The Context Layer Needs a Facilitator
We need to tap into the institutional knowledge to make AI agents effective. But the most valuable knowledge is tacit: trapped in people's heads, pre-articulation, never written down. How do you surfa...
There's been a lot of talk about the "context layer", the institutional knowledge AI agents need to be effective. But the hardest knowledge is tacit, trapped in people's heads. Our MCP server and /harmonica-chat command let you launch and manage Harmonica sessions that surface it, straight from CLI.
25.02.2026 13:49
π 0
π 0
π¬ 0
π 0
The Resonant Computing Manifesto
Technology should bring out the best in humanity, not the worstβa manifesto for resonant computing built on five principles that reject hyper-scale extraction for human flourishing.
Remember when the internet wasn't awful? We can go back to that.
Some friends and I have released the Resonant Computing Manifesto: a call to bring back such a time, to see if we can bring back a world where technology works for us, rather than against us.
resonantcomputing.org
05.12.2025 17:57
π 930
π 329
π¬ 37
π 43
Exactly. It would also make it easier to find the right tab in browser
30.11.2025 23:30
π 2
π 0
π¬ 1
π 0
Hi Nick, can you please fix the page titles / metadata, so when sharing the link with someone the message would contain the event name instead of "smoke signals"?
30.11.2025 06:44
π 2
π 0
π¬ 1
π 0
Making Sense of Open-Source: Ideals, Reality, and What's on the Horizon Β· Zoom Β· Luma
Have you been wondering about the future of open-source? Do you find yourself caught between idealistic visions of collaborative commons and the messy realityβ¦
There is an ongoing war for the soul of "Open-Source". We lost the battle, but we don't have to lose the war.
On Dec 3rd, @kpaxle.bsky.social and I are co-hosting a discussion with members of @scenius.bsky.social: @ranganaut.bsky.social, @erlend.sh, and Regis Chapman.
RSVP here!
luma.com/z9ivf64k
24.11.2025 19:37
π 7
π 3
π¬ 0
π 0
It is - feel free to join our Telegram or Signal group
13.11.2025 17:13
π 0
π 0
π¬ 0
π 0
Seminar: Governance Among Plants & Insects with Dr. Orit Peleg Β· Zoom Β· Luma
Dr. Orit Peleg seeks to understand the behavior of disordered living systems by merging tools from physics, biology, engineering, and computer science. At theβ¦
Wednesday! Join me for a Metagov Seminar with my colleague Dr. Orit Peleg from CU Boulder, a computer scientist who does the most wonderful work on self-governance among plants and insectsβthis is going to be glorious: luma.com/pzbp4zq1?tk=...
10.11.2025 23:27
π 6
π 1
π¬ 0
π 0
Stuart Lynn from CrownShy talking about some very cool digital democracy projects in Scotland and elsewhere! Nice to see Harmonica mentioned @zhiganov.bsky.social ;)
30.10.2025 13:40
π 1
π 1
π¬ 2
π 0
Yes, and this is why we've created CIBC: to connect builders of citizen infrastructure with local communities all around the π cibc.notion.site
22.10.2025 18:29
π 4
π 1
π¬ 1
π 0
I think it's good practice to use existing platforms as MVP to test your hypothesis before building something new... but of course a Goodreads list might not be a good MVP in this case
if someone is already using Goodreads, the friction to participate in the experiment would be much lower for them
22.10.2025 18:15
π 0
π 0
π¬ 0
π 0
Fun fact: before moving to SF Eugenia was a lifestyle journalist who reviewed chic restaurants and night clubs in Moscow
22.10.2025 16:10
π 1
π 0
π¬ 0
π 0
Individual preferences are meaningful within a relationship but create noise for collective sense-making, without shared context. Why doesn't Tinder allow filtering by Letterboxd/ Goodreads/ Spotify likes? Such an obvious way to increase the value. But would it help us run better citizen assemblies?
22.10.2025 06:05
π 1
π 0
π¬ 1
π 0
That's why some people believe Goodreads to be the best social network ever created
22.10.2025 05:42
π 1
π 0
π¬ 0
π 0
basically, any truly prosocial AI agent should:
- be open source
- support BYOM (e.g. public AI or locally hosted AI)
- be enhanced by prompts & datasets provided by data co-ops
- be owned by employees or DAO members
- adhere to democratic standards (e.g. democracylevels.org)
- what else?
16.10.2025 17:53
π 2
π 0
π¬ 0
π 0
Interoperable Deliberative Tools At Metagov, Summer 2025 update
Walk with us towards deliberative tools that we can govern
This project is part of Metagov's effort to build deliberative tools that can't be captured by corporations or governments. Open source, interoperable, designed so communities can choose their own path rather than being forced into someone else's vision. metagov.substack.com/p/interopera...
12.09.2025 11:51
π 2
π 0
π¬ 0
π 0
Our WG at @metagov.bsky.social is working on "AI facilitation primitives" that could empower communities to have better conversations. We're learning from practitioners and synthesizing patterns that can inform prompts, evals and datasets for developers of tools like Harmonica πͺ
12.09.2025 11:51
π 2
π 0
π¬ 1
π 0
It's weird timing - working on tools for democratic participation while watching actual democracy crumble in real time. But that's exactly why this work matters. When the old systems fail, we need alternatives ready to go. Actual ways for ordinary citizens to govern themselves.
12.09.2025 11:51
π 2
π 0
π¬ 2
π 0
A proposal with policy recommendations for supporting prosocial platforms - governing tech design goes "upstream" as opposed to the "downstream" regulations that govern content. toda.org/assets/files...
14.07.2025 13:34
π 3
π 3
π¬ 0
π 0
As predatory markets tighten their grip and the charade of democracy is abandoned, commons will become even more indispensable. I'm thrilled to say that a significantly revised, updated edition of my book βThink Like a Commonerβ is now out, today! Details at thinklikeacommoner.com.
11.03.2025 15:51
π 24
π 12
π¬ 2
π 1
"Facilitated by professionals from Missions Publiques, the assembly featured an AI prototype to synthesize input manually entered by students and groups at various stages of the deliberations and link summaries back to the original discussion with a search function."
08.03.2025 11:35
π 0
π 0
π¬ 0
π 0
Ooh, nice write-up from Yale! Actually the "AI prototype" is called Harmonica (www.harmonica.chat) β we created a custom workspace for ENS students and built a few RAG pipelines based on MP's facilitators requirements (coming soon to our regular app!)
08.03.2025 11:31
π 1
π 0
π¬ 1
π 0
6/ This vision merges collective intelligence, radical interoperability, and agentic thinking to create a pathway toward global alignment of human and more-than-human needs at every scaleβfrom neighborhoods to cities, and beyond. πππ
09.01.2025 11:19
π 2
π 0
π¬ 0
π 0
5/ Imagine these avatars engaging on X / Bluesky. They could:
- Share insights & perspectives with relevant audiences
- Be monitored & guided by their communities
They would dynamically update based on:
- Collective dialogues & deliberations
- Data from scientists & journalists
- Data from sensors
09.01.2025 11:19
π 0
π 0
π¬ 1
π 0
4/ To be legitimate representatives, these avatars must:
- Operate based on constitutional principles
- Have open-source code and opinion provenance
- Be subject to ongoing audits
- Allow real-time adjustments
Transparency ensures they would truly reflect the collective will.
09.01.2025 11:16
π 0
π 0
π¬ 1
π 0
3/ Double interoperability:
These digital representatives could also learn from collective dialogues in Remesh, GoVocal, Talk to the City, etc. if they agree to use a standard protocol. And the avatars wouldn't be locked into one platform. They could operate across different systems.
09.01.2025 11:15
π 0
π 0
π¬ 1
π 0
2/ For example, residents of a neighborhood use Harmonica to sensemake on local issues. Over time, it will gather enough data to create a community avatar.
This avatar could then participate in deliberative processes at the city level, representing the voice of the neighborhood.
09.01.2025 11:14
π 0
π 0
π¬ 1
π 0