Rethinking Open Source W/ John O'Nolan of Ghost  ## Metadata - Author: [[The WP Minute+]] - Full Title: Rethinking Open Source W/ John O'Nolan of Ghost - Category: #podcasts - URL: https://share.snipd.com/episode/c19cbfd9-1d51-4fc0-834b-fd8ed21b1c89 ## Document Notes ## Highlights - **WordPress vs Ghost Approach** - WordPress is a generalized CMS allowing flexibility through plugins but comes with complexity and security challenges. - Ghost focuses on professional publishing with core features and limited external integrations, aiming for simplicity and focus. Transcript: John O'Nolan That the one thing that pulled you back was going to be custom fields, but here we are, forms. Yeah, the blessing and the curse of WordPress in my mind has always been the power and the extensibility of the platform and its ecosystem and all the different things you can do with It. On the one hand, amazing. On the other hand, comes with some downsides, comes with complexity of integration, comes with security issues, update servers that you may or may not have access to all those things And i think different platforms have different answers depending on should have different answers depending on what they're trying to do wordpress is what i would call a generalized Cms you can kind of decide what you want to build with it and it can become that it can morph into an e-commerce platform or a real estate directory listing or a blog or a business website. And the extensibility allows you to fulfill those different use cases. So it's absolutely critical that you kind of have this really deep set of capabilities that plugins can hook into. What we do with Ghost is in many ways kind of the opposite. We've focused in on a single set of use cases, which is publishing and publishing professionally, effectively. And so we try and take all of the things for those sets of workflows and just build them natively into core. So the way we kind of think about extensibility is it's just quite different. If there's something that everyone's asking for, it kind of should be in core. And if there's a longer tail set of things people are trying to do, then usually we try and do that via an integration with a product that lives outside of core, usually by API. Or in some cases, it's just really indicative that you're trying to do something that the product wasn't designed for. And there might be better capabilities or better other options out there that have the capabilities you need. ([Time 0:02:40](https://share.snipd.com/snip/cd10ed0e-e7b4-4460-9d48-1eaf5ad52f14)) - **Onboarding New Developers** - Try Ghost by signing up for a free trial to explore its managed hosting and theme system. - WordPress theme developers will find Ghost's Handlebars.js templating familiar and easy to adopt. Transcript: John O'Nolan It is it's not nearly as extensive you know the scale of the of the WordPress ecosystem means you can find almost anything on any topic tutorial guide video podcast or whatever you can Think of we don't have quite the breadth of content but we do have a reasonable depth. So if you have never heard of Ghost before, you have, but you haven't tried it, you know, our website's a good place to start. We have documentation, we have tutorials, we have examples. Probably the best way to try it out is simply to kick the tires. You can sign up on ghost.org for a free trial, no obligation. We will not be automatically charged at the end of it and have a fully managed instance spun up working and get a sense of how things work. Have a look at the theme layer, have a look at the integrations layer. Ghost theme levels modeled after WordPress. So if you are a WordPress theme developer familiar with, you know, all of the hooks and functions and data outputs that you would normally use there, it's going to feel fairly familiar. There's a couple of concepts which are different, but the syntax is not a million miles away from what you're most likely already used to. ([Time 0:05:07](https://share.snipd.com/snip/50b928f7-ecd7-4d41-aea1-0d2725d3a544)) - **Ghost's Small Scale Vision** - Ghost aims for sustainable operation with a nonprofit model and a small team around 50 people. - John values growing a diverse ecosystem over dominating market share or maximizing company size. Transcript: John O'Nolan Much of the slice are you trying to get? I'm kind of, I think DHH's post this week was fairly good on this topic where he said, you know, that the scale of value that Rails has created vastly outstrips that, which he has personally Captured. And that's a good thing. That's what open source should be about. I agree with that. I would almost take it even further and say that we are attempting to capture the smallest possible slice of the market to sustain our operations and nothing else. So Ghost is a not-for organization, single nonprofit foundation. There's no other companies with venture funding or other motivations. It's just one nonprofit foundation. And our goal, with lots of goals, but one of them is to never grow to more than 50 people in headcount, around 50 people. It might not be exactly that number, but around that mark. And the point in my eyes, or how I think about the future, is not to try and grow the biggest company possible, or to try and capture as much value of the web as possible, and certainly not To try and capture a percentage of the web's total market share but to grow an ecosystem which a core team in the structure of a non-profit foundation is able to support and in my mind that Looks like lots of different companies doing well not one giant company you know making the majority of the money and controlling how everything works, but really more of a chaotic, In some ways chaotic, decentralized ecosystem. I think of it as, you know, like a city. You might have lots of shops, lots of houses, lots of different businesses. The goal of leading a good city is to create infrastructure that can allow a diverse set of people to thrive within it, not to control everything and to own everything and to buy up everything And to turn it into some dystopian nightmare. So we're trying to really do the biggest possible thing with the smallest possible team and make enough revenue to sustain, to make what we do possible, but not more than that in business Terms. ([Time 0:11:00](https://share.snipd.com/snip/97881e8d-1ad1-4a47-81ec-24444f398833)) - **Why Keep Teams Small?** - Large companies can become unruly or lose fun and creativity as they grow beyond about 50 people. - John prefers a tight, agile team that fosters personal connection and creativity over scaling for scale's sake. Transcript: John O'Nolan And no. It's definitely not easy to run a company of any size, and it definitely doesn't get easier as you get bigger. That being said, there's quite a lot of people who are doing it very well for every person who's perhaps not doing such a good job of it. And it's not something we have no reference material for, right? 1,700 people dwarfs in comparison to the 19,000 at NVIDIA or the hundreds of thousands at Microsoft. There are playbooks, there are good advice out there. There's a lot of hard problems you have to solve along the way, but there are also a lot of easy things that you can not step in along the way. And I think the things we've seen within the WordPress community in the last three weeks have not been representative of the hard problems of running a 1700 person company. They have been some of the 101 basics of things you don't do, whether you are a freelancer, a one person company or a two person company. And that's the thing that I've found a bit disappointing. But the stop gap that you were kind of alluring to in our context is less about, I think companies get on, by definition, get unruly at scale. And that's something I'm worried about. It's more that it doesn't seem like fun to me. My idea of a good time is not having eight layers of management and not knowing anyone's name. My idea of fun is honestly harkens back to my WordPress contribution days where I would show up on IRC, I would chat shit with a bunch of the other contributors and we would make cool stuff Together. And at the end of the day or the month or whatever it was, we would release stuff that I sitting on a laptop in a pair of $2 shorts on a beach in the Philippines had written with no status, no Money, no clout. And it would get released to hundreds of millions of people. That's my idea of a good time, having a tight-knit team of people doing really cool stuff at a scale that doesn't make sense just because of the amazingness of open source and how far and Wide it goes. So I simply don't want to run a company with more than 50 people because I think that's the point at which you start to lose that. We have a lot of team members who've come from startups that grew past that point, and they all kind of have the same story, which is like it was great at the beginning it's really fun we Were doing all this cool stuff and then around 60 70 100 people it started to break down new leads of management were required the priorities changed it didn't feel the same anymore it Just doesn't sound like fun to me so there's there's a selfish goal in there of you know, with Ghost being a nonprofit, I don't stand to sell it one day or make a shitload of money off it. So it has to be something I enjoy. It has to be something that I'm fully invested in that feels like something I could do for a long time. Because if it's just a slog for the sake of scale, there's nothing to it. So keeping it small, it focused is a creative constraint we've chosen and what's fun about choosing constraints is invariably you have to try and get around them right you've got this Constraint what are we going to do with it we've only got 50 people well that changes what's possible with the product it changes with what's possible of what portion of the market you Serve what type of market that is it changes all kinds of things that then there's an amazing amount of creativity that comes from those and trying to figure them out. So I'm a huge fan of creative constraints of any kind. ([Time 0:14:56](https://share.snipd.com/snip/9107d6a7-d2dc-46bf-ae41-57b9a5544636)) - **Ghost's Nonprofit Governance Model** - Ghost is run by a nonprofit foundation that sells managed hosting to fund development. - John plans to grow the board to include community-elected trustees and reduce his own control to zero. Transcript: John O'Nolan A lot of the way we've structured things was inadvertently inspired by automatic and WP Engine, believe it or not, 12 years ago. So being in the WordPress kind of core team and seeing how the product was made, I had direct exposure to kind of all of these issues, perhaps earlier than other people. And the tension and the kind of conflicting motivations and priorities of all these different organizations within WordPress, I always found difficult. You cannot have a venture-backed arm on the one hand, a nonprofit foundation on the other, and a.org website on the third, all controlled by the same person, even though they have conflicting Priorities and expect everything to be smooth sailing. It was always going to be a matter of when, not if this happened. The amusing thing is that all these conflicts were what inspired a lot of ideas for Ghost of how not to do that. I was very bought into the idea of democratizing publishing when I was working on WordPress, and that still means a lot to me. And the conflicts of interest around kind of all the positioning that goes on behind the scenes was not what I was excited about at all. So a lot of the ideas for Ghost were based on what would an open source product that's trying to just democratize publishing look like if it lived up to its ideals? Now, I don't know if we've succeeded in that, but the angle we came up with was an attempt at trying to answer that question. So Ghost is a single nonprofit foundation, and that nonprofit foundation sells a managed hosting service, which we modeled after WP Engine. It seemed like WP Engine had a great business model. You had the full power of WordPress, but you didn't have to worry about servers. Whereas by comparison, as you'll remember, WordPress.com was not used by anyone who knew anything about WordPress. These days, you can do themes and plugins, but for a very long time, you had no option to run themes or plugins. So in my mind, I wanted to have the simplicity of a non-profit organization to steward the open source project, the business model of WP Engine to fund that open source project, and the Transparency and integrity of what an open source project should be, which is to have no outside influences or ulterior motives that come whether you want them or not the second you Take venture capital. And so far, so good. You know, we're 11, 12 years in now. We are at $7.5 million a year in revenue. And we were profitable from the very first year. There were enough people that kind of believed in that model, although there were many more who didn't quite understand what was the point. Why do you need this weird nonprofit model? And if anything, I think the last few weeks have drawn attention to why it can be a good idea. Jono Nolan Is there a board of directors on the foundation? How does that sort of oversight work in Ghost? John O'Nolan Yeah, you're right. I skipped the bus factor. That was part of your question. In a non-profit foundation you have a board of trustees and at the moment those two trustees are myself my co-founder hannah and for the first 10 years or up until now of ghost i think that's Been roughly the right model getting something going getting off the, you really don't need decision by consensus or much oversight. You just need to prove that there's a market for the thing you're doing. And that requires simply moving quickly. One of the things I see often is people coming out with a big manifesto and a big governance structure before they've built a product, which is the wrong way around of doing things because People aren't buying your governance structure. They're buying your products. But why people stick around often when something gets popular or large is if it does have a good governance structure, which allows it to persist without inadvertently decimating Itself at some point. And so as Ghost gets bigger, my intention, and I'm going to write a bit more about this soon, is to have the board of trustees grow, to have independently appointed people that we have Selected, but also to have board seats that are elected by the community. And I think it's going to be interesting figuring out what are the different constituencies within that community, how long should a board seat work. But equally, there's some great reference material there. You know, Rails has a foundation that seems to work pretty well. Drupal has a foundation with elected board seats that seems to work pretty well. There are all kinds of organizations outside of software development and direct open source that have collaborative board seats that oversee the larger direction of a non-profit Foundation. ([Time 0:18:59](https://share.snipd.com/snip/b607f0f7-b83f-4438-ac50-ec28679ae4ed)) - **Moving Beyond Founder Dictatorship** - Founders bring vision to start projects but long-term sustainability requires leadership beyond one individual. - John opposes the benevolent dictator for life model and envisions democratically elected leadership for Ghost. Transcript: John O'Nolan One of the cultural moments that we're in at the moment is celebrating founders and visionaries, people doing things in business to an almost insane degree, to the point where they, You know, cult leader, founder, it's not that different. And we've probably gone a little bit too far on that. I do think it is very helpful for getting new ideas started, new products made to have someone who is deranged enough to believe they can change the world or whatever the correct version Of that quote is. And I do think great products by and large are made by individuals, certainly not by committees. I also think that the staying power of a company or an organization has to at some point transcend one individual. And, you know, to pick up on a couple of the examples, you mentioned that apple continue to thrive after steve jobs now we can argue about whether or not they still make good products and Whether there's the right person calling product shots there and i would tend to say the answer to that is no but the organization has outlasted because it was able to transcend it was Able to evolve after the founder was no longer around microsoft's another great example because we have three axes to it. You have Bill Gates, who got it going, had the big vision. You have Steve Barmer, who by all accounts made Microsoft flounder, but had the organization become very profitable, which is kind of what Tim Cook's doing now. And now we have Satya Nadella, who's come in. And okay, we now have big visionary ideas coming out of Microsoft. We have for a few years now. Microsoft is doing some of the most impressive hardware, software, investment, and acquisitions of any large tech company. And it is a visionary individual with big ideas making decisions, but it's not the founder. So to the extent that I think you need one person making big shots in order to do big things, I think that's true. I think I agree with that. Does it have to be one person for life or the original founder of the company? I don't think so. In the context of Ghost, that's currently me. Yes, I am that person. I make those decisions. I think I do an okay job, although some would look at the size of Ghost as compared to WordPress and say I've not done a good job and that's okay. But I don't intend for it to be me forever. I also don't intend for it to someday be a committee. I think you need one leader with clear ideas. I think the key difference that I really hope WordPress ends up at and certainly where I want us to end up at is if you're going to democratize publishing, then that person should be democratically Elected somehow. There should be some ability for the broader community to decide who that leader is and then trust them to make good decisions and vote whether or not they are. Without that, it is just a dictatorship. Not a dictatorship in the sexy good way, if there is one of those. Just a textbook dictatorship. ([Time 0:26:29](https://share.snipd.com/snip/6fc7b7e3-5010-4504-b7ce-e06d0f5882b7)) - **Sustainable Funding Beyond Donations** - Support open source projects by using their paid hosting services instead of relying on donations. - Offering visible incentives and partner recognition can help projects financially sustain contributors and development. Transcript: John O'Nolan Yeah, it's kind of easy. By having these things live in one organization, we have the simplest thing in the world. We can say, if you want to contribute to Ghost, you feel aligned with the project, you would like it to succeed, use our hosting. Our hosting pays for the open source developers who work on the core products and improve it for you. So by simply using Ghost, you are contributing to Ghost. If you never write a single pull request, but you use our hosting, you're directly funding the nonprofit foundation that makes this possible. And that's not a coincidence. The idea of this business model was exactly that, was to make it so that the users of the products are the people who are investing in the future of the product. We have tried having like a donation button or mentioning it on the site. I think we found much the same thing as the WordPress Foundation has found, which is you don't get very many donations. People, for the most part, don't donate to things like tech companies and tech products. It just doesn't, that's not a thing that happens. And I had in my head that when Go started, I was like, oh, maybe, you know, Google will give us a bunch of money like they give Mozilla. These grand ideas of philanthropy, that hasn't panned out either. So making the hosting model work, making sure that we have enough margins to run the business has been super important, but it also hasn't been very complicated. I think when I think about what could be other ways that we would make the organization more financially successful or to kind of have other revenue streams, there's a lot of options, But they all kind of hinge on incentives. Something that I think has been missing in the WordPress space is why should anyone contribute 5% back? What are they getting in return? Because if it's just a kind of give me 5% or I'll go nuclear, it's just not a very good incentive. But there's lots of other projects that have done this better. And we're not one of them because we haven't tried this yet. But if you look at Laravel, they have a sponsors section. And for, I think it's thousands of dollars per month, you can sponsor the Laravel community, the open source project, and you will get a prominent partners page on the website. Listed as an official partner, recognizing what you do, marketing your services if required. So it seems like the price point would be pretty high if WordPress did something similar. It seems like if there was something you got in return for financially contributing to the WordPress project, then you're selling a product. But if you're just saying, give us money and we're not giving you anything back, yeah, that's a harder sell. It's a harder sell for really anyone, but particularly for a software product that you can't see. ([Time 0:30:27](https://share.snipd.com/snip/d0727039-aae4-4282-b1e1-7e9460dc70b6)) - **Recognizing Contributions and Paying Developers** - Plugin and theme creators contribute significantly to WordPress's growth and should be recognized. - Open source projects need to pay contributors to remain competitive and sustain ongoing development. Transcript: John O'Nolan I think the litmus test that I would use, and I'm not suggesting everyone should use this, is would WordPress exist today or be as large as it is today if you took away all of the non-official Themes and plugins. If those just went away and the only themes and plugins available were automatics, would WordPress be around? Would it be as large? If the answer to that is no, and I think it's up to everyone to hypothesize for themselves, if the answer to that is no, then it follows that the people who have created those themes and Plugins have been in part responsible for the growth and the success of WordPress. And if that's true, then they have contributed. In my mind, they've contributed a lot, regardless of whether or not they make money from the things they've worked on. You should be allowed to make a business. No one should forbid you from doing that. So those contributions, I think, are completely valid. And it's not an impossible thing to measure. Yes, it's a hard thing to measure. But again, Drupal does this pretty well. They have a really clear framework for how contributions are measured and what things are included. It's self-reported, but there's oversight. There's a clear decision-making framework around what things are and are not recognized. And they've created incentives around what it means to be a contributor and how you're recognized as being a maker or a taker, as Dries put it. So there's ways to do it. But I think every participant in an ecosystem is contributing to that ecosystem. Is there another definition of contribution, which is purely about core and evolving the core platform? Sure. But what i'm not sure i'm not sure if that should be entirely volunteer based all the time i remember when i you know i was never paid to contribute to wordpress and i remember it was almost A dirty word to consider like being paid to contribute to open source almost like it should be purely altruistic and we should all just collaborate because then the thing gets better And we all benefit from it. And that's a lovely idea. But in the meantime, the rest of the web is moving forward without us. And as Morten has pointed out on more than one occasion, there's a lot of people contributing meaningfully to WordPress who are struggling to put food on their table. And the idea of bug bounties or paying open source contributors is kind of, no, we can't talk about that. But this should be everyone volunteering and working together. And I think maybe, this is just maybe a controversial opinion, I think maybe we're at the end of the road for that ideology of open source. Open source projects need to generate revenue and they need to pay their contributors. And in doing so, create an incentive for people to contribute. ([Time 0:34:26](https://share.snipd.com/snip/feaae118-d71f-4f1c-bee9-3a093046cc82))