Talking to Preston So is easy.
I was nervous before our conversation simply because on paper there are things about the man that are frankly intimidating. Author of Content Strategy for Mobile Karen McGrane named Preston “the smartest guy in the field” in 2024. He was called “probably the smartest person working in this industry right now” by Deane Barker, author of Web Content Management.
But Preston So one-on-one is so personable, so engaging, that he instantly put me at ease. We talked about some aspects of his life and career, his experiences working in Drupal and other content management systems, what his keynote will be about, as well as his love of travel and learning languages.
Preston, you work at dotCMS. Can you talk a bit about what your job is and what it is you're doing?
Many of the folks who know me from the Drupal world are probably a little surprised to see that I've gone over to a Java-based CMS. But I used to work at Oracle also, which was a Java-based CMS. I don't really have a lot of opinions about Java versus PHP, but I know there's some strong opinions on both sides.
But dotCMS is really interesting as a company. We're an open source CMS. You can see all of our code, all of what we do. You can contribute if you want to. So in that case, it's very similar to Drupal.
I joined dotCMS about five or six months ago as our new VP of product. And in that role, I basically oversee all of our sort of product or product-related functions. And that means our product team, our design team, our data function, our developer relations function. And also, I work on our analyst relations functions as well. So I wear a lot of hats at dotCMS.
And it's very similar to what I was doing before. I mean, my background has always been in software, in the actual engineering … in coding.
I read in your bio about your interest in voice interface and voice content. Can you talk a bit about your interest in non-traditional interfaces like voice?
This ties into the writing I've done in the past around what I call the “channel explosion”…. These days, content needs to go to a lot of different places. One of the things that we often forget, especially those of us who have primarily worked with web content, is that content isn't just read, right? It's also spoken. It's also aural. It's visual. It's spatial. There are so many things about content that aren't really … tied to that rectangular box that we call the website or the screen or the web browser.
And a really good example of that is voice interfaces and voice bots or voice assistants. About seven or eight years ago, I was part of a really amazing team at Acquia, [that] worked on the first ever Alexa skill for the state of Georgia, building an Amazon Alexa skill that would allow people to ask questions: like, how do I register to vote or how do I enroll my child in pre-K?
Content needs to come from a single source of truth. You're seeing a lot of these new use cases emerge where people want to serve content to a mobile app, people want to serve content to a Roku device, people want to serve content to an AR overlay, for example, in your Vision Pro.
One of the reasons why I've been so interested in voice is because it really throws out a lot of the prescriptions and a lot of the ideas that we have about content, a lot of those biases that we have towards written, visual online content…. Web content is actually more abstracted away from natural human language and natural human biology than is speech-based interfaces or how we actually converse.
So I wrote a book about five years ago called Voice Content and Usability. In that book, I talk about voice content strategy, voice content design, how do you actually get content ready for a voice interface? And how do you actually implement an end-to- end voice interface that needs to consume content from a CMS?
When that book came out, there weren't a whole lot of Alexa content-driven implementations. It was basically just Capital One balance checking and Domino's Pizza ordering. And that was about it. No one had ever done a content-driven voice interface that was more informational rather than transactional.
Unfortunately, a lot of the things that have happened over the last few years with generative AI have really thrown those approaches out the window, because oftentimes with AI, you don't really feed it content. You're looking at content that is being reconstituted … by the AI as opposed to something that you're actually serving. But for governments, it's a much, much bigger concern for that content to stay up to date.
[You want] to help somebody learn how to get health insurance, or how to file a death certificate, [and that] cannot be mucked up by AI hallucinations or incorrectness. This is one of the reasons why voice content strategy and voice content still remains so relevant.
Your bio says that you're interested in “endangered and underserved languages”. Where else does your interest in learning languages come from?
My biggest passion outside of work, outside of professional pursuits, is travel and languages. A lot of it comes from my background. I spent a good amount of time in Brazil when I was younger, so I'm fluent in Portuguese because I did an exchange program there. I taught English there in college as well. I also spent time in Wales.
Some of the richest interactions and some of the richest experiences I have when I travel are when I'm able to converse in a language that is a very seldom learned language, a very atypical language. It's a language that people don't really often take the time to learn or have much of an interest in learning. But [these languages are the way] in which you can get to know the culture, get to know the food, get to know just the way that people interact in these other environments and in these other languages.
Languages are entire universes unto themselves. Especially those languages that have that rich, rich tradition of oral language traditions or rich literature that stretches back for centuries. I love to focus on languages that I can speak right now with people today.
Right now, I'm focusing on three languages – two of them are incredibly difficult. The third is a little bit easier, and it's all towards a vacation I've got planned with a friend coming up in November. We're headed to South Africa, and so I'm learning Afrikaans, which is obviously at the center of Middle Dutch, the sort of colonial language in South Africa, but I'm also learning Xhosa and Zulu, which are two of the Nguni languages spoken in South Africa.
Can you say a little more about the keynote presentation that you're going to be giving at Twin Cities Drupal Camp?
Over the past four to five years, I've been tracking sort of dissatisfaction on both sides of CMS.
I think one of the things that's really unique about the content management system is that it occupies a very unique ecological niche in the software world. Whereas a lot of other software products have a focus on individual personas, like Salesforce for salespeople. CRM tools tend to be for those kinds of folks.
The CMS has always been very unique in software because it brings together people with very different skills and very different priorities. Two of those personas that are probably the chief personas that the CMS deals with are, number one, the CMS developer. And then number two, the sort of content practitioner or content team or content architect or compliance reviewer or accessibility reviewer, everyone who has a stake in making sure that content is successful.
But we know based on just hearing from folks around the CMS industry that we're starting to see a bit of a schism right now, which is that there is, number one, a trend for developers to go towards headless CMSs, like Contentful, Sanity, some of those, and really go in that direction. But the problem with that is that it kind of leaves content teams with their hands tied behind their back. They can't really do drag and drop layout management anymore. They can't do preview of all of their different sites anymore. There's a lot of issues that come up with headless CMS.
But by the same token, developers today really don't want to work with the sort of monolithic or traditional CMS anymore. I love Twig. I love PHP template. There's a lot of folks who don't. There's a lot of folks, especially who are coming into front-end development nowadays, that really, really don't like to work with those paradigms.
One of the things that I think is really important is that as we contend with this huge influx of new JavaScript frameworks like Astro, SELT, so on and so forth, and also new delivery channels like we were talking about earlier, Blaine, around AR, VR, voice, AI, so on and so forth, it becomes a really big concern.
How do we actually collaborate effectively in a CMS that works for everybody and not just one half of the back office? One of the struggles that we see very often is that oftentimes headless CMSs will say, well, hey, content is just the data. Let us handle the presentation. Let us handle the front-end. Let us handle how things look.
But what that does is it severs all those linkages with how content authors want to preview, with how content editors want to be able to look at and review or schedule content or review things for compliance or review things for accessibility, so on and so forth. But developers also don't want to be held back.
The topic of my talk is really what I call the universal CMS, which is a new pair and I am really quickly getting a lot of traction. It really is about restoring the balance that characterized the early static web CMS era. Basically saying, hey, we could do all these really cool things with the website, but we had a handshake where we agreed that, hey, developers, if you hand over control over layout and control over all of these visual components, I will give you obviously control over how to code the whole thing.
But this unique grand compromise that we forged is something that is starting to come back. We are starting to see headless CMSs build in visual editing features which violate the peer headless architectural prescription. We are also seeing a lot of the old traditional CMSs or monolithic CMSs begin to build a lot more APIs and SDKs for JavaScript developers or mobile app developers to build on top of. And so I think what we are going to start to see here is a convergence between both the headless CMSs and the traditional CMSs towards a new equilibrium which I call universal CMS.
And here in just a few years, I think we are going to get rid of this full distinction between headless and monolithic and all of those tired terms that have a lot of baggage with them.
[Short bio]
Preston So (he/they) is a product executive with over 25 years in software, 17 years in content technologies, and 9 years leading product, design, engineering, and developer relations functions at organizations such as Oracle, Acquia, dotCMS, Time Inc., and Gatsby. He is Vice President, Product at dotCMS and the author of Immersive Content and Usability (A Book Apart, 2023), Gatsby: The Definitive Guide (O'Reilly, 2021), Voice Content and Usability (A Book Apart, 2021), and Decoupled Drupal in Practice (Apress, 2018).
Named “the smartest guy in the field” by Content Strategy for Mobile author Karen McGrane in 2024 and “probably the smartest person working in this industry right now” by Web Content Management author Deane Barker in 2020, Preston is a globally recognized authority on the intersections of content, design, and code. He is an editor at A List Apart and former top-read columnist at CMSWire. Preston is a frequent presenter with 17 years of speaking engagements spanning over 50 conferences, including SXSW Interactive (2017, 2017 encore, 2018) and An Event Apart (2020–22) and keynotes in three languages. He is based in New York City, where he can often be found immersing himself in languages that are endangered or underserved.
