---
tags: dexdex
---

# conversation augmenting

"Rapid (and not rapid) Sensemaking Frameworks"

## notes from call Connor <> Christina Bowen June 11

built-in feedback loop
"what is the actual meaning of "care coordination"?
drastically cut down the time to get data into the map
teasing out the nuance is really important for network maps
deep code part 1, part 2, 5 stages of learning
discernment, alignment, coherence, practice?, mastery
coherence check
what would be the way that you want to configure it?

within the UI, I want it to be a component...

anytime you want feedback, or you want to cocreate something...
a configurable module that you want to add to a survey
if we can automate parts of the process
an iterative feedback loop, that sensechecks you before you move to the next step
LCA goes into effeect mid July

Christina talks about "sense-checking", which is a great term for what rsf can provide. Quick social checks, how do people feel about x?
___

How should consent work?

Game-style... you don't know ALL the rules

Players opt-in to a game, trusting the maestro

Gaming... playing a fun conversational game

Milestones and waypoints... when people participate its great to know where a closure point will be.

___

Metamapper goes in here
https://metamapper.herokuapp.com/
https://github.com/metamaps/metamapper

Metamapper rough work, for University of Technology Sydney

https://paper.dropbox.com/folder/show/UTSMetamaps-Bot-Work-e.iX7ZavGxujPFwhjOZcQrb92y2xOKZ1LNCUUu3e4Pz7IuqDd8Iz

This document on "modules" is particularly interesting
https://paper.dropbox.com/doc/Modules--AbjRg1iil6Fy7SRYKok3CPA2Ag-bYFajWtz5E73r0QpjbX3s
It talks about the different types of modules that would facilitate different modes. It classifies what kind of inputs and outputs each process would expect

### ontology

The fundamental data type in the ontology of this whole thing seems to literally just be:
- a block of text
- we could call this a "statement"... its just a String
- additional data types include:
- A LIST, or array, of statements
- A LINK, between statements
- statements could have arbitrary metadata... key-values

- These statements get passed around as values, they flow in and out of functions and forms, being transformed
- This led me back to FedWiki... recalling how brilliant that is.
- a collection, or array, of ordered STATEMENTS... each has an ID, so that it can have a trackable history
- they call this array a "story", just click view JSON
- there are a set of actions performable on a story
- there are a set of actions performable on a "statement" or an element in a story
- statements also have a TYPE property

https://stormz.me/en/features
reminds me of the work on metamapper, and its facilitation template-type stuff

Loomio explains great use cases and best practices for their suite of CI tools:
https://help.loomio.org/en/user_manual/getting_started/decision_tools/

### Video proof of concept
<iframe width="560" height="315" src="https://www.youtube.com/embed/wqUliWi2wwQ" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

## Metamapper talks Holochain?

what would it take to bring Metamapper into a Holochain app?

https://github.com/holochain/holochain-basic-chat already implements basic chat channel functionality. It would only need to be extended to have Private Chat DMs functionality... and to begin with, that could just be simulated. They wouldn't have to be private private, they could just be hidden.

The UI ideally needs to be capable of rendering a Markdown like formatting for messages.
The syntax would NOT match up with slack. But what if we left slack behind and focused on a custom UI? Slack is the PAST

## Nesta Collective Intelligence Proposal Reworked

We propose to develop the Rapid Sensemaking Framework (RSF), an open source, cross platform collective intelligence bot/service available for popular chat services in use by distributed teams (FB Messenger, Text, Email, Slack, Mattermost). RSF enables a facilitator to request different types of knowledge from a set of stakeholders within and across digital chat channels.

RSF enables a facilitator to query up to hundreds of people with specific types of knowledge requests, polls, and tagging.

Linear back and forth chat is a notoriously shabby way to make sense of complex systems for groups. We propose to test the use of structuring techniques that come from workshop facilitation and encode those in a chatbot, as a way to scale conversations about complex issues to networks of hundreds. This chatbot acts as an extension/prosthetic of the facilitator, to hundreds of individual chat participants - but it also captures and organizes the knowledge immediately, in a way where the conversations build in context.

As so many communities of myriad types use chat platforms, we want to see if we can enable communities to develop new collaborative sensemaking skills, using the combination of automation via bot + human designed facilitation methods/models embedded within.

An outcome of RSF as a facilitator's tool augments the decision-making capability of the group as a whole to more rapidly understand themselves and their own preferences.

## Planning

We would like to develop RSF so that we can do an extended pilot with various communities for controlled tests with groups of various sizes and types.

The technology is currently working, but needs more dedicated testing and development time. We have lessons learned, from several test runs.

Our plan is to finish building the bot and integrations, and test it with partner communities. Our partner communities should include at least one of several types, such as organizations, distributed teams, and informal communities, et al. We also plan to do tests amongst a completely distributed team virtually, as well as in-person facilitated sessions. These teams will range in size from under 10 to ~150.

All work will be licensed AGPL, with communities being able to install it and host it themselves.

1. The tool will be able to be adopted and used by users of main chat apps
2. The knowledge generated by the tool may be useful to the decision-making of the group
3. The data and experience generated by our tests will be of use to other practictioners and researchers as frameworks and processes for running collective intelligence processes within distributed communities.

RSF explores how we may scale social cognition through use of chatbot tools to make better decisions together.

For a decision-maker within our partner communities, we hope to provide actionable insights for groups in how they can use the ubiquity of group chat platforms as a way to unlock the latent collective intelligence of the group. We also plan to have some artifacts of knowledge that will be publishable.

## Nesta Collective Intelligence Proposal Original
https://www.notion.so/b2f2ade4f4f84006ae391ae0d39db687
Prepared by Ishan Shapiro

### **Proposal Name**

Metamapper Collective Intelligence Bot

#### Proposal summary (please provide a summary of your proposal, no longer than 150 words) ***This question is required.**

We propose to develop Metamapper 1.0, an open source, cross platform collective intelligence bot available for popular chat services in use by distributed teams (Slack, Mattermost). Metamapper enables a facilitator to request different types of knowledge from a set of stakeholders within a digital chat channel, and have that knowledge be organized in a dynamic knowledge graph. It also enables automated knowledge capture from group chat channels.

The bot integrates with [Metamaps.cc](http://metamaps.cc/), an open-source visual knowledge graph for collaborative sensemaking. Metamapper + [Metamaps.cc](http://metamaps.cc/) enables a facilitator to query up to hundreds of people with specific types of knowledge requests, polls, and tagging, while ingesting their responses into a curated knowledge graph in real-time.

Metamapper is already a prototype in testing. The code is on Github here: [https://github.com/metamaps/metamapper](https://github.com/metamaps/metamapper). You can explore more at [http://metamaps.cc](http://metamaps.cc/).

#### Please explain what you are wanting to test/learn through your experiment(s)? And what outcome measure(s) do you envisage using to answer this? (max 150 words) ***This question is required.**

Linear text is a notoriously shabby way to make sense of complex systems for groups. We propose to test the use of structuring techniques that come from workshop facilitation and encode those in a chatbot, as a way to scale conversations about complex issues to networks of hundreds. This chatbot is almost able to be an extension/prosthetic of the facilitator, to hundreds of individual chat participants - but it also captures and organizes the knowledge immediately, in a way where the conversations build in context.

As so many communities of myriad types use chat platforms, we want to see if we can enable communities to develop new collaborative sensemaking skills, using the combination of automation via bot + human designed facilitation methods/models embedded within. An outcome of Metamapper as a facilitator's tool augments the decision-making capability of the group as a whole to more rapidly understand themselves and their own preferences.

#### How do you plan to carry out your experiment? (max 250 words) ***This question is required.**

We would like to develop Metamapper so that we can do an extended pilot with various communities for controlled tests with groups of various sizes and types.

The technology is currently working, but needs more dedicated testing and development time. We have lessons learned, from several test runs. We have several communities who have expressed interest in using Metamapper when it is more production-ready. We've got developers, designers and facilitators who are wanting to be able to devote time to contributing to it. In short, our plan is to finish building the bot and integrations, and test it with partner communities. Our partner communities should include at least one of several types, such as organizations, distributed teams, and informal communities, et al. We also plan to do tests amongst a completely distributed team virtually, as well as in-person facilitated sessions. These teams will range in size from under 10 to ~150.

We plan on using the grant to dedicate for stipends to key designers and developers to contribute open-source code to the base, a dedicated community liason, as well as other small hosting and API costs. All resulting work will be licensed AGPL, with communities being able to install it and host it themselves when the experiment is complete.

#### Please outline how this will help create actionable insights for practitioners and have wider applicability? (max 150 words) ***This question is required.**

1. The tool will be able to be adopted and used by users of main chat apps
2. The knowledge generated by the tool may be useful to the decision-making of the group
3. The data and experience generated by our tests will be of use to other practictioners and researchers as frameworks and processes for running collective intelligence processes within distributed communities.

Metamapper explores how we may scale social cognition through use of chatbot tools and network visualization + analysis to make better decisions together. Metamapper is a unique bridge between chat platforms and a graph network mapping platform.

For a decision-maker within our partner communities, we hope to provide actionable insights for groups in how they can use the ubiquity of group chat platforms as a way to unlock the latent collective intelligence of the group. We also plan to have some artifacts of knowledge that will be publishable.

#### Please describe if your experiment is part of a larger practical project or research programme (max 150 words)

[Metamaps.cc](http://metamaps.cc/) is a collective intelligence experiment in action, proudly in beta since fall 2012. It is the only open source, collaborative, real-time graph mapping platform out there (seriously!). [Metamaps.cc](http://metamaps.cc/) has never been 'startup-ized'. It has been developed and used by a small, dedicated and growing community as a platform to collaboratively create and edit concept, or knowledge graphs, together, online. It is an AGPL licensed platform with code found on Github. It supports a small, dedicated community of systems thinkers, facilitators within various types of organizations and communities in support of collaborative sensemaking. We have done pilot programs with the University of Technology Sydney, Waterloo U, University of Southern California and The Value Web, amongst others. We are a p2p community dedicated to contributing to the knowledge commons and building new interfaces to explore collective intelligence solutions for communities worldwide.