Replies: 1 comment
-
|
Great, thanks, that's helpful. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
We're planning to build a public telemetry site at bitterbot.net showing what the Bitterbot P2P mesh is actually doing in real time. The goal is to make the network legible to both operators (is my node connecting? is skill X earning?) and prospective contributors (is this thing alive? what's the activity level?) without making anyone open a terminal.
Starting metrics we're considering:
Open questions we'd love feedback on:
Privacy posture. We're defaulting to aggregated-only — no peer IDs, no IPs, no approximate geolocation. Does that match what you'd want as an operator? Is there any individual-node data that would be useful enough to justify explicit opt-in?
Sybil resistance. Raw node count is trivial to inflate. We're leaning toward "active nodes" = uptime >24h + at least one verified skill execution in the last 7 days. Better ideas welcome.
Pull vs. push. Edge nodes already gossip telemetry to management nodes. We could (a) have the site scrape from a management-tier aggregator every minute, or (b) ingest a live event stream. Curious what people have used for similar dashboards.
Scope. We want to resist dashboard-creep. If you could only see three metrics above the fold, which three?
Historical depth. 30 days? 90? Forever with downsampling after 30d? How important is full history for skill earnings?
Not building in the open yet — PRs welcome once there's scaffolding, but right now we just want to make sure we're not about to build the wrong thing. Drop thoughts.
Beta Was this translation helpful? Give feedback.
All reactions