Interview with Matt Mullenweg on Gutenberg, WordPress, and the future
Welcome to the Post Status Draft podcast, which you can find on iTunes, Google Play, Stitcher, and via RSS for your favorite podcatcher. Post Status Draft is hosted by Brian Krogsgard.
In this episode, I am joined by Matt Mullenweg, the co-founder of WordPress and CEO of Automattic.
Just after releasing WordPress 5.0, and on the heels of WordCamp US, Matt and I review the event, the release, and discuss how he thinks things went, what could have gone better, and what he sees ahead.
We also dig into WooCommerce, various plans around core development processes, Automattic, and more. I hope you enjoy.
And an audio version.
Full transcript is coming soon.
- Matt's blog
- WordCamp US State of the Word
- Post Status's coverage of WCUS
- 5 Stages of Grief
iThemes makes great WordPress plugins, themes and training to help take the guesswork out of building, maintaining and securing WordPress websites. I talk to iThemes CEO Cory Miller during the break to hear about what they are working on, and excited about for the coming year.
Thanks to iThemes for being a Post Status partner.
Brian Krogsgard: Hello and welcome to Post Status Draft. I hope you enjoy this interview with Matt Mullenweg. It lasts for a while, but I think that it's worth catching the whole thing. We cover a whole bunch of topics. I want to thank iThemes for being our partner for this episode. iThemes has been in the WordPress space for so, so long and they make many great products, whether you're looking to secure your website, manage your plugins and themes and so much more. Go to iThemes.com to learn more and thanks so much to the team at iThemes for being a Post Status partner. Now here's the show. So hello and welcome to Post Status Draft. Fresh off of the release of WordPress 5.0 and WordCamp US, I'm here with Matt Mullenweg, the co-founder of WordPress. Hey Matt.Matt Mullenweg: Howdy, howdy. It's good to see everyone again.
Brian Krogsgard: Yeah, it's great to see you. You look comfortable. You must be somewhere that is something like home.
Matt Mullenweg: Yeah, I'm in actually New York City this week. I had a few meetings after the WordCamp and it's also just one of my favorite cities to be especially this time of year. The lights and everything around Christmas time are just gorgeous, so it's fun.
Brian Krogsgard: Yeah. I've never been to New York City in the Christmas season, but it's one of those bucket list type of things. So let's start off with just how are you feeling post-release, post-event? How do you feel like things went overall?
Matt Mullenweg: Super energized. I'd say like day before the release was pretty nervewracking, but it went better than I think almost anyone hoped, including myself. There was a great energy at WordCamp of we were able to kind of … I think instead of arguing about whether we should do a release or not, should it be Thursday or Tuesday or next month or never, or all these sorts of different options, everyone would say, “Okay, it happens. The world didn't end and what can we do next to make things better?” That kind of making things better conversation is where I find the WordPress community is most productive. It's where you can really get all sorts of different ideas and you can really get to the point where it does make things better for our users.
Brian Krogsgard: Yeah. And I tend to think generally people cool down when they're in person versus when they're online. So all the energy, the pent-up energy from working through all the things that have to do with the release over the course of two years and it finally comes on you, sometimes it just feels good to get it out at the end of the day. You let those bygones be bygones.
Brian Krogsgard: I felt like it was actually a pretty drama-free release and event. I didn't see any major bugs come across. I asked some of the big hosts and some of the big plugin companies, I was like, “What are you all seeing? Are you seeing a bunch of stuff come through?” And I didn't get much. Have you heard much feedback in terms of, are there a lot of bugs coming up? Are there are a lot of issues or has it been fairly smooth?
Matt Mullenweg: Well, what I've been telling you all year, that it would be anticlimactic, I was half right. The software was anticlimactic. Everything around the software was pretty intense.
Brian Krogsgard: Very climactic, yeah.
Matt Mullenweg: But yeah, I actually didn't realize it before I got on for State of the Word, but host site GoDaddy, others have already upgraded over a million sites. They're seeing normal support loads, things like that. It's like I said, ultimately what drives a release day is, is the software ready? And the software was ready and all the inputs, all the data, all the sort of normal things we look at were there. It was just one of these things where the people, we didn't seem ready and that's much harder to navigate. Sometimes getting it out there helps a lot, but it's one thing that I've seen throughout my entire professional software career is the longer you go between releases the harder they are. I've seen this in small teams, big teams, public, private, inside automatic, outside, whatever it is. If you can get to where you're releasing more frequently, it just removes a ton of stress and burden on each individual release and I think makes for ultimately a much better process for everyone involved.
Brian Krogsgard: Yeah, and I want to dig into more about the process. I was really surprised. I don't think I saw a single person share your old blog post about the version 1.0 and the importance of shipping 1.0. Maybe we're just getting really old. So referencing a blog post that's that many years old is harder now. But yeah, I've been in the camp for a while now that it just seemed like you've got to get it out there. I felt like we always had an advantage in the WordPress landscape and this is one of the things I was worried about when you announced the big feature release style. Here are our three goals and we're going to make the releases strategic to that. We had the benefit of iteration when we had the regular releases. So overall do you feel like the tradeoff of losing the regular major releases was worth it in order to get the big thing shipped?
Brian Krogsgard: I think of like Drupal, the struggles and the boundaries they had going from 6.0 to 7.0 to 8.0 and it was almost like a whole new thing. This was really our first time, in a really long time at least, maybe ever, that you feel like you update and it's in some ways a whole new thing, and it was a long release cycle that led up to that. So do you think that tradeoff was worth it or do you think there was a tradeoff? Is my assumption a poor one?
Matt Mullenweg: Well, there was in that we didn't do a major release for a year. We did slip some things that would previously be in major releases in the minor releases, in the 4.9.X kind of releases, so that helped. That's a good question. I think that in 2019 we can do both. So we can keep some really super rapid public iteration and the feature plugins, and who knows, I'd actually love to see a few more kind of areas of WordPress that we want to innovate in or expand in, take more of this plugin approach. And then we also have a lot of great stuff in the hopper for some major releases. I think we'll be able to do … there was a lot of talk at contributor day as we were looking at it.
Brian Krogsgard: Yeah, talk about ServeHappy. I mean, for I don't know, five, six, seven years, people have wanted, demanded PHP upgrades and it was a footnote in this state of the word.
Matt Mullenweg: I should have done it as one more thing.
Brian Krogsgard: One more thing, yeah. Had to really get all the developers on your team for that one. So you made a commitment to in 2019 have a staggered upgrade process to mandate certain versions of PHP. Remind me if I'm incorrect, 5.6 in the Spring and 7.0 next winter. Is that basically the timeline?
Matt Mullenweg: Yeah, that's the idea and we're just going to go through. And again this is, just like with Gutenberg, it's not about changing a version number or something. It's really about bringing the whole community along for the ride, so we need to scan all the plugins. For 5.6, I think we'll actually be able to update them or send patches to maintainers, even update and maintain plugins, because those changes are pretty easy. Seven is tougher and I think that is, we're targeting December for that. I could see that slipping, just depending on whether host in the meantime and everyone else is able to get there, but even getting to 5.6 opens up a ton of possibilities, including with things like OAuth and GraphQL and namespaces, and there's some pretty cool features. 5.6 is ancient at this point. It had a lot of improvements over 5.2.
Brian Krogsgard: And a lot-
Matt Mullenweg: And we still have a lot of folks on these older versions, so part of the reason we're giving ourselves time as well is to, if we can run some analytics to figure out which host these old sites are on Ron and reach out to them, try to work with them.
Brian Krogsgard: Do you think a part of the reason why you even think this is possible, for a long time, I mean, people were running WordPress across a huge spectrum of versions of PHP. Do you think part of the reason why you even think this possible is the hosting relationships that had been made between the community and the hosting companies over the past few years? Because that's something that I've felt like has been quite significant. The efforts that, if you look at the big ones, the GoDaddy and the Bluehosts of the world, the amount of commitment they're giving to the community, do you think that's a big part of it? And how hard is the long tail going to be to capture, do you think?
Matt Mullenweg: That's a big question. First, I'll say that the improvement of hosts, particularly the big ones in becoming kind of like WordPress first, providing a really first-class WordPress experience from security updates, everything, it's probably been the biggest change in the WordPress ecosystem over the past five years. And I have huge admiration for them because they're doing this for millions of sites, thousands of support folks. They're really moving some really, steering some really large ships and the results have been fantastic. We got, I think it was up to 68% of our WordPress was on 4.9 by the time 5.0 come out.
Brian Krogsgard: Yeah, that's great.
Matt Mullenweg: But in terms of what's going to make the PHP upgrading easier, it's really a combination of everything, including time, including some of this white screen protection stuff that we're doing. If we can back port that, you know we still backport to 3.7 all of our security updates?
Brian Krogsgard: Mm-hmm (affirmative).
Matt Mullenweg: Which is kind of wild.
Brian Krogsgard: It is. It's great for security of the Web broadly, but it's quite something.
Matt Mullenweg: Yeah, if we can do that for some of this, maybe we can put some of this white screen upgrade protection into those old versions, so that way we can kind of protect people who might be doing plugin and theme or core updates from breaking their site in programmatic way. Um, it's also the work that's gone into updates. We know we migrated to a new data center this year for WordPress.org It's like, I forget the exact or something like five or 10 times more powerful than the old one.
Brian Krogsgard: Wow.
Matt Mullenweg: Almost no one noticed. But you notice that there's no more 503s when we run the updates, and things can go faster. This means we can do auto updates better, which means that we'll be able to do opt-in for core and plugin and theme auto updates and then maybe move that to opt-out 2020. These are all just things that have been going on that represent thousands of hours of work from hundreds of people that set the stage so we can do stuff. It's part of the reason I think that we will be able to move a lot faster in 2019. We had to stop core ,essentially freeze it for Gutenberg, because we were building a new foundation and you can't build the house on top while you're also pouring the concrete, right? Things will sink. They'll get off center. I've never built a house so I don't know this analogy all the way, but now we've got the foundation. We've got this framework and things being built on top of that framework. We're already starting to see the benefits to things like keyboard navigation and accessibility when they're built using Gutenberg fields that already have this kind of baked in.
Matt Mullenweg: So I think that we have the opportunity to do just a massive upgrade not just of the editor but of every single plugin, every single theme, every single person's experience with WordPress because think about it. No one just runs core. Cool if we do Gutenberg in core? Ultimately it doesn't matter. Everyone's got five, 10, 20 different plugins. They've got their theme which might implement who knows how many additional features baked into the theme? Probably dozens, if you think about all the customization options, that's all going to be able to be reimagined now and already is. We see some of this, over 100 themes in the plugin directory. You might have noticed we put the Gutenberg-ready plugins at the very top, even above featured. We're really trying to get everyone on this train.
Brian Krogsgard: Do you think there was a messaging gap in a trying to get across the idea that Gutenberg is a foundation rather than a window dressing or wallpaper or whatever other nonstructural housing component we want to compare it to?
Matt Mullenweg: Well, I think it's easy to say that there was a messaging gap and that communication could have been 10 times better or whatever, because there are obviously some people who wren't on board and still aren't on board. So now, is there anything that could have gotten everyone 100 percent?
Brian Krogsgard: No.
Matt Mullenweg: I don't know, but I think we will always strive to do better. And ultimately, I mean, one of the things that made me pretty confident about the release wasn't that Gutenberg is perfect. It's not. By the way, it's still not right for a ton of people, but that's okay. You know WordPress, unlike almost every other SaaS service or software use gives you control. And the fact that we have classic editor as a first party supported official thing and that anyone for whatever reason, literally any reason, can still use the old experience instead. That's what I meant when I said 5.0 was going to be anticlimactic, not that it meant Gutenberg was perfect or ready for everyone, but that forever was not wordy. They're just a click or two away before or after the release from having the experience that they used to.
Brian Krogsgard: Yeah, one of the things I like to remind myself of is that neither was TinyMCE perfect or right for everyone.
Matt Mullenweg: Well, no.
Brian Krogsgard: And that's why it was being replaced. So I think sometimes we do have to look at what are we trying to replace and what do we have right now? Is that worth getting out into the world and iterating on?
Matt Mullenweg: Did you see those user tests videos?
Brian Krogsgard: I did. Unfortunately my wife said, “Do you want to go see the Harlem Globetrotters?” And I was like, “Absolutely.” And I didn't have WordCamp US on my calendar yet.
Matt Mullenweg: Oh, don't worry, don't worry.
Brian Krogsgard: So we double booked, so I re-watched the whole State of the Word, including the user tests, the Q and A, the whole thing, and it was pretty amazing seeing those user tests. I've seen people do user tests before. They're often really terrible.
Matt Mullenweg: Painful.
Brian Krogsgard: That's why for years I've tried to make my own message that we need to continually support things that make WordPress easier to use, so I think that was a good thing to highlight in the State of the Word.
Matt Mullenweg: And by the way, I probably would've picked the Harlem Globetrotters over me as well, so o need to apologize.
Brian Krogsgard: It was the first one I've missed since like 2012.
Matt Mullenweg: That's awesome. That's great. I'm an average live experience. The Harlem Globetrotters are amazing live.
Brian Krogsgard: They are.
Matt Mullenweg: You can always catch State of the Word on vide later.
Brian Krogsgard: They are. I may as well use the opportunity too to thank [David Bisset 00:15:39], because he did a fantastic job with both tweeting things and just generally keeping people that follow along with Post Status up to date and it turned out to be a good outlet for him because twitter banned his personal account because they thought that he was tweeting so much.
Matt Mullenweg: Wasn't that wild?
Brian Krogsgard: Yeah.
Matt Mullenweg: I know. That guy is a national treasure of WordPress.
Brian Krogsgard: He is. He is. If there was a community spirit award, he should win it.
Matt Mullenweg: Absolutely.
Brian Krogsgard: So he stood up to ask a question during the State of the Word and unfortunately ran out of time, so I would-
Matt Mullenweg: Right, I felt so bad that we didn't get to … There were a lot of people in line, but just some of those earlier questions were a bit longer and I could tell the room was starting to, like I think we were already like an hour 45. We were pretty far into it, so …
Brian Krogsgard: Yeah, and maybe we've put a 30-second clock next year for people to ask a question.
Matt Mullenweg: I've been thinking about that, like there were some definitely things in the room this year, like it was very dark in the audience so it was a little harder to see. I think that also subdued people. There was a little more distance between the stage and the people which looked a little more theatrical, but I think also made the audience a bit more disconnected. And even, there was a funny thing. You saw the spotlight on the question askers?
Brian Krogsgard: Yeah.
Matt Mullenweg: It was so bright and at an angle that it was shining in the eyes of everyone sitting next to them. So after the first or second question there, there was a huge hole in the audience next to the mic because we were just getting like it had this crazy spotlight on it. I think we can figure out some different ways of doing lighting or doing the questions in the future that might be a bit a bit easier.
Brian Krogsgard: Well, you'll have that opportunity in St Louis, right, for the next two years.
Matt Mullenweg: Woo hoo, St. Louis. I'm very excited about that.
Brian Krogsgard: Yeah, I'm excited about that too. I haven't been there in a long time, and what David-
Matt Mullenweg: I love that we're doing the opposite of Amazon. They had the opportunity to pick any city in America and they go to D.C. and New York.
Brian Krogsgard: I know.
Matt Mullenweg: They're both amazing places, but not really shining a light on places that don't get as much love already.
Brian Krogsgard: Yeah, I think my-
Matt Mullenweg: And we've had Philly, Nashville and now St. Louis, I really dig the WordCamp US thing.
Brian Krogsgard: Yeah. I have feeling that Amazon may have been including some political reasoning and in their strategy for that, but that's okay. WordPress doesn't have to do that. We can go wherever we want.
Matt Mullenweg: We can go to Canada, which was one of the questions, which I thought was kind of cool.
Brian Krogsgard: Including Canada, which was a good question. I actually had that brought up to me before WordCamp about the U.S./Europe thing and whether we can learn from that and have WordCamp Americas or something.
Matt Mullenweg: Yeah, what's your thought on Central America in that equation?
Brian Krogsgard: Yeah, I mean I think that there's a lot of cultural differences, but I think anything Americas, there's some common themes of people that work hard, people that explored many centuries ago. So I don't know, I'm probably not the best person to answer that type of question, but …
Matt Mullenweg: I've been thinking about it a lot, like what are the logical divisions and because we do have some rules in the WordCamp world around like what could be … I'd love to see a WordCamp Asia, for example. Asia is obviously geographically, culturally and language-wise, incredibly diverse.
Brian Krogsgard: Right, I think the lingual differentiation would be a challenge, so I don't know what the best regions would be but perhaps-
Matt Mullenweg: But then look in Europe.
Brian Krogsgard: Yeah.
Matt Mullenweg: It has a ton of languages [inaudible 00:19:08] English.
Brian Krogsgard: I always envy friends in Europe because they all speak like five languages plus, and some of them speak better English than I do.
Matt Mullenweg: Oh, for sure. Have you ever seen a map of the lower 48 overlaid over Europe?
Brian Krogsgard: Yeah, it's very, very similar in size, so-
Matt Mullenweg: Yeah, it's actually, you forget how large actually the lower 48 of the U.S. is that it covers kind of like into Europe, pretty far into Russia and Turkey and Middle East.
Brian Krogsgard: Yeah. We think of Alabama and Houston or Birmingham, Alabama and Houston, Texas as pretty close, but that could be different countries if we were in Europe with different languages and culture. It's a reminder of the diversity in our space and how serving different cultures and people from different backgrounds is extremely important. All right, so David's question was, what are the dates for WordCamp next year? Do we have any idea?
Matt Mullenweg: Oh, I think we know them. I'll put them out later. They'll be in a blog post or something.
Brian Krogsgard: Okay, he will be disappointed again.
Matt Mullenweg: Well just, I don't know if I could tell you the day of the week of my birthday or something. I'm pretty bad at dates.
Brian Krogsgard: No, that's fine.
Matt Mullenweg: As you might guess from-
Brian Krogsgard: Is it going to be in December again, though?
Matt Mullenweg: I believe so. Don't hold me to anything there. I will officially find something.
Brian Krogsgard: But late in the year still.
Matt Mullenweg: Yeah, yeah, so I think that we'll put up like a placeholder site for 2019, and there will definitely be some tweets and some official stuff saying the dates.
Brian Krogsgard: Okay.
Matt Mullenweg: But to your earlier point, I actually feel what you just described of going between countries in Europe, I feel that when I'm, excuse me, on the East Coast of America because I can actually drive eight or 10 hours and still be in Texas. I can drive two hours over here and cross through three different states. And that was one of the points that I tried to make. This was actually I think one of my weaker Q&As. I had two answers that were two of the worst answers I've given probably the past few years, and one was what people heard me say, which I wasn't trying to say, to Morton's thing.
Matt Mullenweg: Which I was trying to make the point that in a distributed organization or where we have people working across things, you can read into someone's language and it might not be their first language, so it's really good to give the benefit of the doubt and realize that we're not all speaking our native tongue. And so things might come across as harsh or worse or rude or things, when it's just not at all. It's certainly not the intention of the person to communicate that way. And the other was the answer, I think it was Birgit's question about being hired to work on things.
Brian Krogsgard: Yeah, corporate sponsorship for WordPress work.
Matt Mullenweg: Yeah, neither … I think my brain was just a little fried by the end of the day.
Brian Krogsgard: So what would you adjust in terms of, I think that question was around how to … I was actually with you with Morton, that I didn't notice the issue. It seemed like you were pretty clearly speaking about broader communication things, not specifically with his communication, which is-
Matt Mullenweg: I would never make fun of Morton's or anyone's skills speaking another language.
Brian Krogsgard: Yeah, and that one was pretty clear to me. But the one about corporate sponsorship, do you think there's a place for … How do you encourage independent work on the WordPress community, nonprofit side of things? Is there a place or a way that that can be done without corporate sponsorship? Because I know, I mean I'm self-employed and I can personally understand if I suddenly gave 10 or 20 hours a week or whatever amount of time to the core project in a non-revenue generating way that would be felt. So is there a way that that can be done, or do you think corporate sponsorship of community work is really the only way to go about it right now?
Matt Mullenweg: Well, let me take a second crack at this one. First and foremost, I never want to downplay the contributions of people doing it in their spare time.
Brian Krogsgard: Yeah, and it is, I mean it's astonishing and I could be that person, too. I don't choose to be that person in this stage of my life, but yeah, it's amazing what people are doing completely outside of work time.
Matt Mullenweg: Yeah, and I also don't want to understate how big an impact on the WordPress project, two to four hours a week out of the 168 can have a really huge impact on some big areas. Even a single hour a week, if put in the right place, can move mountains. And WordPress was largely built by volunteers and people doing it in their spare time, and I worry to say, to ever get to a point where we're saying we only value people who are doing it full time, or it's only possible to make a change if you're doing it full time. So that's one point I was trying to make, that let's never downplay the value that people can do with just a little bit of contributions. Five for the Future is just two hours a week in a work week, right?
Brian Krogsgard: Right, yeah, and that was the question [Daniel Bock Huber 00:24:12] had was if you had plans or iterations that you thought would be good for Five for the Future?
Matt Mullenweg: Oh, for sure. The second is that the WordPress Foundation is not designed to be an employment entity. Now, there's no employees. There's no HR. There's no benefits. There's no payroll. There's no anything, and going from zero to one there is actually a pretty big thing. It is designed to give grants. We can give money to other organizations. By the way, I think that there's also a big opportunity for other nonprofits to be part of this. There's no reason to say only for-profit corporations can sponsor people that work on WordPress. I would love to see more nonprofits that by the way have a lot more expertise in some of the areas that we want more contributions in, including design, accessibility, privacy, internationalization.
Matt Mullenweg: These things, there's full nonprofits dedicated to that have often sometimes very large staffs and pretty large budgets, journalism. These I think are really fruitful areas, and in fact one of the things that, I don't know if you saw Morton's talk, a lot of people I thought misread it. Maybe I misread it as well. When he was talking about open governance, my take was that he was talking about getting WordPress a seat at the table, and discussing these regulation changes and et cetera happening. I think the example last year was, there was this meeting at 10 Downing Street. Who was there? Was WordPress represented?
Matt Mullenweg: And he started talking about the Web Foundation, and I began thinking. I was like, “Wow, well, WordPress only represents a third of Websites, and not even, really. It's a third of the top 10 million. Another foundation like the Web Foundation actually might be a better vehicle to try to advocate on the open Web as a whole, versus just the people who happen to be using a single CMS.
Brian Krogsgard: Yeah, and I don't want to speak for him in terms of which audience he was trying to, or which group he was trying to represent in terms of within WordPress, or WordPress within the broader Web. I do think that there are often questions about governance of WordPress. You have long been the central vision of the project, and then delegated the execution of that vision to many other people who then lead this project, which is massive at this point. One of the things I've always asked you for several years now was, how many people do you have directly reporting to you?
Brian Krogsgard: Because one of the things I find to be important is under this model, where your vision is required, then if you're spread out so thin it requires these sprints to focus towards WordPress, rather than always being able to have kind of a constant flow, even if it's a minority of your time, going towards that, establishing that vision for WordPress. So, how have you evolved in that regard? I know you're-
Matt Mullenweg: What was the number last time you asked me?
Brian Krogsgard: It was high.
Matt Mullenweg: What is high? There was one point it was like 26 or something.
Brian Krogsgard: Yeah, it was in the 20s. It was in the 20s last year I think when we talked about it.
Matt Mullenweg: Yeah, it's under half that now.
Brian Krogsgard: That's great.
Matt Mullenweg: So, that's been really good. There's been a lot more leaders, but I would actually argue the point that WordPress has always been sort of my vision being set, or even my direct leadership. There was a good four or five years there where the leadership structure, because we've experimented with lots of different, we don't call it governance, but essentially leadership structures in WordPress. For a while, we had kind of the … It wasn't a committee approach, but essentially like the lead developers consensus approach. We did that for a few years.
Matt Mullenweg: Even from the beginning it wasn't just me. It was me and Mike Little, so it's never been solo. Then we went to where the release lead was the final decider, including over me, so that was probably, I don't know, 3.9 to 4.7 maybe, that including could overrule me as project lead for what was in the release or not, and that was to try to give a little more autonomy and flexibility to release leads. But the big change was a few years ago I said, “Okay, I'm going to take back over core WordPress development,” and that was to try to provide a … Try to make some of these big changes happen. So right now it is much more of a benevolent dictator model, although both of those words are questionable. But, I don't see that as the permanent forever structure, you know? [crosstalk 00:28:54]-
Brian Krogsgard: One problem with that is-
Matt Mullenweg: I don't know if I can sustain this forever. I mean, this is a pretty intense period and we're doing some pretty intense work, and I'll need a break at some point, too.
Brian Krogsgard: Yeah, and how do you, or do you think there's merit in opening that up in some way to certainly have more community input, but sometimes that's not good, you know? Sometimes the central vision idea can be more productive, but when you bring that into open source software versus running a company, maybe there's different consequences. So have you given any thought to what does WordPress look like after this phase in terms of a governance structure?
Matt Mullenweg: It's hard, because I haven't seen any models where anything resembling voting works, you know?
Brian Krogsgard: Yeah, for running software, you mean.
Matt Mullenweg: Yeah, and for choosing a direction. It's not that many models for consumer-facing software. There's some for server-facing software, where sort of like a committee approach works, you know? It seldom creates great user experiences. You kind of want, I'm not saying it always has to be me, but what I want is a strong, opinionated, thoughtful leader doing, setting a bold direction, taking experiments and being willing to fail, comfortable with failure, is I think what you need to create great software.
Brian Krogsgard: So how do you-
Matt Mullenweg: And that's tough, and by the way, we've had lots of those leaders in the history of WordPress, and I think that's been the success. It's not just because I'm doing anything.
Brian Krogsgard: Sure.
Matt Mullenweg: It's that we have lots of folks and we have, as a community including myself, invested that authority and power in them, and that trust. It doesn't mean that it's always been right, like we've definitely done things that we might undo later or that were mistakes, but the ability to make those mistakes is what you need for an innovative organization. When you look at any organization, private or public, it's too afraid to fail, that's when you get this kind of incrementalism. That's when you get this kind of fear of change. That's when you essentially stagnate, and if you were to point to any place that software goes poorly, whether that's in government or companies that don't innovate or get disrupted by smaller upstarts, anything, it's because one of those happened or probably all of those things happened. They got too successful and then they were afraid to fail, because it was too embarrassing and they had too much to lose, and then they stagnate.
Brian Krogsgard: All right, we're here to take a quick break and I want to thank our partner, iThemes, long-time supporter of Post Status. I've got Cory Miller on with me to talk about what they've got going on. Hey, Cory.
Cory Miller: Hey, Brian, thanks for having me on.
Brian Krogsgard: Yeah, it's my pleasure and I appreciate your support over the years. I was hoping that you could just tell me maybe what's a big success iThemes had in 2018, and what you're looking forward to next year?
Cory Miller: Yeah, as you know in January of 2018 now we were acquired. iThemes was acquired by a great hosting company named Liquid Web, so this year has been spent integrating. We are almost done with the first round of integrations for our key product, iTheme Sync, which promotes site management for freelancers and agency WordPress Websites, BackupBuddy, our long-time popular plugin, and securities. So we're almost done with the final round of that. You know in any transition there's always bumps, but it's been a really good ride so far.
Brian Krogsgard: Yeah, have you been getting good feedback from LiquidWeb customers so far, or I guess are you looking forward to seeing iThemes products more integrated with LiquidWeb's stuff?
Cory Miller: Very much so. We saw hosting as the future for WordPress. All of the big hosting companies making significant investments in WordPress, and this was a way to join forces with a great hosting company and for our products, our team, to live on into the future doing really good things in WordPress.
Brian Krogsgard: Cool, and iThemes products are still available outside of LiquidWeb as well. You can use them on any host that you choose, or a naked VPS if you want help managing a server that you want to pull all the lever, you can still do that but you can use all the iThemes tools to help you manage the day, and [inaudible 00:33:31] out of your WordPress Website. I'm really excited for you guys and what you have going on, and look forward to hearing more as next year comes along. Thanks for being on real quick, and if people want to learn more about what you've got going on, where do they go?
Cory Miller: iThemes.com, best place to click some links and see if we are offering something that you're in need of. We'd be happy to serve you.
Brian Krogsgard: Awesome, well, thanks for being such a big part of the WordPress community these last 10 years plus, and a partner of Post Status for as long as Post Status has existed.
Cory Miller: Yeah, we love Post Status. Keep doing what you're doing, Brian. Thank you.
Brian Krogsgard: Thanks, buddy. Bye. We've got another question from Ant in regards to giving people a seat at the table in a governance structure, which was seeking an update on the growth council, whether that was productive this last year, whether it happened. How can it happen going forward? I don't know anything about what's happened in the last year, so what have you learned there, and what's happened?
Matt Mullenweg: Yeah, so we created these two growth councils, one consumer one and one enterprise one, and we did these monthly meetings, video chats. And my goal with creating it was one, to bring some people together, so there are direct competitors in these meetings. It's enterprise agencies that are bidding on the same clients. It's Web posts that are advertising on the same key words and everything. And so you'd have like that as part of the discussion, and one of the goals also was to produce, improve essentially the WordPress experience.
Matt Mullenweg: Some parts were successful, some parts weren't, so I was amazed and humbled by how people came together. I think it really showed something I've tried to demonstrate with Automatic in the past, that you can both compete and work together for a greater good. People were both able, because it was a safe space, they were able to share things that were going on in their companies, things they were hearing from clients that they couldn't talk about publicly, things they were seeing, experiences, in a safe space that generated some amazing discussions. I'll also say selfishly that I learned a ton being part of these meetings, so it was really, really valuable for me.
Brian Krogsgard: And that's about how their organizations work and what's important to them, or what kind of stuff?
Matt Mullenweg: And the people they're talking to, whether that's consumers, kind of the more mass market side particularly with Web posts, but also with people doing trainings, people building small Websites, which is my kind of mental model was like, if it's people spending under $100,000 a year for a site that's the consumer group. If it's over $100,000 a year, that's the enterprise group. Not exact, but that was kind of like a rough mental model. And on the enterprise side, what are they seeing? It was so interesting as well, like most of the big agencies started building sites with Gutenberg kind of like February, March, and were launching them over the summer. Now at this point, they're basically building everything. They've been building everything new with Gutenberg for the better part of a year.
Brian Krogsgard: Yeah, I actually spoke with Tom Wilmot from Human Made and Jake Goldman from 10up both, and they've both expressed that they had an early embrace of Gutenberg and seen success with it. Somebody was telling me, I think it might have been Tom, that they showed off a demo of Gutenberg and instantly blew some Adobe experience manager style feature out of the water. Some company had spent gargantuan sums of money on the alternative experience, and they did a couple customizations with Gutenberg over the course of a couple weeks and showed them this as a prototype. It was already far superior to what their competition was putting forward, so from those-
Matt Mullenweg: There's got to be so much of that.
Brian Krogsgard: Yeah.
Matt Mullenweg: At Gutenberg, people have written bad versions of what Gutenberg is, including myself, for five years now. And now we've got this standard we can all build on. I actually had it right here. Human Made had this really cool white paper. I don't know if you've seen this.
Brian Krogsgard: Nice, yeah, they printed it.
Matt Mullenweg: It's glossy, it's pretty … I mean, these are the types of things that for enterprises are really important. PDFs, white papers, case studies, and enterprise section of WordPress.org, important for that user group. For a consumer, a lot of things that that team talked about was the importance of talking about different use cases and allowing users to self-segment over the job they want to get done, and having some guides for that, some recommended plugins, essentially hand-holding them through the process of it.
Matt Mullenweg: So I talked about what was successful about the councils. I'll tell you what didn't go as I hoped, producing things. I think because it was all often the CEOs of their companies or kind of the head folks, it was difficult for them to get time other than the meetings, in between the meetings, to just do work to produce for example a new enterprise section of WordPress.org. So actually on Saturday morning, I officially dissolved the councils.
Brian Krogsgard: Okay.
Matt Mullenweg: What I told the folks was, we're not going to schedule out these meetings for the rest of the year. I left the Slack channels up so people could keep chatting if they want, and of course I said if anyone wants to keep meeting on their own, they're happy to, or to continue the projects, and both groups said they would like to. The enterprise group wants to get the enterprise subsection of WordPress.org out. The consumer group wants to continue shipping some things, but it was a year and we didn't ship these things, that honestly probably any … It could have happened within a month or two if it had had a lot of focus from any of the people involved. It illustrated to me actually the difficulty of a committee approach.
Brian Krogsgard: Yeah, it makes me think of-
Matt Mullenweg: One of the things that came up was the enterprise. We did drinks on Saturday as a kind of ending thing. It was a combined consumer and enterprise, and one of the things is they were like, “Hey, if you had just chosen one person to be in charge of this, that person could have said, ‘You do this, you do this, you do this,' it would have shipped probably pretty quickly.” But because no one knew who was in charge and who could tell other people to do things, I think it was Jake who said, “I'm not sure if I can tell Tom, ‘You need to finish this page,' or ‘You didn't edit it.'” Who's in charge? And so that's I think unintentionally illustrated an alternative group and management style that, because if it wasn't adjusted and set up correctly, which is my responsibility, my fault, and not getting the results that we wanted. So, great experiment though, and I'm really, really glad we did it.
Brian Krogsgard: Yeah, I wonder, you see movies or you see a senate hearing or something, and you have the primary people at the center table and then you have all their deputies on the outside. They're still in a safe space, but they're the people that actually go out and do things afterwards. It's not like the senator or the CEO on that council necessarily that's implementing, but they bring other people from their organization in to help the implementation component.
Matt Mullenweg: I was kind of hoping more of that would happen. It's true, some people on the councils were contributing individually, and some lead larger organizations so they can shepherd a lot of resources as well. So, I think there's some of both that's needed in the WordPress world. Yoast and myself are probably good examples, Yoast the person, not the company, of a leader who's really passionate about areas, who also brings along a lot of their company to work on it.
Brian Krogsgard: Yeah, they have a lot of manpower there.
Matt Mullenweg: Oh, yeah.
Brian Krogsgard: That they've got dedicated to core.
Matt Mullenweg: Person power.
Brian Krogsgard: Person power, thank you.
Matt Mullenweg: Yoast had an incredible effect on Gutenberg and the 5.0 release, even though they're smaller than all of the hosts, all of everything else. I think it's a great example of Five for the Future.
Brian Krogsgard: So I have one more process question, and then I want to dig into phases two through four, but Darren Ethier I think is how I pronounce his last name, he asked what were your thoughts process-wise on using GitHub for probably the most ambitious-sized project yet with WordPress from a feature plugin or feature point of view? Do you think it was encouraging enough to potentially start moving off track in SVN? I do know there was a make WordPress post that was about this, but I'm curious your personal experience, whether you liked being in GitHub day to day or using Git in the process, or if you missed track in terms of that development?
Matt Mullenweg: I mean, I'm an early adopter of GitHub. My user name is literally the letter M, to show you how early it was, and I really like it. I think it's a great product. Having kind of a parallel process, where I've now been involved pretty heavily in something on TRACK, involved pretty heavily on something on GitHub, there's pluses and minuses. I wouldn't say either is a panacea. The thing that actually makes the real difference is much more your engineering management. It's your triage. It's your process. It's the testing. On both GitHub and TRACK, we made mistakes and had tickets that sat for weeks without anyone responding to them.
Matt Mullenweg: That doesn't have to do with the software. It has to do with the people, and I do worry the distraction of effort and time that switching source control and issue trackers and just also all the work that needs to be done to migrate all the tooling at the same time, if it would distract us from what's actually most important in 2019, which is to tighten up some of our loose parts of the software. So that's the only reason I'm hesitant to commit to saying that we're going to do anything, particularly in 2019.
Matt Mullenweg: Also, I am a big fan of GitLab, obviously. I see a lot of, I noticed a lot of comments on Helen's post for actually advocating for GitLab. All of these things are getting better pretty rapidly. GitHub was slow for a while, and now they're really picking it up. Time is kind of on our side if we switch later, because whatever we switch to will be six or 12 months better than it is today, and probably provide perhaps more of a compelling step function and user experience than is currently there, which is yes, I totally agree the interfaces are more modern and there's better tools, and I do strongly prefer Git for branching and all these other things. There's almost nothing you could tell me about either way that is an argument I haven't heard or made myself. But it's more just about a community focus thing, and what is the issues most urgently impacting WordPress users, both current and potential, today?
Brian Krogsgard: I like what you say there in terms of the urgency factor, because I think I agree with you. Personally, if I were to do something to try to actually make code changes in SVN, I would always have to go remind myself, how do I patch something, and do I have the tools on my modern MacBook to do so? Whereas by … Go ahead.
Matt Mullenweg: I was just going to say, we've built a lot of bridges between them as well. WordPress is synced to Git. You can do things both ways. We have tools for migrating patches between them. We've got official GitHub for WordPress. There's a lot of stuff out there, but there is a point when I really want to focus on the developer experience and really invest some serious, serious time into that, but we've got to get the user experience right first.
Brian Krogsgard: Yeah, I think I'm with you in terms of that priority, but long-term bringing on new people to be comfortable developing on the software, most people in their day to day are in GitHub or in Git more broadly, so even the self-hosted Git version or GitFlow I think in the long run would be good, but I like what you said there about the urgency component.
Matt Mullenweg: Well, that's one of the things I was looking at a lot, because that is one of the I think better arguments for using something centralized like GitHub, is that a lot of people have GitHub accounts.
Brian Krogsgard: Mm-hmm (affirmative).
Matt Mullenweg: Basically there was a while, I've reviewed probably over 10,000 engineering applications for Automatic. Every single one of them has a GitHub link. There's definitely something to be said for that. I didn't notice that big a difference in type or number of contributions, so it is a barrier to entry to make a WordPress.org account. It doesn't appear to be the most salient factor in determining whether people submit patches or are active in TRACK or not-
Brian Krogsgard: Interesting.
Matt Mullenweg: Active in [inaudible 00:46:57] TRACK or not.
Brian Krogsgard: I'd be curious how that ends up working if it was officially there, but nevertheless let's move on to-
Matt Mullenweg: Well, just look at the numbers, because Gutenberg was as official as anything, and the numbers of contributors to that versus the number of contributors that we have in any given core release, kind of similar. I think that the hard part is actually deciding that you're going to dive into the code. The hard part is saying like, “I'm going to take this code. I'm going to figure out the problem and then submit the patch.” And once you've been through that, making an account one place or another, I think it's probably something you'll do one way or another.
Brian Krogsgard: So phase one of Gutenberg mostly complete, or at least out the door, and you're going to be iterating on this every two weeks like you told me before. Let's talk about phase two through four, and let's start with just why do you think that the order that you established things in is the right order? Why do you think this is the right direction, I guess, overall?
Matt Mullenweg: Yeah, some of it is sequential and some of it is not. Are you familiar with the Kubler Ross Five Stages of Grief?
Brian Krogsgard: No, I'm not.
Matt Mullenweg: It's like anger, denial, depression, acceptance. You've probably seen these before.
Brian Krogsgard: Yeah, I've seen this in terms of markets. It's the same stages of grief are identified in market cycles, so like in the stock market it's a very similar thing, but I didn't know it was called that.
Matt Mullenweg: I don't have one on my desk, but I actually have some glasses that I often use for drinks that have little markers on them, and it actually has the five stages.
Brian Krogsgard: Nice.
Matt Mullenweg: We could put that in the show notes for the podcast, a link to those.
Brian Krogsgard: Yeah.
Matt Mullenweg: A common misperception, I ended up actually reading one of her books called On Grief and Grieving, and it was actually finished posthumously, so she actually passed away in the process of it. This is I think Elizabeth Kubler Ross, or I forget the exact name, but amazing book. Good to read even if you're not going to go through grief or grieving, because we all do at some point. One of the most fascinating things I took from that book was people think of those stages as sequential, and actually they can happen out of order, and they can happen simultaneously. In fact, you can go through anger and denial and acceptance and everything all in the same kind of 10-minute period.
Matt Mullenweg: That was a really fascinating insight to me just of the human condition, because we are often more complex and less sequential than sometimes our simplified systems would like to say, or we'd like to think. Development is the same way. Phase one is not stopping where phase two is starting, and I think bits of phases one through four have already happened and will continue to happen as we go through it. It's just more about priorities. We've gotten the editor experience to the point where I think it's pretty darn decent, especially for new users. Now people, if you look at people who started using WordPress in a post-Gutenberg world, of which there are now hundreds of thousands, so they've only known Gutenberg. This is that example of the support person who said, “Bring me back the simple editor,” and they meant Gutenberg?
Brian Krogsgard: Yeah.
Matt Mullenweg: What is the new barrier that they're running into? What's the new wall they're hitting? It's the rest of the site. It's customization. It's widgets.
Brian Krogsgard: So for those that don't have these phases locked down in their brain, phase two is customizing outside of the post content itself as the next point of emphasis. So this could be widgets, menus and miscellaneous content. David posted a funny picture on Twitter where you looked like you were in deep thought during the State of the Word and he said, “I wonder what Matt's thinking about right now?” You said it was probably the process of transporting menus to Gutenberg.
Matt Mullenweg: Oh, yeah.
Brian Krogsgard: Moving, I feel like this is something that's happened every few years [crosstalk 00:51:12] in WordPress, it's like the old menu system, then the new menu system, then the customizer menu system. Now we're talking about menus here, and also not calling it menus but reshaping to navigation.
Matt Mullenweg: And navigation block, I think is probably a good way to put it.
Brian Krogsgard: Yeah, it seems like we've got some debt in regards to spending time on things that then we don't use for a super long time, so how can we do the Gutenberg way of site customization and content and try to be efficient and long-lasting with that effort?
Matt Mullenweg: Well, I mean first we'll start doing it in plugins, in the Gutenberg plugin actually, just like we did for the editor. And so we can have, one of the reasons Gutenberg is where it is is because we did do a lot of user testing along the way, and it wasn't just testing mock-ups or ideas. We had the actual working plugin on actual working WordPress sites so they could run. That makes a world of difference, so we'll be able to do that for the customize, replacing widgets.PHP, piece of cake. That's just taking over the page. The customizer is a little trickier, but we can do it because all of Gutenberg was designed to be responsive.
Matt Mullenweg: Essentially, the customizer side bar is very similar to a mobile view, a narrow screen. It's kind of like instead of a wide version, it's a narrow version of that interface, so I think we can slip in there. Mainly it just requires a lot of rethinking, and there are some tough concepts that currently blocks don't have, like the idea of widget visibility, which is [crosstalk 00:52:49].
Brian Krogsgard: Yeah, I was about to say conditionality, or like show it here but not there. That's one of the experiences that seems like it could get really confusing, because someone's on a post and they change a sidebar, and therefore they think they changed it just there but they changed it everywhere, vice-versa. There's a lot of UX things that seem like they could get complicated.
Matt Mullenweg: Yeah, so we've got to figure that out, to be honest. I don't have the answer right now. We've got to experiment and test our way into what works there, because there are some powerful and complex concepts that I think we can maintain, but just make a lot more intuitive and also make not so in people's face. So if they don't need to have widgets or blocks that only show up on the sidebar of search pages, it's not kind of in their face and making them think about things. It's that whole thing, making it more powerful and more intuitive at the same time.
Matt Mullenweg: But currently, blocks don't have a concept of that, but if we solve it for sidebars, widget areas, whatever you want to call it, perhaps we could solve that for blocks everywhere they're used. Showing different headers on different pages is not something that is currently supported by most themes on different types of templates, but why not? That would actually be pretty cool, different footers on different types of pages. I could easily imagine different types of sites that would do this.
Matt Mullenweg: It also raised a good point, which is kind of what has always been one of my bugaboos about front end editing, is that when I'm writing, I want to focus. I don't want to see my header and footer and sidebar and all this other stuff, much less edit them. I think, I mean I've always talked about zooming in and zooming out, Gutenberg essentially being like a new 3D interface for WordPress. I think we want to make it very easy to kind of zoom in and out, including focus on one single area or even one single block, if you really want to work within that.
Brian Krogsgard: And that means in some ways potentially front end editing too, right? When I think about Gutenberg, the one beef I have with writing in WordPress is all the other stuff beyond the editor. Whether it's TinyMCE or Gutenberg, I write mostly text and a few images. I'm not doing a lot of page layout, and I love the tools that Gutenberg's brought in. For instance, you brought up some of Nick Hamze's blocks. I talk to Nick sometimes, and he's got so much creativity and I love that it empowers that. But for 90% of the things I'm going to do, I'm just writing. A clear, focused writing area is what I like, so I actually just don't like the fact that I can go into the appearance editor and the plug in screen. Those are all one click away, rather than just saying, “Here's the page.”
Brian Krogsgard: The best Gutenberg interact I've had yet was actually WordPress.org/Gutenberg, because I was experiencing Gutenberg. It's a live demo of Gutenberg. I was able to experience Gutenberg, and none of the other stuff was there. It was just a column of content, and I really enjoyed that, so is that a part, a way that you think we can focus on how to get to writing and not worrying about everything else?
Matt Mullenweg: We can support both. I mean, you currently have a version of this in Gutenberg where it can show the menu or not, or you can go to the mode which is my preference mode, where Gutenberg takes over the whole screen so you don't have a sidebar menu there, even minimized. I actually strongly prefer that.
Brian Krogsgard: See, I didn't even know that. I've missed that.
Matt Mullenweg: Oh, yeah, yeah, so go in Gutenberg to the top right, the dots. I'm trying to think exactly where it is. Are you opening your WordPress as well?
Brian Krogsgard: I am.
Matt Mullenweg: We'll go through this together.
Brian Krogsgard: I might take over my screen share here, but that's okay.
Matt Mullenweg: Oh yeah, do you want me to screen share?
Brian Krogsgard: No, I mean I'm recording us.
Matt Mullenweg: Oh, got you, got you.
Brian Krogsgard: So people, this might end up on YouTube or something.
Matt Mullenweg: So in the three dots, there is three kind of options there.
Brian Krogsgard: Oh, yeah.
Matt Mullenweg: Tool bar, spotlight mode and full screen mode. I actually am in full screen mode. I strongly, strongly prefer it.
Brian Krogsgard: Neat.
Matt Mullenweg: Top tool bar I don't think is that useful, but they know that. This was basically, there was a huge kind of fight earlier. If you look at Yoast's original post about the plan for Gutenberg, there alternative UI is basically all about the top tool bar. Then spotlight mode I also don't love, but some people really like it. It kind of grays out the other blocks and then makes the current block you're in more visually prominent.
Brian Krogsgard: Cool, and for anyone that happens to be watching this, I just popped it up onto the screen where you can see exactly where that is, so that's really cool. I learn new things every day.
Matt Mullenweg: Yeah, and so even this menu, like I'm looking at this menu right now. Do you have it open?
Brian Krogsgard: I do.
Matt Mullenweg: I want to talk through some things that could be really improved here. So for example, originally we didn't have this descriptive text under each of these three views.
Brian Krogsgard: Mm-hmm (affirmative).
Matt Mullenweg: Well one, the text should be views, not view, and I think that there is, we can improve that text. Under editor, I don't know if visual versus code is actually very intuitive, and two, under code editor, I think it's a beautiful place to expose that you can actually go into the code editor on a per-block basis, so just a little bit of text there that says, “Hey, also per block you can go into this.” Manageable reusable blocks needs the square and the little pointy thing that shows that that's going to take you out of Gutenberg, and I think it does open a new window.
Matt Mullenweg: And I think that we should put some sort of star or cool icon next to keyboard shortcuts, because once you learn those, you are off to the races. So just a good example that as far as we've come with Gutenberg, you can look at a single screen and immediately think of four or five things that are very improvable and that are certainly worth testing and getting out there.
Brian Krogsgard: Yeah, that's neat.
Matt Mullenweg: I always joke I'm the unhappiest WordPress user in the world. It's hard for me to look at anything in WordPress and not see all the things that could be made better, and just be terribly impatient to get them better as soon as possible.
Brian Krogsgard: So, speaking of impatient, you're going to have to … Well, you might be able to work on some of these things at the same time but also there is a phase, a priority, and number three was about collaborative editing. For me, I think of this not even necessarily like me and you might collaborate on a post, but sometimes I want to collaborate with myself. My favorite tool from Automatic that you guys have built is probably Simplenote, because it's like the only to-do app-ish thing, and it's very unstructured, but it's the only to-do app type of thing that I've ever really used in depth, because it's really easy to open. It's really easy to start writing, and it's instantly on my phone. So for me, collaborative editing in some ways is just this simple process back and forth.
Matt Mullenweg: Across devices.
Brian Krogsgard: Yeah, across devices.
Matt Mullenweg: Totally.
Brian Krogsgard: Because I do a lot of stuff through my phone, but your collaborative editing is much more than just cross-device. It's also cross-individual, and people building things together. Why do you think that's an important thing to be in core WordPress experiences?
Matt Mullenweg: Hmm, that's a good example of something that there's not a bajillion people asking for right now. But just like post revisions, which is one of my favorite things I've forced to be in WordPress that I think is really fundamental to have the entire system works, this-
Brian Krogsgard: Yeah, a lot of people have been saved by revisions.
Matt Mullenweg: Yeah, I think this idea that when you're editing in a post, the post is almost like an object and people from different devices, different things, can all come and be simultaneously working on that same object, is a mental model that is technically extraordinarily difficult. But if you get it right, it's just a magical user experience. Even simple stuff, like do you remember when Netflix started saving where you'd watched up to across devices?
Brian Krogsgard: Mm-hmm (affirmative), yeah, that's great.
Matt Mullenweg: Magical, and it's so simple.
Brian Krogsgard: Yeah, I use that a good bit. We use that a good bit with our son, because the easiest way for him to be happy in a car is to have somebody's phone. You can pull up Super Wings or whatever, right where he left off, so yeah, I do agree. And even when Google Docs came out with that, it was revolutionary in a way but people didn't realize how much it would transform the way they write documents, so …
Matt Mullenweg: Totally, and we've got, the Web now supports technologies like WebRTC, that will perhaps enable us to do this even without centralized service, which is super cool, too. Because before, I was … Conspiracy theorists won't believe this but I was actually really disappointed that it looked like the only way we could do this was in Jetpack. Now, it looks like we'll actually be able to do this in a much more distributed fashion.
Brian Krogsgard: Interesting. Funny enough, I didn't even think about that conspiracy theory and everything. As delightful as that sounds to think about, I didn't consider the centralization aspect of it.
Matt Mullenweg: Yeah, so it's definitely, there's some stuff out there that's possible. By the way, there might still be something where Jetpack can enhance it, or maybe we use Simperium, which is the engine behind Simplenote, to do something even fancier, but we can do a lot without that. But this also requires us to rethink our user flows around what versions, what it means to edit something. Because right now, you and I can kind of be on different paths, and we can save each other and merge over each other. And what is the relationship between what's on the edit screen versus what's live? That also is going to apply now to the entire site, so maybe what you and I are working on together is the new version of Post Status, and you're changing the typography and I'm moving the widgets around and things like that, and then we're setting that to go live at 12:00 pm on a Tuesday.
Brian Krogsgard: Yeah, that's really cool. Another element of that that is a quirk in WordPress I guess, but you could publish a post and now I want to update it, but I don't want to update it right now, but I want to save what I've updated.
Matt Mullenweg: Mm-hmm (affirmative).
Brian Krogsgard: Right now it's either, it's all about your current session. Once it's published, you either update it or those changes are basically gone. You don't kind of save a draft of an update, so a work flow like this would inherently basically have to solve a problem like that, so that would be nice.
Matt Mullenweg: Basically, we're going to reimplement Git in the post-content table, in the WP Post table.
Brian Krogsgard: Yeah, that sounds like no problem.
Matt Mullenweg: Piece of cake.
Brian Krogsgard: Yeah, on that note, do you think there are underlying architectural changes that WordPress is going to have to go through to enable both this experience, because this is Gutenberg, but also I mean you're probably faced with some of the problems that a tool like WooCommerce has in terms of scalability and being the right tool for the job for that next level of application. These aren't blogs anymore, but we basically have the same data base, the same data base structure. You can extend it. You can change things, but do you think there's rethinking that needs to be done in that regard?
Matt Mullenweg: Sure, and I actually think that [inaudible 01:04:12] is I think we have the opportunity to readdress and in a very backwards compatible way provide a path for people to upgrade some data structures, some things that have been hanging around for a while. An obvious one is widgets are a serialized array. Now we can move that to being JSON. There's the native structure, data structure that Gutenberg uses behind the scenes. For some reason, we have to store this as like HTML plus comments, we can just store it in the native data format, which is pretty cool.
Matt Mullenweg: By the way, as we increase the MySQL version, we can actually query and get things out as JSON objects directly from MySQL. We don't have to take them in and out of strings, so there's just some clean-up, some fun stuff we can do. When we get higher PHP versions, we can name spacing doing different things. J-Trip's been talking about doing some pretty cool composer stuff. All of this is, we're getting the user part back on track and so that also means we can start to invest a lot more in that developer experience, which I think is crucial. Also I feel, I have a ton of empathy for, I actually for better or worse, I interact every day a lot more with developers than I do with users, so I really try to get out of my way to interact with users, but you can imagine the gripes and pains of developers is something that is more of my daily experience.
Brian Krogsgard: So we'll have time to talk about all these phases over the next few years, but quickly what do you see in terms of the number of sites where having a multilingual setup, and this is around the focus of phase four, how often are you encountering Websites like that, and how much do you think that this being a part of the WordPress experience would be empowering for people to actually support more languages and more language versions of their site?
Matt Mullenweg: I think it would be a really big fundamental concept. The Web is global. English or single language is actually I think a small minority. People who speak one language is a small minority of the seven billion people in the world. This is obviously very prevalent in places like Quebec, where there's officially French and English, and of course all over Europe. This is why I get the question every year for Camp Europe. I am monolingual, ashamedly so.
Brian Krogsgard: Yeah.
Matt Mullenweg: But the world is polyglots, and that's why I named the mailing list WP Polyglots, not WP translate or WP anything, when I set that up almost 15 years ago now. Even for monolinguists like me, I got someone contacted me the other day saying, “Hey, can I translate this Gutenberg FAQ into Japanese? Do I have your permission to?” And they put it on a different site. How cool would it have been if I just could have said, “Hey, by the way, collaboration and work flow, here, you now have access to make this post into Japanese,” and now at the top it'll have a little link to it, to the Japanese version, and there will be sort of a neural structure in a canonical place for the MA.TT version of the Gutenberg FAQ translated to Japanese by this volunteer, where anyone in the world could live. And by the way-
Brian Krogsgard: I was about to say, what if it was permissionless, like just if someone wants to make something on my site in other languages, do it. As small as Post Status is, I've had people that want to recreate the members side in Dutch or something.
Matt Mullenweg: Cool.
Brian Krogsgard: I'm like, I can't do that, but that sounds amazing if you want to do that. But I'd be totally down if the structure was there for it to say, translate this post, and it's almost like forking software, or almost the way language packs and plugins work. You don't have to be responsible for every language pack if you're the plug in author. Someone can just come and do it, and I think that'd be really cool for posts.
Matt Mullenweg: That would be so cool, and by the way, layer that in with other plugins. Layer that in with membership, so you only allow members to translate things, or layer that in with one I'd love to re-evaluate which is [Blickey 01:08:23]. So there is public revision history and sort of moderated submissions, so maybe I have, maybe my colleague [Nalco 01:08:34] can moderate Japanese translations, but maybe she doesn't have time to translate every single one of my posts herself. But when people submit one, she could take a few minutes to just read it through and make sure they're not having [crosstalk 01:08:47].
Brian Krogsgard: Yeah, translate my post and make it sound completely different.
Matt Mullenweg: Yeah, I think it's actually WordPress.com.
Brian Krogsgard: Assuming good intent is part of the thing there.
Matt Mullenweg: Do you remember the Happy Birthday incident?
Brian Krogsgard: I don't.
Matt Mullenweg: Someone, because when we opened up kind of like more totally community translations on WordPress.com and made it easy for anyone to do it, someone changed where it said comments, like leave a comment, into Happy Birthday! I forget what language it was, Italian, so all Italian sites under every single post would say Happy Birthday!
Brian Krogsgard: Just Happy Birthday!
Matt Mullenweg: Which I think is kind of like the most brilliant, if you're going to troll or vandalize us, that was a really lovely way to do it.
Brian Krogsgard: Yeah, we could probably learn some things from Wikipedia's own editorial flows and how to implement something like that. I do think that would be really neat. Do you think that still leaves room for other plugins that are doing various multilingual things? Does it leave room for them, or do you think it kind of eats them up and makes that business model go away? Because there's been some good-sized businesses in that realm.
Matt Mullenweg: It leaves room for them 100%. In fact, I think it increases them a ton. Same thing with page builders. If you look at where the business model is for a lot of these translation plugins are going, they're actually helping you do the translation. That makes total sense. If I'm a business, I might want my site in five languages. If I can click a button and just say, “Hey, I'll pay you $1,000,” and by the way the plugins can be in the middle of that or there can sort of be an API for that. I mean, this is kind of incredible, but the downside of them is very similar to page builders, which is each kind of has its own way to do the data and the storage.
Matt Mullenweg: Some put it all in post content. Some put it in custom post type. Some put it in meta fields. Some put it in separate tables, so you essentially have actually each of the possible ways you could kind of logically do this exists. So now, if I'm building a plugin that wants to work with multilingual versions of the site, how do I do it? There's no common foundation. Exact same issue we had with blocks. There were blocks on WordPress for years now, but every single visual composer and Beaver Builder and Elementor and Avada and Divi, they all did blocks in a different way.
Brian Krogsgard: Matt, there's like 100 business questions I want to ask you but we're over an hour already, so I will-
Matt Mullenweg: Can we take a few more minutes, like maybe wrap it in like, say 10?
Brian Krogsgard: Yeah, sure. If you're down, I'm down.
Matt Mullenweg: You're one of my favorite people to talk to.
Brian Krogsgard: Well, I appreciate it. I love talking to you as well, and I appreciate your openness.
Matt Mullenweg: Between you all, honestly you all and the Tavern folks, these are so much fun. I actually listen to a lot of these podcasts. It's really enjoyable for me, so I appreciate you doing it. My ask of you for 2019 is, I'd love to see a bit most Post Status on the scene, you know?
Brian Krogsgard: Yeah, that's a task for myself as well, so I did a relatively personal update recently.
Matt Mullenweg: I saw that.
Brian Krogsgard: One of the things that I've struggled with is I haven't been in the thick of things in about four years, so I'm actually not spoiled, but I'm doing some part-time work actually being in WordPress, and back into doing stuff.
Matt Mullenweg: Cool.
Brian Krogsgard: It's actually on the product side, not the service side. I've worked in an agency. This is going to give me a chance to do some part-time work on the product in SaaS side of things, so I'm excited about that. I know from history, when I'm working hard on something, whether it's building my own site or working in the industry, it gives me much more to say, so I'm excited about being inspired by working within our space more, and within the experience more. Because it's been four years of not doing that, and over those years I've tried to rely on other people to have a voice more as I felt like mine was fading. Kind of like what you were saying, you're around developers all the time and not, like you want to be around more users. I was looking at the industry a lot, and not being in the industry I felt made it more difficult to have good things to say.
Matt Mullenweg: Although I would say the outside perspective is really valuable as well.
Brian Krogsgard: Yeah, I think at least for a season I need, it'll be really beneficial for me.
Matt Mullenweg: I can't wait to see, and by the way, as a member of the Post Status community, include us along for the ride. I'm sure after that personal update you got a ton of people contacting you [crosstalk 01:13:14].
Brian Krogsgard: I did, which I was very thankful for.
Matt Mullenweg: It's one of those things, I think what was beautiful about that is you opened up about the struggle, not just about the good stuff. It's really scary to be vulnerable like that, publicly. It was scary of me to put the things we learned and did wrong in the State of the Word, right? I just wanted to talk about all the good stuff, but it's like it's important for us to talk about what I messed up on. People help when they see that.
Brian Krogsgard: Yeah, that was very-
Matt Mullenweg: And I think the most valuable thing you built is like overall a pretty positive community that cares about WordPress deeply, so that's awesome.
Brian Krogsgard: Yeah, and thank you, and I was very appreciative of the various responses that I got between feedback and ideas and job offers. But I'm excited to do some work within the space, and you know actually it did give me, it was beneficial because I got my head kind of out of the space a little bit, which I needed. I've been thinking about WordPress a whole lot since 2010, with basically no break, and I can't imagine how you feel at times. But for me to still have a focus on WordPress but at least spend part of my week thinking about other things, doing other things, it gave me some personal benefits I think, and allowed me to look at WordPress things with a bigger picture. So I'm excited for 2019 and I think it'll be a lot of fun. Now for you, just-
Matt Mullenweg: Well, I often tell people if you're ever feeling not inspired or burnt out, go to a museum. Watch a show. All of my best ideas for WordPress have always come from the intersection of different fields, different areas, and I think that's true of most great art, most great software, most great anything. So if you can get out there and keep, the WordPress stuff or whatever you're working on will be in the back of your mind anyway, so if you can kind of focus on something else for a little while, actually really it's fun.
Brian Krogsgard: It is.
Matt Mullenweg: I can't wait to see what you do. As always, let me know if I can ever be of help.
Brian Krogsgard: Well, thank you very much. So a couple of business-y things. One is, what's your biggest product right now, from an Automatic perspective? Are you guys seeing a lot of growth in the WooCommerce space, or is dot com still the main revenue driver? What are you seeing there?
Matt Mullenweg: Sure, let me just think through what I can say publicly.
Brian Krogsgard: When are you filing your S-1, Matt?
Matt Mullenweg: Yeah, we make the same joke in Automatic that they made at Apple, which is we're a ship that leaks from the top.
Brian Krogsgard: Hey, I have to commend you, as someone who doesn't like it but I can appreciate that this is the case. You guys are pretty good about being tight-lipped on the leaky boat side of things, so it's a tight ship. I don't get near as many of the leaks as I would like out of Automatic.
Matt Mullenweg: Well, I think a big part of that is just most of what we do is in the open. If you want to see the pricing tests we're running on the new blogger and e-commerce plan that we just launched on WordPress.com, it's in the Calypso GitHub so you can see which prices we're charging which people, and how the test is going, and what percentage it's going out to. Kind of, it's all out there. WordPress.com is still the largest, and it's still growing like a weed. I think in 2018 WooCommerce will actually grow faster in a year over year percentage, which is kind of what I started saying. We're starting to see what I predicted when we actually first brought WooCommerce in, which is that the e-commerce opportunity has the potential to be bigger than all of the rest of Automatic's businesses combined.
Brian Krogsgard: Yeah, what's Shopify, like eight billion plus market cap, maybe 15? I don't know. I think they doubled their size in 2018.
Matt Mullenweg: Yeah, market cap is just vagaries of the market, but revenue, they'll be about a billion dollars this year. So I think it's always good to look at kind of, when sort of comparing companies, look at the revenue not just the market cap.
Brian Krogsgard: Sure.
Matt Mullenweg: So they'll be at a billion dollars and still growing at a really, really good rate. Now, the other reality is that nothing that we do is completely disconnected. A lot of the WooCommerce growth and increasing user experience is sort of amplified, I think, because we've been doing it in Jetpack, which is also amplified by running Jetpack code on WordPress.com and having that sort of stream of new users that are coming there, and so we see the problems that they're having, and the struggles, or we build the infrastructure.
Matt Mullenweg: You asked about data structures earlier. I think one thing that we'll start to see a lot more of in WordPress, and actually a very common feature of Jetpack that people don't know about, that the Jetpack business plan, you can have full Elasticsearch version of your site. So there are certain use cases for which I don't think MySQL is the best medium with which to address or query your data, and Elasticsearch is pretty much the cat's meow. So the fact that we can in real time, create real time Elasticsearch indexes for every single Jetpack business site, which only costs $300 per year. Before that, it used to cost … Actually, I think VIP used to charge $5,000 or $10,000 a year for this, maybe even a couple grand a month, actually. Now we can do it for $300 per year.
Brian Krogsgard: Wow.
Matt Mullenweg: And that's just one of the features in Jetpack Business. That's in addition to the real time backup and the activity log and all the other things, the security restores. It just shows that we're really surfing the outer edge of Moore's Law, and I think that the stuff that we're going to be able to do over the next few years is going to be pretty amazing, and I'm excited about Automatic. We also just moved DNS Perf, we moved back into second place for fastest DNS service, so by the way, way ahead of Amazon, way ahead of Google, and second only to Cloudflare, which Automatic's network is tiny but mighty. Cloudflare invests I think over 100 million dollars per year in capex, building out their data centers in their kind of points of presence. We've reached a number two performance ranking with a fraction of that investment, so very proud of that team, very proud of all the folks at Automatic that are really building this out. By the way, then that benefits all the users of Jetpack Photon, the CDN, just so much cool stuff happening.
Brian Krogsgard: That's awesome. I heard recently that you all have also been able to lower some of your prices for VIP as you're deploying like VIP Go and some of that stuff with scalable Websites. Back when I had my hands in a few VIP sites I think the lowest pricing back then was maybe $3,000 to $4,000 a month, and now you can do it as low as $1,000 a month and you all provide a lot for that, so that's impressive.
Matt Mullenweg: That sounds about right, yeah. I'm not up on the latest there, but fundamentally my business philosophy is to get the best stuff in the hands of as many users for as low a price as possible, and we're just relentless in kind of cutting the costs, bringing things down, trying to leverage the best and newest technology to do so.
Brian Krogsgard: How are you attacking the very high end, so that a store that may grow up on WooCommerce doesn't feel like they have to leave WooCommerce once it hits a certain scale? One of the things that I feel like people have been, I've been hearing from consultants especially, is that they get to a certain size and then they're really struggling with WooCommerce, and then something like Shopify ends up being the answer. How is WooCommerce going to compete in that realm to be the platform for Macy's or who knows what? Some big company with lots of orders and lots of stuff, so that you can keep them throughout the life cycle?
Matt Mullenweg: Yeah, I think my take on that problem, because ultimately what we want to do is be more scalable in Shopify and also a better user experience, and right now I would say they're ahead in both. There hasn't been as good a feedback loop. The thing that made WordPress so scalable was there was a tight feedback loop between the host, the agencies building the biggest sites, WordPress.com itself being a big site, and the core development. So that meant that when we had bottlenecks, we were able to solve them often in core in ways that were ultra scalable in Webscale.
Matt Mullenweg: WordPress.com is on its own one of the largest sites on the Internet, so that's the same software that you can download and use, which is kind of wild. For WooCommerce, that loop isn't as strong so one of the things I've been encouraging the team to try to do myself is get a tighter loop between the host and the people building the big sites and the WooCommerce core team, so that they can figure out versus having five different plugins that try to address the scalability in different and often subpar ways, really get that built into core, with the sort of Web scale, knowing everything we've known in 15 years of building WordPress, what's going to make it so it can scale to thousands of orders per second, hundreds of millions of rows.
Brian Krogsgard: Two more questions, real quick.
Matt Mullenweg: We can do so, by the way. It's a very tractable problem. We will solve that. Solving that is something that will happen within the next six to 18 months. Solving the user experience to catch up with Shopify is also improving the user experience, or BigCommerce rather, I think will take longer. That's more of a five to 10 year goal.
Brian Krogsgard: I personally think that gives both the WordPress ecosystem, the WooCommerce ecosystem and Automatic directly a massive value proposition, if that can be accomplished. I think WordPress now people, you never feel like you have to leave WordPress because you got too big.
Matt Mullenweg: Yeah.
Brian Krogsgard: And as that can be accomplished, and that was a massive accomplishment for WordPress. For a long time that wasn't true, and now that, if that can be done on the e-commerce side of things, or commerce broadly not just necessarily WooCommerce, I think that will be fantastic for everybody involved.
Matt Mullenweg: By the way, I just want to point that out, just to pause. Think how little, how many other examples of software you can think of that the consumer version also works when you run it at an industrial global scale?
Brian Krogsgard: I have no idea. I'm not very-
Matt Mullenweg: It's not very common. I think it's one of the things the WordPress community should really be proud of, because that is highly unusual. And as we're one of the only ultra-successful consumer open source startups, projects, communities, whatever, I think pointing to that sort of scalability is a big thing as well.
Brian Krogsgard: So, one question is Automatic has investors. We just saw Slack, Lyft and Uber all file for becoming public companies. When I think, what's the option for Automatic, one of however many successful startups where you have investors, they know you're a successful startup. They know that there's opportunity for them to succeed in that. That's long-term revenue stream, remaining a private company. It's private equity. It's going public, or it's acquisition. Do you have any thoughts in terms of what makes you happiest as a company and as a business owner, and what do you think is best for the way Automatic interfaces with the broader WordPress community?
Matt Mullenweg: It's hard for me to imagine being a public company CEO, any harder than leading an open source project.
Brian Krogsgard: I've thought about that a great deal. I think you actually fit the mold for what Wall Street might like in terms of the leader of a public company. I will say, that's my bet. My best is one day Automatic would go public, and we figure out how that works in terms of all the private, public components and the open source side of things. But yeah, I'm curious-
Matt Mullenweg: I think it's actually a possibility. By the way, I could see other WordPress companies going public, probably WPEngine is the closest to the gates there.
Brian Krogsgard: Well good, GoDaddy and EIG are public.
Matt Mullenweg: They are, yeah, so there will be other open source WordPress either based or WordPress will be a part of their business public companies, likely before Automatic as well. I think that's a great path for us. Automatic has investors. They are minority investors, so the sort of existential future of the company is in the hand of myself and the other people who own the majority of the shares of Automatic. I think there's lots of options. The people you point to, Uber or Lyft, Slack, et cetera, to varying degrees have built some pretty impressive things. I admire some more than others, but also have reached a scale that's kind of far beyond where companies generally go public.
Brian Krogsgard: Yeah, they really have. It's been-
Matt Mullenweg: What that illustrates is that the private market is actually a lot more liquid and has a lot more capital available for when and how that needs the company or the ecosystem or whatever to grow.
Brian Krogsgard: So I could surmise from that that you don't feel like there's a rush?
Matt Mullenweg: I won't even speak to my own personal things, but say just in the industry there's less of a rush, so that is the plus and minus of so much capital moving from the public markets into the private markets. So people like Fidelity, T. Rowe Price et cetera that used to only do public companies now invest in private companies. That's great for the companies, because by the way as a private company, you don't have the kind of quarterly scrutiny that you do when you're public, and I think it's actually easier to out-innovate and out-flank public companies when you are private, which is one thing I personally like. But that said, it does keep more of that growth in the hands of only sophisticated investors.
Brian Krogsgard: I completely agree.
Matt Mullenweg: So there is a lot more of that upside that is no longer in the hands of what is commonly referred to as the retail investor. I would love for anyone who loves Automatic's products to be able to own a share of the company, even if it's just one share. That's not really possible under the current sort of regulatory frameworks, and for good reasons, by the way. As we saw with the kind of ICO craze, there's a reason all these protections for consumers and public market, public company standards exist.
Matt Mullenweg: But in the meantime, at Automatic we will continue to hold ourselves to public company standards, but really build in an extremely long-term way. If there's an idol there, I'd say it's Amazon, and try to make the best possible user experience, best possible products. If we continue to do that, I think from a business point of view we will have an array of options available to us and any number of choices.
Brian Krogsgard: And as a leader yourself of the WordPress community and Automatic, what's one big thing that you learned this year that you hope to improve upon in 2019?
Matt Mullenweg: Hmm, I know you're probably thinking about like a big thing, but I'm actually going to say a small thing.
Brian Krogsgard: It can be small.
Matt Mullenweg: It's really important, which is how much you can … What is the one thing everyone listening to this is doing right now?
Brian Krogsgard: Focusing, I hope?
Matt Mullenweg: Breathing.
Brian Krogsgard: Oh, okay.
Matt Mullenweg: And you probably weren't thinking about it, but now you are thinking about it. You're breathing right now.
Brian Krogsgard: I am.
Matt Mullenweg: And it is amazing how much you can influence your mindset, your state of mind, your mood, your everything, through breath. Just as an exercise, where everyone puts their hand on their belly and takes a big breath and pushes your hand out as you take it in, Brian, are you doing it?
Brian Krogsgard: Oh, sorry. I do. I feel good.
Matt Mullenweg: It has an immediate effect on you.
Brian Krogsgard: It does. You know, actually that's how we discipline our son often. A three-year-old is a very emotional being, so if they're upset one of the first things that you can do with them is just tell them, “Let's take a deep breath together.” So you take that deep breath and the response is also very immediate, the way they calm down and settle down and think more clearly because they took that breath, so that's interesting.
Matt Mullenweg: That's all of us, and also by the way, this is not something new. This goes back thousands of years.
Brian Krogsgard: You didn't just invent this?
Matt Mullenweg: No, no, you know there's so many, there's so much prior art here, but I'd encourage people to check it out. Also check out, when the videos go up, the Joanna Barsh talk, and Alexis Lloyd both had sort of almost like guided meditations and exercises. I think this is one of the next big areas for people who aren't familiar with to discover. They're actually really, really powerful, and it's been huge for me. In an extraordinarily busy and what might otherwise be seen as extremely stressful year, it was actually one of my best years of my life.
Brian Krogsgard: Awesome.
Matt Mullenweg: I do have to watch a clock, and I do have to run now.
Brian Krogsgard: That's fine. I'm happy to hear that. Thank you for spending so much time with me. It was great talking to you, and we'll catch everybody next time.
Matt Mullenweg: Thank you. I'll see you all, and thank you to all Post Status readers.
Hi Brian I personally thought you did a great job it was one of best interviews of Matt Mullenweg I’ve listened to in quite a long time.
Thanks Jonathan — I am biased, but I really agree about this one. 🙂 I got a lot from it and listened to it twice over the holidays. What makes a good interview, in your view? What do you look for? Are there other podcasts you enjoy (WP, tech or otherwise) where you see interviews being done especially well?
Comments are closed.