Whilst communication face-to-face is generally to be preferred, this is clearly not viable for large communities. However, collective decision-making within large communities grows ever more important. This page investigates early steps towards achieving IT tools, mediating between individuals, that would help in enabling meaningful and rewarding outcomes in large communities. It also explores how such interactions within separate large communities and subcommunities may be woven together into one structure. Such interactions would enable a radical new crowdsourcing of ideas and a highly significant impact on creativity and innovation in modern society. Swarm intelligence would be harnessed both in the generation of consensus on the site and in maintaining its smooth operation.
Here we examine an experimental collection of themes aimed towards improved community amongst large groups. It is intended to stimulate discussion and thereby to improve the ideas contained here.
The aspiration is to work towards achieving large-group community, through an online tool, where:
• all can express their opinions freely and fully (subject to community guidelines), thereby harnessing the resources of the crowd and providing it with a sense of empowerment, promoting engagement and responsibility;
• unity is achievable between and within communities;
• all information is accessible in one place;
• outcomes are democratic in nature;
• the functioning of the site is primarily people-driven through swarm intelligence and the crowd, in order to keep the connection between the site and the people it would represent direct and tangible; and
• views can be explored by viewpoint or community.
2 Human-to-human Interface
Human-to-human interfaces include speech, body language and written communication (traditional or electronic).
Here we investigate how IT tools could be designed to facilitate and mediate in many-to-many communication and decision-making.
The notion of using an IT-based system as a community platform will strike some as overly clinical, detached and impersonal. However, this would be offset by large-scale interaction evolving in real-time, and by plug-ins for chat-rooms, meet-ups and similar. In addition human-computer interaction (for example through brain-computer interfaces) is set to develop meaningfully and rapidly.
3 The Nature of the IT Core Mediating between Users
3.1 Text Basis of the Mediating IT Core
The fundamental basis of the mediating IT core would be text, with the platform resembling Wikipedia in many ways. It should be editable by all users at any level of detail, such as: a single word, a collection of words, phrase, sentence, paragraph, section or page. Each user would be able to express anything on the platform.
Fundamentally possible edits to these blocks would be:
• substitutions (replacing a block of text with another (or removing it) – somewhat like spinning a single wheel in a many-wheeled slot machine (or adjusting it to show a blank));
• moves (changing the position of a block of text – similar to altering the sequence of the wheels in a many-wheeled slot machine); and
• additions (adding a block of text if it is genuinely original – similar to adding an additional wheel to the slot machine).
These edits may achieve the adaptation of any piece of text (or collection of text pages) into any other (similarly to the so-called librarian’s nightmare in mathematics).
3.2 Structuring the Text within the IT Core
The collection of such edits can be greatly simplified by structuring the text on the site through dependency parsing, through which data is organised according to its grammatical structure. This would greatly simplify the selection of possible changes that could be made and therefore the underlying data structure.
• universal parts-of-speech; and
Larger blocks of text (such as paragraphs and sections) could be tagged according to theme in order to facilitate choices between them, including omission: users should be able to see a list of paragraph ‘titles’ and choose between them, viewing them more fully as necessary.
3.3 Interacting with the Text within the IT Core
Two key forms of interaction are proposed:
• at a deeper level users would be able to make the full range of edits as described above. These edits should be tagged with their purpose in order to attract the votes of other users;
• functioning in a layer above this, users should be able to vote on the existing changes, with the help of the above-mentioned tags:
• The possible changes available to a user at any position within the site may be seen as a ‘cloud’ of issues.
• Various mechanisms could exist to make the most interesting issues easily available to users.
• Through these votes each community would be able to display a ‘most-favoured’ version of any particular page – a unified voice of that community.
3.4 Operation within Different Communities and Text Versions
It is intended that:
• different communities operate within the same framework.
A user should be able to select which community they would like to operate in at any given time. It is anticipated that each user could be a member of separate or overlapping communities, or those that are subsets of another.; and
• within each community different versions of the same text be accessible.
Within each community, users should be able to navigate to the version of the text most similar to that which they would like, by selecting existing edits. During this process they could also make votes and edits as wanted.
Once they reached the version similar to the one they would like to see, they would be able to make edits there.
3.5 Reconciling the Choices from Different Communities and Text Versions
This constitutes the most complex element of the proposed tool. It is tentatively suggested that:
• there be a single ‘master version’ of all the text on the site;
• sub-versions according to community or user-selection would effectively be filtered versions of the master version;
• edits or votes for existing changes would be recorded within all communities containing (‘upstream of’) the community in which the user made the edit. This would include the master version;
• a user making an edit would need to organise the correct placement of equivalent edits ‘upstream’. IT-based- or swarm-intelligence could assist or replace this.
4 Swarm Intelligence
Swarm intelligence includes the decentralised, collective behaviour of individuals systems, such as in flocks of birds, ant colonies and fish schools. This concept may therefore potentially be applied to the collective behaviour of users on the site, both in forming and shaping the content, and in managing and policing it. Swarm intelligence may lead to intelligent behaviour of the group as a whole, unknown to the individuals involved.
5 Potential Additional Elements of the System
5.1 Delegation Mechanisms
In order to function in a meaningful way the site would need delegation mechanisms to facilitate users in being involved in a broad range of issues. This opens various possibilities for how this would be managed and its utility maximised.
5.2 Electoral Systems
Since the site involves voting elements, different electoral systems would be possible. This would certainly be the case for delegation mechanisms. There is the possibility that it might also apply to voting for different edits but this is left here as an open question.
5.3 Maximising Interpersonal Connection
In order to maximise the interpersonal nature of the tool, functionality should exist for forming groups (including friendship groups), online discussion/ and meetups etc.
5.4 Verification of facts
PeopleTree would exist in order to enable collective opinion-making. As part of this, mechanisms should exist to establish the truth of certain bits of information. These should include methods such as referencing published sources and input from established experts in a relevant field.
6 Control of PeopleTree
A community like PeopleTree, established for the greater good, could potentially be self-governing. However, where this were not possible (for example before potential self-governance) then control should be by carefully appointed trustees, potentially established by government or international bodies.
7 Funding of PeopleTree
PeopleTree is envisaged here as being funded by subscriptions.
As a basic outline: these would be collected by the smallest sized communities possible. So for example communities of a geographical nature might start with a subscription to a neighbourhood (a street or group of streets). Neighbourhoods could collectively pay subscriptions to a borough and this pattern would continue up towards nations and the international stage. The same could apply to non-geographical groupings by topic.
Mechanisms should be found to make subscriptions means-tested.
8.1 Security through nested communities
The security of the system is potentially a concern. However, the design outlined above for funding the system would also have security benefits. If each user has to engage with the system first through a small community then identity can be more easily identified at that level.
• individuals within a neighbourhood may be easily identified;
• neighbourhoods in a borough may be easily identified;
• boroughs in a city may be easily identified, etc.
Certain types of community are better suited for establishing identity, such as a neighbourhood or football team. Conversely a community of those with a common interest but who are only connected through the internet would be less suitable, such as those sharing a set of political values.
8.2 System outages
Object of the system would be to forge real community. Even in the even of system outage the ‘imprint’ of the community connections would survive within society.
Issues arise concerning access for all to the platform. There are two key aspects for a response to this:
• currently delegation (through Parliamentary democracy) gives everyone very limited access to democracy (once every four or five years). Those with limited access to IT would be able to delegate their voting to others. This would still significantly increase their capacity to engage;
• in our view the answer is to increase access to IT, not to be held back by lack of it. Creative solutions need to be found in this area.
Verification of individuals should be managed by any community holding ‘certified’ status (i.e. trusted to verify individuals). At the lowest level such communities would most naturally be those in which face-to-face contact is possible, such as neighbourhoods. Neighbourhoods could be verified by boroughs, and up the chain to states and the international community.
Blockchain mechanisms could also be used.
The processes of verification and funding would naturally sit together.
Multiple verifications should be encouraged.
This is intended only as a set of initial ideas about PeopleTree. Input is welcomed in order to develop further the various concepts involved.