well, sort of. I mean, at least in the web space, particularly for government agencies and nonprofits. This was during the Obama administration.
Brie [00:01:08] Right after the 2008 election, the new administration wanted a new whitehouse.gov stie, one that was more focused on civic engagement. And this ended up being a big win for open source software. This signaled that it was safe to use in the federal government.
Blake [00:01:23] But Tim traces that history back even further to the 2004 presidential campaign of Howard Dean.
Tim Lehnen [00:01:28] If we go back earlier, for example, than the whitehouse.gov scenario, probably the first place where it really got that attention was also in the political sphere because Drupal was used in the Howard Dean campaign for the so-called first digital presidential campaign. Actually, a member of my team, Neil Drumm, was kind of central to the organization that did that. And that was, you know, end of Drupal 4, Drupal 5 era. That kind of kicked off the evolution of Drupal from being, hey, it’s another CMS option–like Joomla!, like WordPress, like Typo3–to well, actually, maybe it can be a citizen engagement platform.
Brie [00:02:11] Ah, Howard Dean, former Vermont governor, ran a long shot campaign for president as a Democrat and actually became the front runner. And then it all fell apart in the Iowa Caucus. He came in third and never recovered.
Blake [00:02:24] And then the scream.
Brie [00:02:25] Right. And then the famous Dean scream. And are we still talking about Drupal?
Blake [00:02:30] Okay. Okay. Yeah. So there was this group called Hack4Dean. They wanted to sync up all the different Howard Dean supporters, groups like Doctors for Dean, Michiganders for Dean, things like that. So they used Drupal to build this tool where you could easily launch a web site for your Howard Dean group. And then from there, you can create mailing lists, communicate with other Howard Dean sites, and ultimately raise money for Dean’s run for president. It was called Dean Space.
Brie [00:02:55] And because of Dean’s campaign, Drupal went to South Carolina and Oklahoma and Arizona and North Dakota and New Mexico. And it went to California and Texas and New York and then went to South Dakota and Oregon, Washington and Michigan. And then it went to Washington, D.C. to take the White House because those Dean Space people went on to work for the Obama administration four years later.
Blake [00:03:19] Including Neil Drumm, who Tim mentioned.
Brie [00:03:22] When Obama arrived in D.C., he switched Whitehouse.gov from a proprietary CMS to Drupal. But that’s all history. This started a boom in Drupal sites starting around 2009. Then there was explosive growth up until about 2015.
Blake [00:03:37] And if you look at builtwith.com, which tracks the technologies used to build a web site, you can see Drupal’s usage has leveled off since then. But a lot of the initial growth took place in Drupal 7, which was released in 2011.
Brie [00:03:50] And all this season of the Agile Podcast, we’ve been talking about Drupal 7 and it’s retirement and how it will affect people powering their websites with the CMS. And if you look at where Drupal has been since Dean Space, you see that this is a critical moment for Drupal. They’re doing a big overhaul with the move to Drupal 8 and 9.
Blake [00:04:10] I asked Tim about this transition. Here’s what he said.
Tim [00:04:12] Drupal 7 was probably the most important milestone in Drupal’s history. It was the period of the most explosive growth for the project. It’s the version that I think the most people are familiar with that powers still today quite a lot of the most significant Drupal sites that are out there in the wild. So, I mean, it’s a really big deal. And it was a really important piece of software. That said, you know, the first version of Drupal 7 did come out in 2011. It’s been aging for a while. And even though it’s end of life has been extended into 2022, the writing is a little bit on the wall. And I think that there is opportunity to take advantage of some things that changed in the architecture of Drupal going into Drupal 8 and continuing through Drupal 9 and 10. So with everything that Drupal 7 was able to do, Drupal 8, I would say kind of refines those things and advances them in a few significant ways. Like, for example, views is probably the most essential feature of Drupal. And it was only a contrib module in all of Drupal 7, right? It wasn’t architecturally foundational. And it became that as part of Drupal 8, the same way that the rest APIs and the kind of API first principles of development became fundamental to Drupal in Drupal 8. So a lot of those elements that took Drupal and went from, hey, this is a super advanced CMS to this is a framework that is sort of a CMS at its core, but can also be your in-flight entertainment, can also be your headless solution for various multi-channel kinds of things, can also be all these other kinds of web applications, became codified in Drupal 8 and then that’s continuing through 9 and 10. So, there’s this hard upgrade path between 7 and 8 that’s still still troublesome. But fortunately, the 8 to 9 path is much easier.
Brie [00:06:12] So that’s the pitch, right? That this upgrade is tough but that new versions of Drupal will be easier to migrate to.
Blake [00:06:18] Right. And D8 is better for the user than D7. More contrib modules in the core, object oriented programing.
Brie [00:06:24] OOP?
Blake [00:06:24] OOP. But we’ve talked to some of these people who just can’t stomach the cost of a million dollar migration on a Web site that might have cost a tenth of that when as first built in Drupal 5, 6 or 7, which put up this question ahead about who Duplass for.
Brie [00:06:38] Yeah. The question of enterprise sites. This move to Drupal 8 and 9 is too expensive for small sites? I mean, does that only leave the big budget sites?
Blake [00:06:48] Right. Well, Tim and I touched on a lot of this in our conversation. Here’s my interview with Tim Lehnen chief technology officer at the Drupal Foundation. I started by asking him about who Drupal is for.
Tim [00:06:59] You know, the way Dreis, the Drupal Project Founder, likes to talk about it is that Drupal is for ambitious experiences. And he specifically uses that word rather than enterprise for, I think a good reason, although it’s a little bit hard to track it first. So I’ll do my best to explain it. If we’re talking about enterprise experiences, usually what we mean is sort of like large corporate, big I.T. team, lots of data kind of scale and sure, Drupal is great for those environments. But ambitious can also mean a startup trying to do something innovative like use Drupal as a content store for both an retail outlet on a normal Web page and also kind of an augmented reality display inside a storefront. Right. That doesn’t necessarily mean it’s a big enterprise solution. But it’s an unusual and ambitious idea.
Blake [00:07:52] I mean, it’s not for the masses. You need to have a little bit of tech saavy.
Tim [00:07:56] Yeah. I mean, I would say … We were we were afraid of saying this for a long time, and we shied away from saying that Drupal wasn’t for, you know, insert a persona here. For a long time we clung to the idea that it was for anybody could do anything. To a certain extent, that’s true. But Drupal wouldn’t be my choice for making, you know, grandmas quilting store. And it wouldn’t be my choice for making, you know, just a restaurant menu site or those kind of brochure level, relatively simple experiences. You know, there’s the Wixes and the Squarespaces and the Weeblys of the world. And for a use case where there isn’t some sense of that technical ambition, where it’s like you need a web site just because you need one, not because it’s the core of that work you’re doing. Then maybe those are better choices. But when you’re really thinking about being digitally present with what you’re doing, you start to care about more sophisticated features, users and permissions content relationships, different sorts of channels for displaying that content. And that’s where Drupal comes into its own for sure.
Blake [00:09:00] So here’s the big deal for a lot of organizations. I talked to a guy from Lullabot and he was serving a client since D6 and he was telling me that I think the D6 web site cost about $50,000 when they migrated to D7, it cost about $100,000. And their estimate to migrate from D7 to D8 is about a $1 million. And the client just can’t. They just don’t have the money.
Tim [00:09:26] Yeah, it’s a tough time because there’s the constraint, because budgets are frozen everywhere. There’s the constraint of thinking, well, you know, what I bought 15 years ago, is that what it should still cost now? My hope, and I’m almost certain this is true, is that, you know, as much as the prices, you know, 10xed or whatever compared to an earli er version, the sophistication of what they’re doing and hopefully the data and the user interaction that they’ve built in that intervening time is also 10x right. So, you know, my expectation is that that different and cost is not just, hey, the implementations harder. It’s also what you’re doing with a digital experience on the web is probably more complex.
Tim [00:10:06] And just vastly even driven by more data. That said, yeah, it’s tough. It’s hard to look at the Drupal 7 to 8, 9, beyond upgrade and think, oh, this is just gonna be easy because it’s it’s not necessarily. I do think that it’s still worth doing. Right. You do have to consider it, in some ways, you do have to think about it more like a replatforming than you do just an upgrade. And that’s what gives people so much pause. I know that’s part of what makes it so difficult. But there’s so much value in the feature set that was enabled by the architectural changes in Drupal 8 by the development model, changes that mean that we get new features every six months instead of years and years in between that happened towards the end of the 7 cycle. Being on board that train with 8 and beyond just so worth it. And the other thing about it is, despite having to kind of tear off the Band-Aid of the 7 to 8 upgrade, 8 to 9 to 10 and probably to 11 and beyond, those are, in fact, easy updates. Like once you get there, there’s a commitment from the leadership of the project and there’s a technical path to making sure that those things continue to be easy in the future. You know, like any project of that scale, you’re going to have to analyze the pros and cons and understand your use case. But kind of the best thing I could say is if you were starting from scratch, if you were just about to launch this for the first time, if you weren’t upgrading Drupal 8 or Drupal 9, should still be probably your first choice or certainly in the first three that you’re considering anyway.
Blake [00:11:43] I guess there’s a couple organizations that just didn’t have the money to get out of D7 and just like D7. And so there’s like this fork of Drupal 7 called Backdrop CMS. I’m wondering If you ever heard about it, what do you think of it? What do you think of stuff like that?
Tim [00:11:58] I am familiar with Backdrop CMS, it was started by a couple of the Drupal contributors who are definitely heavily involved. I think Nate and Jen Lampton in particular sort of led that effort. I think in a lot of ways it came out of some of the frustration with the delays in Drupal 8 and a little bit of the anxiety about whether that kind of major architectural shift that happened from 7 to 8 was gonna be worth it, was going to pay off. It’s kind of like the Drupal 7.5 Kind of thing. It’s sort of that way to sort of stay on a Drupal 7 like thing. I think it seems sort of fine from that point of view, for those folks who who use it. Personally, I’m not a huge fan of open source software forks. I mean, I know that’s a fundamental part of the freedom of using open source, which I fully support. But, you know, one of the things I value so highly about the Drupal community is the continual collaboration and everybody kind of working in one place. So for me, I feel like the things that it genuinely provided was continuity with the user experience that people were used to and with the software that people already have in a relatively easy upgrade path, I think that five years later there are some migration tools, there’s new user experience innovation, and there’s new features that would make it worth, you know, sort of sticking with original flavor Drupal or choosing original flavor Drupal, even if you’d gone to Backdrop for a while. But it definitely filled the niche for some of those folks who just felt kind of stuck between the old way and the new way a few years ago, for sure.
Blake [00:13:38] It’s kind of interesting because we all witnessed kind of like the D7 being pushed back. And I’m wondering if it’s because, like, I mean there’s so many, like big D7 sites that are still out there and like D8s been out for like five years, but they still haven’t felt compelled to move. What do you make that and if that had any influence on the decision to extend D7?
Tim [00:14:03] One of the issues, I think, that slowed down the upgrade process id D8 was admittedly delayed, like multiple years delayed in its update process. And then even on launch, a lot of the contributed module ecosystem wasn’t caught up and wasn’t ready for Drupal 8 yet. And so there was a lot of things that I think would make someone who has a perfectly stable, successful, functioning web site feel like, OK, I don’t really need to do this now and I’ll just let it bake and let it mature. And I think that was the mindset that a lot of people were in that kind of carried them through the first couple of years of when 8 was out. And then, you know, momentum is easy to maintain. But that isn’t, I would say, why we extended the end of life. It’s related. But as you said, D8s been out for five years. It’s been a worthy product for a long time.
Tim [00:14:56] People have had all sorts of warning that this is coming. I think we really only extended it because we knew that the Covid situation would freeze budgets across all these places. So even the people who had it in their plans, OK, finally, we’re gonna go ahead and upgrade. Maybe we’ll go straight to 9. Like, a lot of those people we know are going to have to push that back by a year or something like that. So, that was sort of the main reason to do that. The other thing is, you know, that caution because of the delays and the readiness of the ecosystem, was reasonably well founded. And I think the community and the project had to earn back some of the trust in what we could produce, and what I’m very happy about is that after 8’s release, when we started doing the six month point releases of the software, we’ve never had a late release. And for Drupal 9’s launch, it was something like 60 percent of the top five hundred contributed modules were ready the day it launched. We had people upgrading on Twitter on launch day and saying, cool, 30 minutes later, I went from 8 to 9. And it’s all there. It’s all working. So that’s been really, really cool because I think we learned a lot from this kind of stumble around that that transition, that tearing off the Band-Aid process in the first part of 8 and really got over that and made things a lot better in this 9 process.
Brie [00:16:21] That was an interview with Tim Lehnen, chief technology officer at the Drupal Foundation. And what really struck me about that was Tim mentions letting go of certain parts of the CMS market, saying that brochure web sites and smaller sites shouldn’t really use Drupal or they aren’t best suited for it. It’s too complicated. It’s too expensive. It seems like it might be a change in Drupal’s direction and seeing the CMS better define its target audience, unlike, say, WordPress, which is trying to be the CMS for everyone. Drupal is trying to create a niche for itself.
Blake [00:16:57] Yeah, I guess it’s kind of like shooting a mosquito with an elephant gun.
Brie [00:17:00] Yeah, kind of like that?
Blake [00:17:01] Maybe, although. But I do have to say a couple of things. I think, first of all, the big effort, the big cost, the big move is really going from D7 to D8 or D9. But if you were building a website from scratch, say that you’re building some brochure web site from scratch in D8 or D9, I don’t think it is that much bigger of a lift than doing it in WordPress. So I think there’s a little bit of a caveat. So I mean, I don’t think he’s turning away these smaller, you know, opportunities necessarily, although I think it does require a more, I guess, tech savvy, sophisticated developer to do Drupal. So I suppose the cost will be higher. But I mean, I guess you can’t be everything to everybody. I think you have to do a niche.
Brie [00:17:51] And then there’s Tim’s take on content management systems like Backdrop, which was a fork of Drupal 7. And Tim believes that forking takes people out of the community and that is generally not what he wants. But he is also really involved in supporting the open source community of Drupal. So that makes sense.
Blake [00:18:09] Well, I don’t know. I mean, it sounds kind of ironic because, I mean, I don’t think Dries built Drupal from scratch. I think he forked some other piece of software and created Drupal. And likewise, I don’t think Matt Mullenweg created WordPress from scratch. I think he forked something that became WordPress. So, you know, it doesn’t it doesn’t bother me and it shouldn’t bother him. I mean, Drupal is so well-known and so widespread that he shouldn’t feel threatened by Backdrop CMS. So, I mean, I’m okay with that. You know, it doesn’t bother me.
Brie [00:18:48] So at the end of the day, for lots of sites, the move to D8 or D9 is enormous. And I understand that it’ll be easy to go from one version of people to the next. Now that Drupal is built on the Symfony framework. But that still doesn’t put my mind at ease in terms of the next migration, which Tim expressed a lot of sympathy for. Which I think is all he can do. I mean, moving out of D7 is important, but there’s no masking that its an enormous effort.
Blake [00:19:17] Yeah. I mean, we’ve heard stories of people where basically it costs like, what, $50,000 to build the site in D6. $150,000 to build it in D7 and $1.5 million to build it in D8. So I mean, that’s huge. But I tell you what, that’s why the Backdrop CMS is really attractive, because, you know, if you’re already in D7 to go from D7 to Backdrop is probably tens of thousands of dollars, but going from D7 to D8 or D9 within, that could be hundreds of thousands of dollars. But you’re not getting all the new benefits of the D8, D9 framework. So, you know, for a fraction of the cost, you can kind of like stay where you are. You shouldn’t feel compelled to move.
Brie [00:19:59] Right.
Blake [00:20:00] Brie, you run a D7 site and you have to move. How do you feel after listening to Tim talk about the future of Drupal?
Brie [00:20:07] I mean, I feel a little bit more confident knowing that Drupal has a direction and it fits with my organization’s mission? I mean, that’s all great. We’re doing ambitious work and we need a Web site that matches that. And Drupal is committed to that.
Blake [00:20:22] Yeah, well, my big question is that, you know, does your CFO have any idea how much it’s really going to cost?
Brie [00:20:30] You know, I bet they have some background experience with the web site migration costs. But as we’ve discussed before, it completely varies. You know, like it’s so hard to predict exactly how much it’s going to cost. So, I bet they have some relative idea, but it’s probably like a wide uncertainty error band if you know what I’m saying.
Blake [00:20:53] Yeah, well, I do know what you’re saying, but here’s here’s my personal experience is that if you’ve never migrated a Web site from D7 to D8, you have no idea how much it’s going to cost. And so you’re just basically taking a wild guess. You know, and so it’s very likely that you’re going to underestimate and and the problem is that, you know … Until this podcast came out, I mean, I’ve never heard any real numbers about the real costs of migrating from D7 to D8. So the risk is, is that you’re going to tell a client. OK, it’s going to cost you. You know, I don’t know, $250,000 and then you sign a contract and you get halfway through it and you realize this is not 250. This is 750. And then you have to go back to the client. You’re going to say, I need another five and a thousand dollars. And not every client has that kind of money. It may take another another fiscal year to get that kind of money. And you’ve totally destroyed your reputation with that client because now they know that you had no idea what you were talking about.
Brie [00:21:55] Right. Are there any other concerns you have about Drupal going forward?
Blake [00:21:59] Well, I mean, not really. I mean, I’ve started learning Drupal 9. I like it. It’s easy for me and I’m not really what I would call a developer. I mean, I’m not a computer science graduate. So, like, I’m getting the hang of it. I’m learning it. I’m understanding it. I do think that the you know, when you look at Google Trends, I do see that, you know, Drupal … the search for Drupal over the past 15 years is dropping down to about the point where it was introduced 15 years ago. So I think it’s becoming a very, very niche product, I guess mainly for, you know, government agencies and universities. I think Drupal is perfect for government agencies and universities compared to like alternatives, like proprietary alternatives by Microsoft or Adobe. I mean, Drupal is still pretty cheap, but it’s still pretty powerful. So, I mean, I think I think it’s got its use. I think it’s got his niche. And so I think it will I think it will survive it. I think it will continue.
Brie [00:22:58] I guess my last question is, is it niche (nitch) or niche (kneesh)?
Blake [00:23:03] Yeah. Well is it raison d’être or … I don’t know.
Brie [00:23:06] It’s that. I like that one.
Blake [00:23:08] I don’t know.
Brie [00:23:11] Well, there’s one more episode this season of the Agile Podcast. So stay up to date by subscribing to the Agile Podcast on Apple podcasts or wherever you get your podcasts.
Blake [00:23:22] And please leave a review while you’re at it. Thanks for listening.
Brie [00:23:26] Until next time.
Blake [00:23:27] Bye-Bye.
Blake [00:23:31] The Agile podcast is produced by Agileana, a DC based web development company. To learn more about Agileana, you can visit our web site Agileana.com. Or send us an email to email@example.com