tolarianwiz
Joined 6 years ago
Karma
160
Submissions
44
Member since
Jul 23
[email protected]
View on Hacker NewsRecent submissions
Comment · 3 months ago
Thanks! Yeah, I think project/contributor discovery are closely related to this, but I haven't built that part yet since I want to get the basic community model right first. One reason I wanted communities to be tied to repos is that, over time, you get a much better signal than stars alone. Out of curiosity, was the bigger problem for you finding contributors for your own project, or finding interesting projects to contribute to?
Comment · 3 months ago
Hi HN, I built Earde because none of the community tools I was using felt like it was designed for open-source projects.
Discord is good for realtime chat, Discourse and Zulip are better for things that should be preserved, but I kept wanting one place that did both and was actually tied to the project itself.
So on Earde you connect a public GitHub repo and get:
-live chat -a public forum indexed by search engines -turning useful chat conversations into permanent forum threads -connections between related project communities, so the same discussion can be shared between them in one click -usable directly in the browser -one feed for all the project communities you follow
I know you can approximate most of these things by combining existing tools, the experiment here is whether they make more sense as one thing, specifically for OSS projects.
It's still very early, so I'm interested in what feels wrong or missing, especially from maintainers who already have a Discord/Zulip/Discourse/etc. for their project.
You can browse the app without creating an account here: https://www.earde.com/feed And here's a demo of a public technical thread shared between communities: https://www.earde.com/c/northstar/t/116-client-enters-a-reco...
Comment · 3 months ago
Yes, absolutely, that's probably the biggest risk. Earde has to be useful to a project on its own first (through live chat, indexed forum, chat promotion). Our hypothesis is that we need to onboard small clusters of related projects, rather than accumulate isolated projects and hope that the network eventually connects. Whether the standaolne value is enough to get those first people onboard is the cold-start problem we still need to validate.
Comment · 5 months ago
Thanks, I’ll look into those cases.
I think the hard part is that Discord solves presence, Reddit solves discovery, and forums solve memory/structure, but none of them combines all three very well.
The question is whether the pain is strong enough to justify trying to build a new kind of platform that combines those three things gracefully.
Comment · 5 months ago
Thanks. Have you seen any communities that are actively struggling with this right now?
I’m trying to understand whether this is just an abstract annoyance or whether there are actual communities looking for a better platform.