Digital Infrastructure and Why Open Source Is Not Enough Anymore

Digital Infrastructure and Why Open Source Is Not Enough Anymore

0:00

/3664.536

1×

Wordpress co-founder Matt Mullenweg has been under fire recently. And while the drama is messy, petty and also kind of entertaining, it also highlights how we, as a society, need to find a way to take better care of our digital infrastructure. Tools that our society depends on are more and more of a digital nature. And yet they are treated differently than roads or the water mains.

I talked to tante to think of new ideas and how we could build a future where software that benefits democracy can be governed in a way that guarantees the level of maintenance, care and user involvement that it deserves.

Book recommendation: Deb Chachra: »How Infrastructure Works - Inside the Systems That Shape Our World«

Transcript

(The transcript has been generated using Macwhisper and might contain errors.)

Hagen: Hi and welcome to my new podcast. Like I said in the announcement this will not be a regular occurrence and only happen if there's a topic I feel the desire to talk about to gather new ideas for things that I feel like are under explained or concepts where I want to learn more about them. So that's also why it's a podcast and not just a blog post even though this podcast is meant as an extension of my blog because sometimes my own ideas just aren't enough so I decided to just talk to people who I think can help me to figure out new ways to think about problems I'm seeing and for the first episode I chose to talk about this whole WordPress situation but not really.

If you don't know what I'm talking about we will explain it in the episode it's just a symptom of a bigger problem we have where digital infrastructure isn't maintained the same way public infrastructure is usually maintained in the meat space but i think that could be a problem and i don't want to sound dramatic but i actually think it could be a problem for democracy itself so i feel the urgent need to figure out new ways how we could govern the internet or govern big, important online infrastructure in a way that makes it impossible or at least very unlikely that something like this whole WordPress thing happens in the future. And for that, I decided to talk to tante, who you may know as someone who has a lot of thoughts about exactly these things. But if you don't know him, let him introduce himself. Both of our blogs are on WordPress and how long has your WordPress instance been running?

tante: I used to say a jackasss with a Twitter account, but I no longer use that fascist service. so I'm just a jackass now, I guess. No, I am an independent writer and do a lot of writing and thinking about open source as well as other techno-social topics, so to speak. I checked that a few days, no, weeks ago by now. I think it started 2012.

Hagen: So, it's been running for a decade now, or over a decade now, and I have the feeling, like over the last decade, WordPress has become the default platform if you want to host a website or run a blog on the internet, if you want to self-host at least.

tante: yeah and if you're not a nerd like i mean some some very tech oriented people they have all kinds of different types of technologies that they want to use and like and and they have their advantages but especially if you're not super tech savvy and you just want a thing that does the thing you want a box that you can type into and it's supposed to generate a website for you without much digging around wordpress is definitely the go-to because it runs on every cheap space that you can rent like pay like two bucks a month and you have a web server that can run wordpress decently. Yeah. Yeah, it is all the like the naked WordPress installation is passable, but it's okay. But there are so many themes that you can get for free. There are so many plugins for everything you might want to do. And also, most of them are free or have a free version that will apply to most people. This infrastructure that you just hook into by just choosing WordPress. All this code and functionality and visual that are available to you at the click of a button.

I mean, you also don't need to understand anything. You just click add plugin, you search for a thing, and then it gives you plugins. You click one, it does the thing, you're done. The reason for WordPress is not that it's such a great tool and the editor is so nice or the code is so great or efficient or whatever. No. A, it runs on every cheap VPS, and it has this huge ecosystem that everyone can just use.

Hagen: I think that's a huge thing actually because I'm one of those nerds who liked dicking around. So I switched CMSs like other people change their underwear and I never blocked. And then I decided, no, I should actually start writing again, properly writing. And then I installed WordPress because it just works. It really just does. It runs. You don't have to do anything about it. It updates itself.

And I think that's something I've been currently running into because things happened and I did the radical move to switch away from WordPress. And I think you're still thinking about it. WordPress is run by something that's pretty common in the open source space, like this whole benevolent dictator for life role. In this case, it's Matt Mullenweg. And he turned out to be not so benevolent. I think that's a nice way of putting it.

tante: I'd say emotionally unstable.

But the interesting thing is not that WordPress is run by one guy who's in charge. I think the interesting thing is that for the longest time, people, either due to laziness or because it was convenient, ignored that because it wasn't framed as such. Like there are, the Linux kernel is run by Linus Torvalds. What Linus says is the kernel. Everyone knows that. It's not hidden. And some other projects work this way as well. They have like the guy in charge. It's always a guy. Like I've never seen a project like that run by a woman in this shape or form.

Hagen: I mean, the current example would be Mastodon, which is also one guy.
tante: Like they exist.

tante: It's run by a dude, yeah. And there's many, many projects, but WordPress presented itself differently. There's the WordPress Foundation, and there's also many open source projects that run like that, where there's a legal entity, often a foundation, or in the U.S. these public benefit structures that you can set up, or in Germany, an eingetragener Verein, like these instances that allow you to not pay taxes, and that allow you to build like democratic structures or whatever structures, government structures to manage this project that don't depend on one person's will. And WordPress kind of presented itself as such. Like there was the WordPress foundation. It still is. Like there is the WordPress foundation that owns WordPress.org, we thought, where you download all the plugins, all the themes, the source code.

And it says the foundation kind of manages the development of the open source project. So it's in their hands. And Matt Mullenwick, the founder of WordPress or one of the founders of WordPress, also has a company that's called Automatic that runs WordPress.com where you can pay money to host your WordPress for you.

Hagen: Which seems like a reasonable setup in the first place. Like you need money to develop something like WordPress and not everyone wants to run their own servers. So just buy it from them and they can run it for you and you can just write. Sounds like a reasonable idea in the first place.

tante: Totally. I mean, the interesting thing is that there's this reasonable setup. I can host my stuff myself, or I can just pay Matt Mullenweg's company and have this good feeling of not only do I get my shit hosted and professionally run, but I'm also kind of investing in the open source project because automatic also funds a lot of people who work on the open source project it sounded like a win-win and i think everybody felt like that like that was the setup and nobody really looked closely at what was going on i mean it wasn't hidden but no one because that was the gut feeling and everything worked until like a few few weeks ago by now where people realized that this setup is is more like it looks good but that's not the facts ...

Hagen: Exactly, it's not the facts because it turned out like Matt Mullenweg now claims he personally owns the WordPress.org domain, so he claims that isn't part of the foundation. There are some disputing facts in what has been registered at which point, and it turns out that the whole board of the foundation is under his control so they aren't independent from each other they don't vote like wordpress the foundation is exactly the same as automatic it's it is matt malnweck at this point and he decided to start some beef with some other people in the community and that went so far that he took over some plugins like just removed the owners from them and said they are mine now we are in control now which has sometimes huge security implications for customers because huge huge businesses run on wordpress who depend on the thing being consistent and transparent and not just arbitrary they really depend on things being predictable

tante: Yeah, I think it is relevant a bit to go into the details, not like the full turn by turn of everything, how things went down, but like, okay, he has beef with his other hosting company, WP Engine, who also offers you pay them and they host your WordPress, like so many other companies. Like there's a bunch of them, but WP Engine is very big. he complained that they don't invest enough into the open source project, which I think is probably fair. I think they could invest more given the money they make, but whatever. That's one conversation. It's beef between two corporations who gives a shit, whatever. But what he then took is when you said he took over plugins. Yes, those were plugins that WP Engine, the company that he said doesn't invest enough in the open source project, developed as open source projects. Like people WP Engine employed develop these very core WordPress plugins that many, many installations use as open source tools on the WordPress.org repository, like in the way that you're supposed to do things.

So the thing that he complains they're not doing enough, when they do it, he takes it away from them. Which is a really weird move that goes beyond just okay. I'm taking over our name or whatever but like it shows how inconsistent and how unstable his behavior is and you

Hagen: Yeah, there were important plugins, security-relevant plugins that people depended on, And then it's more out of control where if you wanted to develop or contribute to WordPress as a volunteer, like working for free on WordPress, you had to say, no, I'm not affiliated with WP Engine in any way, shape or form, which many of the contributors couldn't do because most of them are developers for a living. They develop for WordPress for free, but they also work for clients who might be hosted on WP Engine. So it all felt a bit like the whole joke that happened at one point where the artist Anish Kapoor got exclusive rights to Vantablack, this one very, very black color. And he got the exclusive usage rights for art. and then Stuart Hicks, I think his name, created the Pink is Pink and said, everybody can use it, it's fine, unless you're Anish Kapoor or affiliated with Anish Kapoor, which was a joke because it's a ridiculous thing to say. That was never meant to be taken seriously. And Matt Malnweck just did it like he had a tantrum in kindergarten.

tante: And the interesting thing is that Matt Mullenweg didn't do that as the CEO of Automattic. He did that as the foundation. Like the platform that like webpress.org is supposed to be where the foundation does their thing. It's where the foundation's bylaws lie and where their rules are and all these things that are relevant for what the foundation wants to do. He changed it on the foundation website. He didn't change it on like you could as a business person, you could say, OK, we're here at Automatic. And if you also work with WP Engine, we don't want to want to do business with you. That's fine. Like you also host on WP Engine. You can't buy more hosting space on our website, whatever. That's not what he did. And I think there's distinction between automatic where you think, okay, you're a private business. Do whatever the fuck you want. And like the foundation that so many people on this planet like kind of trusted. It's like they use, I mean, many people use WordPress without knowing they use WordPress. They're just happy that they have a thing where they can type it to a website and it appears. But like people who make those decisions for them, like the agencies or developers who set those things up, like they trust in WordPress because they thought there's this entity that covers the project and it can't die, it can't go away. There's processes of how things work. There are rules that govern how things work. And now you see that this one guy who has a company can just put on a different hat. Now he's the foundation. And now he says, yeah, you're fucked now, by the way. That's what distinguishes it from, for example, Elon Musk, who also is an unstable white dude. But there you know, like, yeah, it's his company, whatever. Do your thing. Ruin Twitter.

Hagen: Yeah, we knew Twitter was fucked the moment he bought it.Like, it was clear from the beginning that this was going to happen the way it did.

tante: Everyone knew, okay, Twitter is a company. He bought the company. Now it's his toy, and he can break it if he wants to. But the difference in WordPress is that there was supposedly this thing that is bigger than the guy that WordPress could survive whatever Matt Mullenwick did. And now people realize, oh, he has his fingers basically everywhere. So if he wants to kill this thing, he can't.

Hagen: That led to a lot of, I think, understandable outcry. I was one of the people who participated in the outcry, and then I decided, you know, I can't support this behavior, and it's not good for independent web publishing for one platform to have this amount of power. And then I tried switching away from WordPress, and that is, on one side, it is pretty easy because WordPress is so big and powerful that every other CMS out there basically has an export tool. It is very easy to get your data out of WordPress and to get it into another CMS should you choose to leave. The problem is that over the last decade, two decades, the ecosystem of WordPress became so powerful and so influential that basically no other CMS seems to have survived that can compete on the feature set even a little bit. Like I tried a couple of things, like a couple of static site generators like Kirby, but now I switched to Ghost, which seems to be the other, not big, but bigger option. They are very focused on not doing traditional websites. They are very focused on blogging and newsletters with membership integrations and things like that.

tante: And they don't have comments, for example.

Hagen: No, they have comments for members. Like, if you are a member of the page, you get comments.

tante: Yeah, but if you want to host your own, you can't, Unless you have a third party that hosts it for you, and it's a whole different problem.

Hagen: Exactly. There are a lot of, like, you could implement Disqus comments. TalkYard is another option. You could self-host a TalkYard instance. I looked at it and decided, comments aren't that important, are they? That was basically my, because, yeah, no, I'm not going to do all of that. Write me an email. Don't write me an email. Just won't happen. Let's be honest. And this very podcast here will be hosted on Ghost because I decided to integrate it with my blog.

And it works as in it's possible to generate a podcast-compatible RSS feed. It's just not meant for this. Like, the MP3 length tag will be broken. No podcast client cares, but it technically is a broken feed. The audio will be pasted into the field that is usually reserved for the Facebook excerpt. It is very hacky, very broken. And for WordPress, there were many, many plugins. There were commercial plugins to host your podcast. There are projects like Podlove that allow you to host your podcast very, very easily. You don't have to touch any code at any point. It's just you install a plugin and it tells you what to do and then it's done. And even Ghost, the other option that is used by commercial entities that is used for professional publishing, like 404 Media, for example, publishes via Ghost. It just can't do that. And that's also a company who do their own hosting who have an income. like ghost is also a company like very similar structure with the commercial entity and the non-profit entity like wordpress and it just can't match it like and for me i just have a blog and and this podcast that's easy but i think you for instance have a multi-site wordpress installation is what you said is there any way for you to get out of wordpress at this point

tante: I mean, I could host like four versions, but if I were bigger and had like many active WordPress websites, I would at least have to like multiply the effort. Like I would have to say, okay, no, I'm hosting ghost or Drupal or whatever a few times. So I have to keep them all up to date and have all this code running extra that I need to need to take care of. It's harder, but I think that the multi-site thing isn't that much of a killer. But whenever I looked at alternatives, I realized, okay, it's a lot harder to run.

Like WordPress is, here's a bunch of PHP files, unzip the thing and drop it into your web server. And now it's done. It works. Like if you want to run Ghost, you have to run a Docker container. And if you run it without Docker, nothing works. It's broken.

Hagen: No, that's fine. I actually have to say I'm not running it through Docker because I hate Docker with all of my heart, I don't want to touch Docker.

tante: But , it's a technicality. Like I don't want to get too deep in the depth. but it's a lot more complex to run.

Hagen: And it's heavy on the system. I see the RAM demands. It's way worse than WordPress when it comes to efficiency. Like you need beefier hardware to run it.

tante: Yeah, I mean, hey, Ghost is, it's built for a different thing. I think it does that thing well. It's not a bad CMS at all. It works well, especially if you just want to run like a professional newsletter with for-pay stuff and whatever. I just pay Ghost to run it and it's fine. And I know I could export the thing and run it myself and things went badly but for many people that they chose wordpress exactly because they didn't have to do that because it was like okay i have this little i don't know you have a little football club or like just a bunch of friends who dick around with i don't know board games and you have a website and it's a thing blah blah you don't want to pay a lot of money and deal you just want a thing and drop it there and the web space is maybe even free because you don't need a lot of it, and you can do all the things you need to do. And that was a huge benefit for the very small variant. But you could also host WordPress as a very big and highly scalable thing. People probably know Ars Technica, which runs on WordPress. This is a website that gets more than a few hits every second. They had to think about how to run it, but they run it through WordPress, and it works well.

Hagen: Vox Media is moving all of their properties to WordPress. They're using WordPress.com, like the commercial instance, but Vox Media is completely switching to WordPress from their own CMS because they say it's not worth it to maintain your own CMS.

tante: It absolutely isn't like there's so many great cms options but like if you have a staff of professional developers and designers who can do your thing like your vox media whatever they have a tag team. You can pick from a whole bunch of CMS versions and CMS solutions that have different properties and there's great stuff from like the, okay, here's just building blocks to build your own thing to add. This is basically done. You just slap a theme on it. It's all there. I think WordPress had this very specific niche of for people who don't have like that staff. Browse through the free themes. Then you realize, okay, they're not fully there, but then you pay like a hundred bucks then you have a very well-looking theme that is greatly integrated it also just asks you okay here's your social media thingies enter them the ones you have and it gets the icons and it looks all very professional like what wordpress i think brought was every website you build with wordpress unless you start really messing around with it looks very professional or can look very professional with very very little work that was great for many people who don't know how to design a website. Like the navigation is probably still fucked because they have no idea how to set that up reasonably. But the website looks good. Like it was easy to get to a point where you could, you have this thing, it looks nice, and then you have new, I don't know, demands. You want to have a calendar on your website for events that your association hosts or whatever. And it's one plugin. You click it in, now you have a calendar view, and it does the thing. And it has more features than you probably need because it can also do ticketing and whatever. It had this large infrastructure that was so easy accessible. But that all rested on the trust in the foundation and the fact that everyone believed, okay, WordPress.org is where all the plugins live, and I could get it from there, and they're open source, and blah, blah, blah, blah, blah. But that's really what Matt Malamook's actions shown to not to be true. At some point, he blocked people from WP Engine to access the website. Actually, his legal argument at some point to block people was, this is my website and I get to decide. Like, they can publish their shit wherever.

This is my website and I decide what goes. Like, as a person, it's mine. It's not the foundation. He never said, I do that as the director of the foundation. that wordpress.org is mine brackets matt malewick's website there are like a few hacks where people change like have user scripts whenever you go to wordpress.org it inserts a big matt's website on the top because that's what it is it's one dude who can just decide to change things and especially for if you just have a website that runs on wordpress that doesn't need to concern you too much you have your thing it works you probably don't even know what that this thing kind of affects you. But if you are like a professional, like you have an agency that builds websites for businesses and whoever, and now you no longer can trust the platform that you built your career on and built your business on, you cannot trust that the plugins and building blocks you need will be there or that you will still be able to access it. If you're a dick on Twitter to Matt Mullenwick, who still uses that, he might just block you. You don't know. Could be. It's reasonable. Like he, He blocked all kinds of people like the WordPress Foundation had like a Slack where all the people who built open source plugins could get access and they talked about it and like overarching projects were in there as well. Like, for example, the accessibility group for WordPress, like developers who try to make WordPress more accessible. They met there and organized there. And whenever, like in Slack, when these things, when these decisions became public, people asked Matt, like, what the fuck is going on? What does this box even mean? And whenever you criticize him, he just kick you out of the Slack. Like it's all just his toy. And I think this isn't just bad for like WordPress as a thing, which, yeah, trust in that infrastructure is really shaken by the people who know that this infrastructure is what holds estimates, say, 40% of the web together. But I think it's also it shows a larger problem with with open source projects that often forget that they are organizations and that need governance and that too often like to Rest on like a strong guy like the founder or whoever who gets to decide what's going on And yeah, if that person is nice, that's great. But when your king gets mad things get bad Oh that rhymed cool you

Hagen: And I think that's the reason why I wanted to have this talk in the first place because it raises a lot of questions of where to go from here. Because I think we laid out pretty well why WordPress isn't just a CMS. It's basically infrastructure at this point. Because the whole web ecosystem for indie publishing, if you don't want to pay your own developers, is just built around WordPress. This is the one you want to have if you don't want to tinker, if you just want something that works and can work in every direction that you want, from a portfolio to a blog to a podcast. doesn't matter wordpress can do it and we just don't have an alternative to go from here so what does have to happen for us to protect things like this infrastructure i mean you also mentioned linux and we are all like yeah linus torvalds just decides and to be fair most times it has gone well more or less linux works i mean there's a reason why basically every web server runs on it but it doesn't have to be this way it can be way more fucked up than it is like i know there are a lot of problems there as well but how do we protect essential infrastructure i mean just imagine something like this would happen with wikipedia which is also has been one guy jimmy wales's face now is a foundation and it seems to work more or less pretty well and every country is a bit different but think about these huge open source projects that we have like that large chunks of the internet and with us all moving to the internet like large chunks of human civilization run on and then it comes down to one dude how do we fix this

tante: Well, I mean, even for WordPress, people are trying to work on things. Like there was a fork a few years ago called ClassicPress. They didn't like the new editor that WordPress introduced, so they kind of stuck with the old one, but it never really gained any traction. But there's now a project called AspirePress. And what they're trying to do is not just fork the WordPress code, like the PHP for the WordPress software. They're trying to A, build governance infrastructure, like how can this project be run democratically and reliably, and B, to kind of replace WordPress.org. like build a plugin, themes, blah, blah, blah, infrastructure that is managed by a community where not one person has all the power, where there's like transparency. But it's a long time. It takes a long time to set that stuff up, not just because of the programming and the technical challenges, but also like, okay, what kind of organizational structure do you want? Which juristic region do you want to put it in? What's the right legal shape it should take? It takes a long time because these people also all have to work. Like they do that in their free time because they need to pay rent and all this eat and all those annoying things. But so there is movement to try to transform WordPress or whatever WordPress is now and take it away from Matt Mullenwick's hands. Because forking just code is easy, but that will not get anyone to change their behavior or give people an actual out.

Hagen: I think this is one of the important things, like just install a fork. Who cares? But a large chunk of the appeal of WordPress is the centralized infrastructure to some point, where all the plugins are at this one point. Like, it's easy to update because you can just point your update mechanisms at this one source of the true information to get the updates. You can just go through WordPress.org. So I think we need what you explained, that people are working on an alternative to WordPress.org, but I think it has to happen. There has to be this one central location where everyone agrees this is the source of our plugins and the source where we will do the development and it can't splinter away. That is the main problem that I see currently.

tante: Yeah, but I think that it's important to build these more robust structures for a bunch of reasons like a because mitmahal wik is a dick but also like what what program is called the bus factor like if your project is run by one guy and when that guy is run over by a bus the project is dead that's not smart especially if it's a core project that so many other people or other projects depend on so you want to spread like the the responsibility uh a bit but also you want to have checks and balances you want like the the reason we built a democratic systems the way we do is because we don't want to have one person who has all the power because power corrupts and that's just not a good situation to be in on the other hand like i think people were very comfortable with the setup even though i mean all the details were public people believed wordpress.org was owned by the foundation but in general people were happy with that setup because matt malwick did put a lot of money into the open source project like he paid for a lot of shit he paid for a lot of developers who work on the thing. He did a lot of outreach and advertising for WordPress. Like, it's not like he didn't do anything. He put a lot of his basically personal money behind the project.

Now, if you set up a different structure that goes away, you no longer have that. And will people who used to use WordPress for free, who didn't pay anything except for maybe like their neighbor's daughter who set up the WordPress for them a few bucks, will they start like subscribing to a foundation or donating to a foundation will that generate the kind of money you need to keep a project of that size going and that's that's really hard to say and i'm not sure if we as digital citizens if we want to call it that like or people who also live in the digital world if we do have the infrastructures to set those things up robustly like probably if it if if a a good alternative to wordpress.org forms they will get some buy-in and i think they will survive and they will maybe even thrive depending on what matt mullenwick does but how many of those foundations can we actively support given how people how donation structures work how much money people have to give like we have the same problem with firefox which is like the only free browser that is not actively run by an no longer true firefox also bought an advertising company and is now itself an advertising company but like if you don't want to run google's code that is optimized to what google wants if you want to have an alternative that's currently firefox it's basically the only other relevant rendering engine i mean there's whatever apple is doing but that's basically just trying to hold the web back so it doesn't work properly so they can sell more iPhones.

But Firefox is in the same position. There's a company that does weird shit, spends money in weird ways that gets a lot of outrage in the community, like what the fuck are you even doing? But do we really have any infrastructure that could take over? That could say, okay, we're going to fork Firefox and now continue to develop a browser that more aligns with what people want? I don't see that. And I think that that is the bigger question we have right now, saying, okay, we have all these critical infrastructures that we rely on, and browser is one thing, operating system is another thing, and then tools like publication tools, and all these different small tools that people kind of need and want, whatever, and who can carry that? Who can be like this scaffolding that holds that up? And currently, what we're hoping that people who do that out of of love because they like running the thing do that, but maybe they get, at some point they get their family and kids and their free time kind of melts away or they get older, which is a problem with a lot of the open source projects where a lot of the people working on it are, their hair is gray.

Hagen: and they have to compete with commercial offerings. Like, the rapid development that happens if you can pay people to work on it will outpace it. And at some point, the software that is updated and updated and updated because people are paid to work on it will be, by definition, so much better than the free alternative that people only can work on in their free time that you basically don't have a choice to not use the commercial offering anymore. because it's so far ahead.

tante: the choice, but you pay for it. I run Linux on all my machines and yes, sometimes tools don't work. That's the price I have to pay. And that's annoying. But I think that it's very easy to ask for foundations and form the project as an entity. And many projects do that. Even Mastodon that we mentioned earlier, they have this gig, GmbH, this public benefit company in Germany than like a US non-profit that you can donate to and whatever.

Like there's structures that could survive, like the person who started the project going away or no longer caring about it or no longer being able to run it, whatever. But the question really is, does that scale and how far does that scale? Like if you gave five bucks to every open source project that affects you, that you rely on in some shape or form, you probably wouldn't eat. That's just how things work. So how can we build better infrastructures that can last and who can maybe, where can we get the funds for it? And I think that's an unsolved problem and it works for bigger things like Linux gets funded because there's corporate donors that give a lot of money and like there's a lot of money slashing around to do that because corporations have a large interest in keeping Linux free and open and running and whatever.

Hagen: At the same time, like the thing that WordPress.com did, like they developed WordPress. But of course, the incentives are not the same that we want because we are like small, independent web publishers who might not even do it for money just because we want to. We love the freedom to publish ourselves. Like I make my money as a journalist, but of course, that has to go through an editorial process. that's different than I can publish my thoughts on my domain, on my server, which is all mine, that is owned by myself. But I will not make money for any of these companies. Their development priorities will never be, let's support the indie publishers that do it for fun, for the love of the game. So there will always be different incentives.

tante: Yeah, the question is how do we align these incentives and how do we also spread like, there is money sloshing around in the open source space, but where does it go and where is it invested in? Like currently, I mean, not for long, but currently Mozilla has a lot of money and they invested in basically AI right now and buying advertising companies, which is not what I'd want them to do, but okay, like they are a corporation, whatever.

Hagen: And they have a real reason to be afraid that the money is going to go away because all of the money they have is because google pays them to be the exclusive search engine and it's pretty clear that google will not be allowed by law to continue that practice which is as a whole i think a good thing but also a problem like okay apple gets 20 billion dollars a year from google for doing nothing but they don't need the 20 billion dollars a year mozilla will not be able to survive the

tante: Exactly. I totally understand that they need to find other ways to fund themselves. And that is the challenge. I think they made a few bad bets, but I understand their struggle. The struggle is real and the danger is very real.
Hagen: way it did before without the Google money. So they need to find other revenue sources.And they're not the only, like they're very big and very visible player in that space. But so many projects have similar problems. And there are, of course, initiatives that try to address that. Germany has the sovereign tech fund, but that's a bit different. You can apply for a grant there, but you don't get like continuous funding for basically forever. You can't be in there and now you know that you will be able to fund developers for the foreseeable future or whatever. Most of these things are project-based. You say, we want to develop this thing or we want to do this security audit. It's a project. It has a time frame.

It has an effort that needs to go into it. Then you get the funds for it. The EU has projects for that as well. There's a lot of stuff where you can apply for grants, but I think for many, especially for these critical infrastructure components, I think that's not enough. I think that's not how you run these projects. Because if you say you have an open source project, and the open source project has 10 features that you have to maintain, that is a lot of work. It's annoying. People write bug reports. They are sometimes dumb, sometimes smart. Sometimes you have to put a lot of work of fixing them. Now you can apply for a project grant. You say, hey, I'm going to develop these two extra features, and I need, I don't know, 100,000 bucks to do it, and it's going to take two years. Now you're paid for two years. Great. When the two years are done, you have no funding, but you have 12 features to maintain now. Like the load grows by applying for these grants. Like your task gets bigger, but the money doesn't continuously come.

Hagen: Yeah, and these infrastructure tools will never be done.

tante: No.

Hagen: Like, we need new web technologies to be implemented. I mean, remember the time when websites needed to become responsive, which was a good thing as a whole, but it took a whole lot of amount of work to make every website responsive. And this will continue forever.

tante: Yeah, and there are security updates, and you want, like, people's needs change. And that's not a bad thing. good that things are fluent that people can work on things that people can experiment with things but i don't know if we currently have good infrastructures that can support those projects in the long term and i think this would be for example something where governments could fit in and i'm not saying the government should we should have a government browser that's not the point but governments have the funds to set these things up to have like okay we you set up an agency that that basically takes over certain, that takes over, but like funds certain things indefinitely or whatever. Like there are ideas how you could set these things up, but they cost money and it's very hard to legitimize a lot of the time. Like you're spending tax dollars, if you believe that the thing exists, and you don't get like a direct benefit. You don't get economic activity,
like you're not funding your own little companies that will now become unicorns or whatever. You're just giving a few people money who work on shit and no one makes any money. That's sometimes very hard for neoliberal governments to understand and make sense of.

Hagen: which I would argue against because tools like WordPress are essential for small businesses who can't afford to have web developers. And if you wanted to do it in a way where the government doesn't control the trajectory there are tools for that as well like germany has the public broadcasters who are set up in a way where the government they're as far removed from the process as possible so there are ways to do the funding but i see one problem there as someone who grew up in leftist spaces the tendency to want to discuss everything as a democracy i don't know if the term is in english also plenum where everybody has a vote and everyone wants to discuss everything to death and nothing ever gets done. I feel like at least at the starting of something like this, it might still be a good way to have one person who just points at the way and this is the trajectory we go and then we need to find a way to transfer it into, okay, this is becoming infrastructure. Now we need to have a governance.

tante: Exactly. When you're just dicking around with a thing, you don't need all these processes. You don't need a foundation or that. You're just dicking around with a thing and you don't even know if it's relevant yet. But at some point, as you said, when it turns to infrastructure, maybe you need to rethink what you're doing.
Hagen: And I think at that point the problem is different as well because then it's important enough for enough people to care about it so that there will be a structure that can vote on it because it's important enough for everyone to be actually involved in it. A vote doesn't necessarily mean that every little feature gets voted on by everyone. Like there are structures, either you delegate votes to certain people or you vote on like a product management council and they actually make the calls based on the resources the project has. Like we all like every project to develop exactly the feature we want, but they also have like, okay, we have these many hours of developer time and so we have to prioritize. And I think that that's still doable, even though, yes, it's less efficient than one guy just deciding what he wants to do.

But I think for infrastructures, that is very important. And especially if we then move it towards, hey, this is infrastructure that the public funds, at least to a certain degree, then there needs to be a democratic process. The problem with these projects sometimes is that the people who can vote on these projects are oftentimes only developers. If you have code in the main line, then you might have some influence, maybe like a formal vote. But for example, with WordPress, I think it's relevant to say that if it runs 40% of the internet, there's millions of people who can't write a line of PHP or JavaScript or CSS code, but who are deeply affected by what happens with this project, who have a very important stake in this. And who should, to a certain degree, have a say, I think. It's kind of weird that we don't demand more often that as users, who have no idea how the project works, but the project wants us to use it, wants us to put parts of our life onto it, wants to be infrastructure for parts of our life, that we should have some form of representation, some form of influence. And it's hard to do and it's inefficient. And I think it hasn't been done very well yet. But I think that is something that we should strive for if we want to move towards infrastructures that don't go off the rails, either because of one weird dude or because of like Mozilla, an organization that's really just grasping at straws to find some magic money tree again to keep on doing what they are doing.

Hagen: Yeah, I think you nailed a pretty big problem that I have with a lot of open source tools is they very often seem to be playgrounds for developers, which I get because they started out as tinkering projects from developers. I get why it went there. But yeah, a lot of them don't seem to have a lot of involvement of UX designers, and it shows. Also, this is something I noticed when trying moving around CMSs. I'm not a programmer, but I can read a JavaScript file. I can understand the code if I see it, even if I can't write it myself. And the amount of times I had to open a text editor was like, I just want to run a blog and a podcast, which are pretty common things. I'm not doing anything complicated. I'm not doing code injection. I'm not even trying to implement a third-party comment engine. I just want to run the fucking thing. And it's hard, man. It's still hard. I don't know if this is something that could be mandated if you had a governance structure like this, where you say, yeah, the budget is X man hours for development, X man hours for usability design, something like that.

tante: If you had like a structure that has like, I mean, it's the same for every company. If you have a company and you're developing an IT product, you have product management and they have to like see, okay, how many hours of designer time do we have? How many hours of developer time do we have? What are the features that we need? and you prioritize them, and then you kind of try to make it work with the resources that you have. I think the thing is, currently we have commercial things. You buy it, and then you probably these days you just rent it a lot of the time. Or you have open source project. But open source project just means, okay, the license tells you you can have the code, which for many people doesn't mean shit. Like, even if the readme gave them the commands to run, they wouldn't know where to put those commands. And I'm not talking down on them. I know how to do that because I studied that. That's my day job. But many people don't because they have interesting things to do with their life. Which is why they maybe want to write. Because they want to write about the interesting things they do. Or do a podcast about it or whatever. And I think maybe we need to reframe the kinds of projects that we just talked about. They have this infrastructure quality that go beyond, okay, it's just an open source project. If it dies, okay, that happens. But that have reached a level of like, I hate the term, but I don't have a better one. They have enough market share or relevance that we can't let that happen.

And then say, okay, building up wherever the money comes from. There's an infrastructure that you can attach to and get some money out of, but that requires you to have certain things in place, like certain formal governance structures, user participation, I don't know. The structure even gives you best practices. Say, okay, hey, if you have no idea how to set that up, here. Here's three templates that we know that work. And I think it is an interesting task to kind of define what this thing is.

tante: What a project like the one we just described, what makes a project like this, And what should it get from wherever?

Hagen: And I think that is actually doable. could put it in writing as the requirements for this to happen. You can define infrastructure as like market share you should define pillars for value for democracy i don't think every tool needs it but i think a wordpress that helps everyone to become a publisher and publish their thoughts on the web it's important for democracy to give everyone a voice firefox the same to be able to browse the internet without the involvement of companies would be something like that wikipedia would be something like that linux would be something like that but I can think of a lot of tools where I don't think that's that necessary at this point.

tante: The way I think about it, this doesn't take over projects. I think it doesn't make sense to say, we're going to take this. It shouldn't be like the job of that kind of institution, but that kind of institution should offer guidance and should offer projects that realize that they might have grown into the need for this. Allow them to apply for us. We're going to check if you really fit into this, because I think it's hard to really say.

Market share is hard to measure sometimes. user amount of users is also hard to do, but you can apply. There's like a transparent process where people check it and make a decision whether you fit or not. And if not, what's missing, which is also being transparent, like this is why you couldn't fit in, but well, and then, you know, okay, you have a contract and that contract has requirements saying, okay, you need to guarantee this. You need to guarantee user involvement. And– yeah.
Hagen: user involvement wage caps. So if you say you need the funding, you can't have CEO salaries that reach into the millions. That just won't work. But I think coming back to Germany, we have examples like this in the real world where you can have voluntary firefighters, but if you don't find enough people, the state is forced to hire firefighters. Then they have to be commercial firefighters. And I think this is a nice equivalent. Like, if a project works like this and it's fine and it doesn't need the funding, sure, run it your way. You're funded now.

Hagen: But at some point, if you need funding, no, there are rules that you have to follow then. But also, this will come with a guarantee from the government.
tante: We take you under our wing. You're cool now. You still might want to do more fundraising because you want to do more, whatever. But we guarantee a certain baseline that has been agreed upon beforehand. I don't think it solves all the issues. but I think it could create a space where digital infrastructures or like digital commons and not data commons like Wikipedia or Wikidata or that kind of stuff but like commons of infrastructures But yeah, that would be great.

tante: that you can use which is software which could potentially even be compute I'm a big fan of setting up compute like data centers in that way so you can have like a fair access to certain amount of maybe space or computer or whatever you want to call it. But I guess the second step.

Hagen: I remember the times where the five Euros/month server instance to run my WordPress would have been tough on me. Every citizen should be allowed an amount of space to publish their thoughts on the web. I 100% agree that there should be this public infrastructure.

tante: Yeah. And at some point you grow out of it or you want something that the thing doesn't provide, That's fine. The market offers everything you want. Yeah, I think it's just important for us, with our lives being so heavily based on digital infrastructures and so large parts of our lives being run on these digital infrastructures, that we find ways to keep those infrastructures alive outside of corporate control. And the thing that we do with open source currently, I think it rests, it's kind of a house of cards. And I think in 10 years, we'll see a lot of projects having big trouble to stay alive because the people who ran it and to started it, aged out of it or died, which I already see in certain projects that have a very old developer base. And this is something that we need to work on now. So in 10 years, we don't sit there and say, okay let's let's all go back to microsoft they'll run our lives

Hagen: you just invented communism.

tante: i tend to do that.

Hagen: yeah it happens to me as well yeah

tante: let's call it social social democracy it sounds more palatable to the liberals

Hagen: but in germany there's there are people attached to that term that i don't want to be associated with necessarily

tante: yeah yeah yeah true true true

Hagen: One thing we have to all be fine with and that i would be fine with is that it essentially would be choosing winners. Like you are infrastructure now because we don't need multiple CMSs like this being funded because you are public good now. We only need one highway system to talk in a way that the liberals will love. And that's one thing I do not get. Like the government loves to pave the shit out of this country for the public good because they say it's good for business, it's good for businesses. But that just doesn't happen in the digital world. There they love funding some weird startups, some weird helicopter plane companies who claim that they're the next taxi. But they will not fund common infrastructure.

tante: I think that you used a very important term here, picking winners. And yes, that's what it would be. It doesn't make sense to give every CMS on the planet all the money. maybe a bunch of them because they serve different needs or whatever but not many so that's where i think some of the organizations that hand out money also have a hard time to do that like because i think say you are that organization and you're picking winners you probably look for a certain level of consistency this is basically our target platform that we think most of the things should.

But I think that you used a very important term here, picking winners.work on because we think that is important. This whole idea of picking winners is, I think, A, important, but B, also not trivial. To structure this organization in a way that the way they pick winners makes sense and doesn't just create friction and alienation where, okay, they pick this project and now the others hate the other project because it needs to be a very clearly outlined reasoning behind it.

Say, we didn't fund you because you suck or whatever, but we have characteristics that we look for in a project that your project doesn't serve. And I think that in order to get that going, you'd need a lot of money to start. Like, it's not something that you can do with like, here's 10 million. Like 10 million sounds like a lot of money, but it's not. Not if you want to develop software. It's ridiculously little, to be honest.

Hagen: True but I don't think the amounts will be as large as like we are also used to the big tech amounts and I don't think we need that kind of money as well

tante: It's hard to say depending on what you want to take over. If you want to take over browser development, that is expensive. Like that is fucking expensive because it's hard. It's one of the hardest things you can build today, I think.

Hagen: sure but I don't think this is something that should happen on the country level but at least EU level maybe it would make sense to work together with the US as well at that point honestly especially with a browser and with Linux you could make a defense argument and get some of that sweet, sweet NATO money if you really wanted to.

tante: That's true. That's true. But on the other, Europe loves to play innovation with their legislation or whatever. We are the first with the AI, blah, blah, blah.

Hagen: Which they weren't. Just to remember. China had the first AI law, not the EU.

tante: Exactly, exactly. But China are the bad people, you know. I was set. Yeah, but yeah, they were first.

Hagen: I never said it was good, and I wouldn't agree that, and I would agree that China does a lot of bad things, and they just were first at this point.

tante: But like for Europe's size, if you're the EU, the money you need to put into this thing would be negligible. It wouldn't hurt that much. Of course, you'd have to cut down on something. You're no longer funding blockchain shit or whatever. I don't give a shit. Find some money.

Hagen: Just don't fucking build the highway that is running right in front of my door that nobody fucking needs. And that's going to destroy a lot of clubs.

tante: Yeah. But like finding a significant amount of money that you can reasonably say, hey, we can carry a certain amount of load. And probably it's not going to be the Linux kernel. I don't think the Linux kernel wants that. But there's a lot of things where you could look into, okay, what do we want? And there's a few hard decisions where things that I personally would probably like to be funded probably won't be. Like I think the GNOME desktop for Linux, like the free software desktop, should be funded. But I think it's not having enough traction, not enough relevance for people. But I can see a lot of things that would fit and that could where you could try out those things. And many projects maybe don't need billions of money. Maybe they just need a few thousand bucks a month, a few thousand bucks a month or a few thousand bucks a year to keep going in a reasonable way where a person says, hey, I just need like half my paycheck to come out of this and this project is golden. And we can guarantee that this that it exists because I no longer have to do it in my free time or just thinking about these kinds of initiatives that go beyond this whole. Here's a project. And I know that even with government, it's hard to do because they always just they have their funding periods and it's either a year or maybe four years, whatever. And then everything gets reshuffled again.

Hagen: Yeah, but that's why you need to build agencies that like, I mentioned the public broadcasters. There are other examples of just institutions that run parallel to the government that can be influenced by the government, but who do their thing and who aren't political institutions in themselves. And one last thought I wanted to bring into it with the whole, yeah, it would be choosing winners thing. One requirement would be it's open source. Okay, it's a winner, but somebody else could build on it. And even to make a capitalist argument in the end, I think that could even bring a lot of business in. Like, let's go back to the Linux kernel and what Valve is currently doing with their proton layer. And what the fuck? Why are we tying ourselves to Microsoft? Just let us do our own thing and we can build on this free thing and do some software magic that can be done by a few dozen people. Like, Valve isn't a big company. And now every game runs on Linux. And it's better than Windows. It can run on small handhelds. It just seems to work.

tante: Yeah, like I don't super love the term, but it would be a great antidote against what Cory Doctorow calls "enshitification".

Hagen: And that's because somebody else funded the infrastructure of Linux. And that would happen with all of these things. You could take what the infrastructure fund has funded and do your own thing on it. And it would save you a ton of money because you wouldn't have to start from the beginning.

tante: Because these projects wouldn't need to "enshitify", because they know that they're taken care of. That is what we all get so tired of in this digital life where you find a cool tool and you like it. This really solves a thing that I had. And now, like even WordPress, even months ago, they started putting AI everywhere. And now it wants to generate my posts for me.Like, no, I want to write this thing. That's why I have this. I don't want. That's what it's for.

Hagen: that was the whole idea! That's like buying drawing supplies and it's like yeah we can draw your image … but i bought them because i want to draw for fun like i don't publish blog posts as a business it’s literally what i want to write that I don't make money on because otherwise i would pitch it to make money on it it's literally the things i want to do for fun let me have my fun

tante: Yeah. Here is a joint, it smokes itself. Exactly.

Hagen: the important term that should carry over is infrastructure and people have to see tools like WordPress like Firefox as essential infrastructure and not a piece of software it's like the things we need to run the internet on hich means to run our current form of civilization on.

tante: Yeah, there's a great book by Dev Chakra. It's called How Infrastructure Works, The System Shaping Our World. I recommend everyone to read it. It's brilliant. And it really gets down to, we think about infrastructure, this is this technical thing. and some engineers built or whatever, but she reframes it. No, infrastructure is a gift we as a society give to us.

It's something we give to us and our futures and our kids and whoever. It's a thing that makes everyone's lives and everyone's experiences better if we build them. I think infrastructure should be open source, but so far we often stopped at, okay, now it's open source, now it's fine. And I think it's not enough. And the next few years should be about how to move further than just open source.

Hagen: And I think the description of infrastructure is pretty good because I wouldn't have time to do this podcast or to write a blog post if I had to shovel my own shit because I wouldn't be connected to the sewer system. It frees us to do the things we love because we don't have to do them. Somebody else did it for us.

tante: Yeah, exactly.

Hagen: And with that, I think we're done.

tante: Yes, thank you for the invitation.

Hagen: And that's it for the first episode of this new podcast. The book that Tante mentioned is of course linked in the show notes. And to be perfectly honest, I hate call to actions. So it's just bye.