- This is a continuation of a piece I wrote in August 2025.#
Yesterday I wrote a this.how doc about getting WordLand to run on sites that don't run on WordPress.com or have Jetpack installed. What this means is that their sites are accessible through the fantastic wpcom API, and with that API I have been able to build a very nice writers' workspace for WordPress sites. I call the ones that are left out the WordPress.org sites, though I know that's not an entirely accurate way of describing it, but neither is "self-hosted sites" because you could have a hosting service that doesn't install Jetpack. WordPress could be used for things that are quite different from the way it's used today. Important point. #- I sent links to that post to two people, and encouraged them to pass it along, as a way of starting a discussion -- how are we going to get WordLand running on those "other" sites. And we need names for lots of things, but first we have to get the software working the way we want it to, then the naming will be more obvious.#
- Jeremy Herve has already written a response (as I asked him to). As always, very helpful. But I want to push back on the way he frames the project -- as if it's up to me to do this work, and that I dispute. It think it's up to us. WordLand isn't just a product, it's a challenge.#
- To make WordPress a fantastic platform for developers who want to build on its ability to write and read and link together writing and reading on the freaking web. That's why it's so important to understand what the web is and what it's capabilities are. We've been missing some of the big stuff since the big silos took over. I want, before I hang up my hat, to fix that. #
- I'm going to outline here how we start, so it's documented so more people will see it too. #
- I have written a package, MIT-licensed, with an awkward name for historic reasons called wpIdentity. It started off as just a simple way to use WordPress identity with FeedLand because it made sense to do it a couple of years ago because Twitter was destroying the utility of their API, and we needed a new way to do identity. So I stopped everything and implemented both email identity and WordPress identity, or more specifically, wordpress.com identity. This leaves out all the wordpress sites that you can't log onto this way. Not a good situation, I don't like it because I want WordPress to be one large happy family of users and developers. So that's what I'm seeking to instigate here. #
- Then wpIdentity grew and grew. At first it supported all the endpoints in the wpcom API, and made it possible to build WordLand, which itself only communicates to WordPress via wpIdentity. And later when I needed user-specific storage that is not part of the WordPress API, I built a very nice little storage system, it's efficient, highly scalable, and (I think) well debugged. If you read my initial Think Different piece you'll see it gets a mention in this part. Storage, we will find out, is the big missing piece in the web, has always been the missing piece, the thing that keeps us from having a real developer community of web apps. How do I know? Because I'm a freaking developer. I don't work at a company that can insulate me from this need. That's why I see this need and so many others don't.#
- Now here's something else that's cool. I didn't stop with wpIdentity, I made a JavaScript function that simplifies the wpcom interface for people writing browser-based JavaScript apps. It radically simplifies it. I didn't invent this way of doing it, someone at Facebook did, as far as I'm concerned. We may not like Facebook the company or the boss man, but they have employed some brilliant programmers over the years. I think this idea came from the developers of FriendFeed, but I could be wrong. What they did was take the interface code every app developer has to do, and made that the API. I remember when I first saw it in 2013, it blew me away. It took me a full month to build support for Twitter in my app, but it only two days to do it in Facebook. I did one of those for WordPress. It's part of wpIdentity. Here it is. Please if you're a contributor to wordpress.org -- study that thing, because we're going to be working on that code -- together. This is where the "we" part of it starts. But there's more.#
- You don't have to read the code of api2.js to understand how it works, there's an even easier way to do it -- wpEditorDemo. It's a simple browser-based editor that does what WordLand does, without all the flourish. Bare bones. Also MIT-licensed. Nothing but what's required. You can try it out if you have one or more qualifying sites (either on wordpress.com or having Jetpack installed). The project we need to do is this -- make wpEditDemo work on any WordPress site, no matter where it's hosted. And that's going to be done by extending api2.js. That's where the differences will be encapsulated, and above that level, in wpEditDemo and WordLand and all its competitors, nothing changes. This project may be complicated or simple, or somewhere inbetween, but imho this has to be done. And this is the first place I want to join the open source community for WordPress. After talking with Jon, I became optimistic about this happening sooner than later. #
- I zoomed through all this and more with Jonathan on Tuesday. This is what I want to talk about at WordCamp Europe later this year (I have submitted an application). I wanted to show Jon and a few other people I met in Ottawa in October how the pieces fit together. But zooming is the right word, Jon was ahead of me at every step, he just understood what I was showing him. I was blown away -- because this had never happened before with the architecture I had built. Not because it's so complex, it isn't -- it's all about simplifying things, with lots of factoring, finding a repeating pattern and then locking it down in code, so no one has to ever worry about the complexity again. I think it was hard for people to see because they aren't expecting it. #
- That's where I'm going to leave it for now. I suggest if you have a response or comment, please put it on your blog and send me a link. We're going to create new ways to link blog posts together, but for now we'll do it by hand. #
- PS: Hat-tip to Matt Mullenweg for building WordPress, and for having the idea that Tumblr could be rebuilt on top of WordPress. When he announced they were going to do that I knew I was on the right track. We were thinking about WordPress the same way. #
- PPS: If people have a hard time understanding how I could do this -- remember: 1. I've had a lot of time. 2. I understand WordPress at a deep level. 3. I'm good at building and selling APIs, I've been fairly successful at it. And even better, the WordPress community has been a very good caretaker of this work. They support a remarkable amount of it, and it all still works, and they often wonder why the rest of the world doesn't love it like they do. I feel the same damn way. ;-)#
