🎛️🔬🧬 Redesigning Hylo: When community stewards guide platform evolution
A Q+A with Hylo Co-Founder Aaron Brodeur and a big report on our collaboration
We are seeking a Social Media Fellow, and we are still looking for an Open Source Developer for PSI, a Head of Engineering for Local Lab, and an Operations Fellow.
In fall 2023, New_ Public partnered with Hylo, a platform that is, in the words of Co-Founder Aaron Brodeur, “a place to build community and get stuff done.” Hylo combines tools for communication, project management, events coordination, and more. We interviewed community stewards who use Hylo and sketched out new concepts, which we then had the stewards appraise. Much of what we learned was eventually incorporated into our work, and Hylo was influenced by the collaboration as well, culminating in their recently released redesign.
To celebrate this milestone, we’ve compiled our findings into a large report titled “Building social media with prosocial norms” which we’re sharing here for the first time, and I also interviewed Aaron to learn more about Hylo’s work, redesign, and theory of change.
–Josh Kramer, Head of Editorial, New_ Public
Who Hylo is serving:
We're focusing on people who are coordinating or developing patterns of coordination that are novel, that have to do with regeneration, agriculture, or making healthy social spaces and public spaces. And the reason we're focusing on those verticals is because we're making a broadly useful tool. We have to pick an entry point.
We didn’t pick business because that's not specific enough. We picked these people because they're trying to build a better world, reaching for a tool that doesn’t exist. We want to know what they’re reaching for, in the moment of their need, and design that.
One of our biggest and longest term communities is Harvard's Planetary Health Alliance. They have thousands of folks and they've used Hylo to do all sorts of really interesting things. Last year they used it to coordinate the development and design of their delegation at the United Nations. As part of that they had to prepare a document that ended up being 50 pages. They collaborated as a community to get the document done, and to figure out how to get there, because many people in the community were also attending.
It's things like that that I get excited about, because as a designer, that's the outcome I'm looking for: something happening in the real world that I think is positive, prosocial, and creating a future worth living in.
Hylo’s origins and how Hylo operates:
Tibet Sprague and I built a solar company that exited for over $150 million. Which at the time to us was like, “wow, we made it.” We climbed Startup Mountain and planted our flag at the top. But we also saw the underbelly of that process. Prior to that we’d experienced a form of private equity capture that happens to a lot of folks.
In order to avoid that capture, we've structured Hylo as a non-profit open source system. And we’re hoping that eventually the community — and we’ve been seeing it, increasingly — steps up to take over and to govern it as the people. We are seeing that somewhat, but it's going to take time. In the meantime, we've been building Hylo through partnerships. Usually somebody's about to build an app and they come to us and they're like, "What about my idea? Can you tell me what to avoid?"
We're very picky about who to work with and why. So that's how we bring in revenue right now. We do recognize that in order for the platform to sustain itself governmentally, for the people to step up and take control, there needs to be some resource flow beyond that partnership system.
So we are in the midst of building out tools like letting people charge for membership to the groups and some small tax goes to the Hylo Foundation. We do recognize that that sort of revenue stream is necessary, because partnership work is not a dependable source of revenue. It's not something you can cultivate and rely on. It's kind of spotty.
Working with New_ Public and Hylo’s approach to content moderation:
Having a team of people who are not just doing the interviews, but turning those interviews into plausible, reasonable ideas, developing those into wireframes, showing those to folks, getting feedback together as a community — it was so good. The insights that were generated definitely were integrated into Hylo.
For instance, now we have community agreements and community moderation, tied in with roles and responsibilities.
We took a lot of the ideas, and we basically created a moderation system where, instead of a moderator who makes a decision on behalf of the group, the community gets to decide together whether or not something is inappropriate. That decision is also public.
Say you posted a picture that is inappropriate in the community. I see it and instead of flagging it, I say, “this breaks an agreement” and I have to describe how it's broken. I'm not required to, but I can say how to fix it. Then I hit "post.” That puts a screen over the post, and it blurs out the content very slightly. You can still see what's going on, but it's not fully defined.
Then the next person comes along, sees that screen, clicks on the moderation flag that appears and they see “Aaron says that this breaks this agreement” and that person can say whether they agree with me. If they disagree with me, the flag goes away. If they agree with me, now it will take two people to remove the blur. And as more and more people agree with me, the blur becomes increased until the post becomes minimized and then it disappears from the main view. A lot of that came out of the insights generated during the stewardship part of our work together.
Roles and responsibilities in the redesigned Hylo:
Our roles and responsibilities are also influenced by our work with New_ Public. All of the responsibilities of the group have been atomized. A group starts out with default roles: moderator, administrator, and host. The administrator has all of the responsibilities assigned to it and the moderator has most of the abilities assigned to it. The host role allows you to invite people, make pinned posts, and remove posts.
You can also create a new role, name it anything, and create new responsibilities that are not system responsibilities but are recorded and displayed to other folks. So you could say, “Josh is our accountant” and then you could also assign that accountant role to other people later. Everyone in the group can see who has responsibility in what dimension. That was definitely informed by our work together.
We're really excited because we're going to be testing out a lot of really different ways of being online. We're really interested to see what people do with it. The way it's emerged is totally distinct from any sort of product development experience that I've been a part of. I’ve built things that were in front of millions of people. I’ve been in really interesting environments, producing really good stuff, but never something that has emerged out of so many competing needs. There’s usually an underlying single point of need, or focus, whether it's profit or the major donor.
So yeah, there's some way that all of this is coming together, this redesign, that is like a beautiful organic expression. I'm excited to see how it lands.
The problems of scale and group size:
Two people is the smallest group size. If you and I need to communicate, we're not going to go through the trouble of setting up a Slack. No, I'm gonna reach out to you via email or text messages. But of course, if there's a third person, then it becomes tricky and we'll probably pick WhatsApp or something like that because we don't know if everyone's on the same device.
Then if our group grows and we stay on that system, very quickly, if it grows beyond 20 or 30 or 40 people, that mode becomes unmanageable. It's just kind of chaotic and some people check out. But then if that same group tries to migrate to somewhere else, that group usually loses steam.
Then there's platforms like Mighty, where you start with the intention that the group is going to be large, but you need a little bit of a group in a space like that to feel cozy, otherwise it can have an empty room effect. Those can grow quite large, but they also have a threshold. And then you have something like Reddit, or forum software, where hundreds of thousands of people can collaborate and hang out and be in a community. So we realized that there are these inflection points where a group needs to evolve, and if the environment doesn't allow it to, it can become stunted, like a pygmy tree growing in a salt flat. How do we design an interface that acknowledges that, anticipates that, and plans for that moment?
How Hylo builds for adjusting scale:
There are a lot of chicken and egg problems that every community leader comes to. We’re hoping to address some of those. Here's an example: We make it really simple to make a group. That's one of the priorities of the redesign and I think we achieved it. It's just one click.
Next is not having the interface be very complicated. Start with a chat. Make it easy to invite people and add people to the chat. And as they add complex pieces of content, whether it's a link or a photo, it becomes organized in the sidebar. So if I create an event, all of a sudden the calendar shows up. If I upload a bunch of pictures, now the gallery is there. It wasn't there before because there was no need for it. So the interface evolves as the group grows. It becomes more complex and is now customizable in a way that it wasn't before. You can drag things around and order them the way you want. That's one way in which we're trying to grow with the groups. We don't know what the ceiling is for this new type of pattern.
We're bolting on some organizational features and hoping that the combination of these two modalities, which is the stream of conversation plus posts, can support a lot of folks.
Why co-design is important to Hylo:
There are some shared team beliefs about why that’s important. One is that we recognize that anybody who builds something is incorporating their biases into what they build. Of course, their insights are necessary to build something, but the biases also provide the negative space out of which that thing emerges.
If there's no accountability for them, those biases can, with the power of internet technology, propagate across an entire culture. Facebook is an amazing example to me. If we really zoom in on what was happening at the human level when Facebook was created, a young man was new in a place he had never been to and wanted to become a part of it. He made something for his community to rate women, and those biases have propagated across the entire culture. As builders we don’t want to incorporate our blind spots — the bad parts we can't see.
The other component of why it's important is democracy. That’s the simple way to say it. Nobody should have this much power. Nobody should be able to buy the town square. We think it's the people on the ground who are actually engaged in producing a world worth living in who have the needs that will motivate a product worth being on.
What will make the redesign a success?
For me personally, I have two major success conditions. One is getting my parents on Hylo and chatting with them there instead of using iMessage. Because then the family can finally be united beyond the Android/iPhone divide.
I showed it to my dad again recently. He was like, "Woah, this is so much better. It's so much simpler. It's easy to use." I was like, okay, great. That's one success condition.
Beyond my personal desire to liberate my own family from Facebook Messenger, this year I really want to see more local activation. There’s a movement called bioregionalism. A lot of people are going to be using it this year to coordinate local stuff. And so I'm really excited to see all of the work of the last four years come into play. Does it make a stream clean? Can this software do that?
Those are two things I'm personally looking forward to. Figuring out how to make something that does good well is something we're constantly in conversation about, beyond the metrics of growth, engagement, and retention, all the normal metrics. How do we measure the other stuff? How do we measure whether or not technology is actually helping the process, or is it getting in the way?
Thanks Aaron! You can see all of New_ Public’s open jobs or join our Talent Pool here.
Putting some more plants in the ground this weekend,
–Josh
 | A guest post by
|