Tagged: ha-ha

centerless politic and old people!

I think I’ve worked out the social form that creates a dynamic, centerless politic! It’s come out of the experience of the ha-ha’s, and involves compression and expansion of the phase space of £-people-time. Circulation of money in equity cycles, what I originally understood as horizontal and vertical but is now sequential in time, makes money ‘well-behaved’ while giving rise to a subjective enumeration network which tracks our own individual personal value. Lovely.

StandardTori_701

A centreless politic is something I have had my eye out for over a decade. How to create a social dynamic that is centerless, and yet stable? Our current politics and economics suffers from emphasis on the centre, on the pinnacle of the pyramid, on ego, on selfishness and ownership. We area culture dominated by ‘leaders’, and worse, institutional forms that places control centrally regardless of the capacity of the people who occupy central positions. A centerless politics provides us with a completely different social dynamic.

Because of the initial conditions of giving money before deciding where the money is going to be directed, the ha-ha’s create a centerless structure. However, it was only half of the dynamic of the torus — the compressive aspect. That is, when a bunch of people agree to give-it-forward to a smaller number of individuals, represented as the movement from outside the torus to the inside. The other half was not obvious, and is the reason why the ha-ha’s failed — the expansive aspect. What is the social force equivalent to more people being involved? Yup, the Invitational-Protocol. During the co-design period in August, I put it to everyone for the second meeting that we needed to come up with the IP that would work given the limited technology. We don’t have a means of tracking our Invitations, so we settled on using an off-the-shelf ‘crowd-funding’ template.

Anyway, if we sort out the IP, being careful how money-people-time expands, so we have the two parts to making this centerless politic exist!

how many people will we need to make it work?

Because we are dealing with fluids, it is not a straight forward ‘number of bodies’. Just like one’s pulse can be measured not just by number of heart-beats per minute or the blood pressure, the Chinese have several more subtle indicators.

We need to consider density — the number of people within a period of time. We might have several ha-ha’s running simultaneously, each with their own settings. Most of the feedback during the ha-ha’s involved suggestions to improve the tech, making it easier for people to be represented, dreaming up new bespoke tools — check these rather interesting social toys, by the way, specifically Points of Unity.

We wanted to get 100 people in the last ha-ha, to create the conditions that is beyond a small group. We simply don’t have any direct experiential feedback at this level, apart from when we work in companies and governments, or when we are audiences, or take part in self-organising systems like driving on the roads. We didn’t create the conditions, we didn’t meet the right density.

We may also want to consider density in terms of the Invitational-Protocol, how many people we invite, over what period, and with what quality, and range of gleick-group we are tapping into — it is wiser to invite only a few from each social group we may have, rather than all from a social group of like-minded people, which is our traditional way of organising ourselves.

and where to the old folks come in?

Because it works scalably, I have no idea of the threshold numbers from this end, from initial start-up situations; less than a hundred, perhaps only 40, but over a few ha-ha’s. From the back-end of global unity, using the old back-casting thought-experiment we should be familiar with by now, I know the peer group that can make it happen — old people! When we are talking about change over a couple of decades, over a generation, there is only one generation that can almost guarantee that they will not be about, thus deriving our centerless politic. Ie, if it is started with old people, rather than young and wilful, they may be able to kick-start the conditions that in a decade bring fruition for us all globally. It requires this level of vision, something we tend not to think old people have. However, with their age, this kind of period is well within their grasp to understand and to operate by. Think about old folks and their grandchildren, and the legacy they are leaving them globally.

I tried this idea out with a housemate, and the idea got immediately dumped. He only saw the problems that may arise from children worries about their parents being fleeced. Yes, there are problems, and problems which we may not be able to overcome. Still, global unity isn’t going to come about without challenges. This is a challenge that they may need to take on. It is not up to us.

So, I believe this is a rather beautiful invitation for old people to regain some of their lost respect, and bring into existence a sustainable world. I doubt if anyone will get this in this day and age, and also because very few (less than a handful) have any insight of the potential of the protocols, but my, it is beautiful. And the invitation goes to those who are terminally ill, or those who are suicidal — a possible way for them to contribute towards creating a centerless politic. And those mature souls who are playing a less-ego-bounded game; for though they may live to see it happen, they are centrelessness enough in their daily practice, that they have some chance of avoiding the power that comes with social attention.

imgres

inaugural happening-hangout

$10 for a piece of future-history!

Inaugural Happening-Hangout tomorrow http://isithappening.org/campaigns/inaugural-ha-ha/ uses the give-it-forwards mechanism integral to ecosquared financial protocols.

photo

Learn experientially — shift from discussion and community building to direct social impact.

There’s a free option, but it is recommended to put in $10 to get a full experience. Although only a handful of people may turn up (this is only one day’s notice), the social insights the first players get will be essential to the success of the experiment.

building a house ecosquared style

How would I go about building a house if I wanted to do so using ecosquared protocols? I will look at three conditions of social saturation:

  • Red just starting with only you using the protocols
  • Green with teams of people connected in a network using the protocols
  • and Blue where a wide sector of people are sharing within ecosquared protocols

There are many factors to building a house and different ways to categorise its construction, let the following payments are estimates for building a three-bedroom house using traditional economics suffice:

  • land cost, buying land, £100k
  • labour, 40k
  • materials, £120k
  • selling of house £260k +/- whatever the ‘market’ price of such a thing is in the current market

Screen Shot 2013-08-23 at 14.34.29

red solo

Because you are the only person using the protocols, their use is rather limited. Land will be bought, as normal, though you may want to consider the land to be ‘commons’. The materials will also be bought. However, it may be possible to phrase the contracts with suppliers as an MTTP contract. Certain resources may be accessed again in the future, and rather than thinking the money is for the materials, it is more for the relationship to the materials supplier. They just happen to be giving you the bricks, cement, cables, etc.

Labour is conducted as MTTP, whether people understand it or not, it does not matter. It may be possible to open up the potential for a ‘crowd-funded’ house. That is, when the house if finished, the distribution of it worth (if it was sold) would be shared by the value attributed to people’s contribution.

When the house is built, it is ‘open-source’. It is nominally, legally owned by you, but in your mind it is simply the place where you happen to live. Neither the house or the land on which it is built is owned.

Building a house during Red Conditions, means this is more a mental exercise, a thought-experiment, than anything else. It does open up the necessary mentality for future growth however, both in terms of the people engaged to build it, as well as engaging others who have self-built.

Screen Shot 2013-08-23 at 14.36.39

green team

A team gathering to build a house is a little like a ‘barn-building’ community, or a ‘building society’. People are aligning to create a house together, and if they can build one, they may help others as a team so that everyone who contributes ends up living in their own home.

The land is ‘owned’ but is shared in a collective sense. Again, not much meaning here unless the team actually build on mutual, shared ground.

The labour can be paid for via MTTP, and if within the team, it is volunteered. Subjective enumeration is the basic means of tracking people’s contributions. This should map in some way to the current market costs, for laying foundation, brickwork, panelling, flooring, electrical system etc. It should be possible to estimate the ‘return’ to each contributor via SEA should the house be sold.

Resources are bought off market. Again, this is a standard ‘trade’ exchange, money for resources, however it may also be clothed as an MTTP contract, to invite suppliers into the ecosquared supply chain.

The finished house is owned collectively, but is occupied by whoever is living there. The current inhabitants are stewards of the building. This should enable greater flexibility in terms of movement should people want to relocate to other parts of the city, country or the world, by swapping households.

The ‘market-value’ of the house is in terms of traditional economics. In terms of ecosquared, the building has no financial value whatsoever. We may attribute subjective enumeration, presumably by the inhabitants, but this relies on allowing SEA to apply to things. The principle, the practice, is that once the house is built, it is ‘free’. It is a resource to be tracked, to be used, but not to be tabulated with a monetary value.

old fashion barn raising

blue network

Let’s assume that the ecosquared network has end-to-end chains of supply. This means that the build team are simply sharing their resources, as do the suppliers. There is no payment for anything. Materials are given. There may be MTTP contracts, and of course subjective enumeration, depending on the scale of the network.

If this is integrated with an end-to-end supply chain of food too, so that material suppliers and build teams are given food, then the system gets close to becoming fully self-sufficient. Of course, this is still within the current market place, with externalities like oil perhaps, or electricity from a central power grid, and so on, which need to be ‘paid for’. This is balanced by ‘income’ by the entire network as normal jobs. Some participants may take an income as ‘joiners’, ‘bricklayers’, ‘construction workers’, ‘farmers’, ‘teachers’ and so on. Such individuals sell to the current market, and this money is distributed via SEA to all participants within the network.

Because most of the daily resources are internally sourced, whether it happens to be the bricks for the newest buildings or the lettuce in the salad, there is no need for money, personally. Luxuries may be bought by individuals, but this kind of behaviour is frowned upon since the community is still paying for certain collectively required resources, such as electricity, petrol, etc. Once these collective requirements are covered, which is a calculation based on resource allocation and nothing more, the moneyflow which ends up with individuals is spent as each individual sees fit. Because of the way the system works, those who accrue the most money are those who produce the greatest value within the community. This may be related to jobs that people do not want to do, such as working in the sewars, or who happen to provide more invisible skills and services, self-development, social integration, educating kids, etc.

The point is, with the accumulation of material products, as houses, schools, farms, factories, enter into the ‘commons’, so there is no need to pay for these. These resources are shared. Their use is what is tracked using SEA as well as resource allocation economics. There are no owners, so no burgeoning externalities. People are being useful to one another, giving of their produce, whether it is planting a seed and then giving the resulting carrots to others, tutoring kids at math, or constructing bricks. Everything is given freely, and tracked with SEA. Money loses its internal necessity at the rate of which things are being produced within the ecosquared entity; money is no longer needed to ‘buy in’ resources.

 

challenging observation

You decide when these periods of time occur, when we shift from Red to Green to Blue conditions. The rate of social adoption depends on each one of us, individually and collectively, personally and globally.

At some stage, entire communities can become self-sufficient, where money is no longer needed to ‘buy-in’ resources, since they are all generated within ecosquared, food, materials, energy sources. This constitutes the Yellow Condition. Once these conditions apply to all communities globally, then we will have shifted from our current economic system to an economics based on giving, and GIFT will have been completed.

Could something like the Ha-ha’s be used to build a house? Current estimates are about 6 months to build a house. Would 1000 participants be willing to put in £100 to get a house built, using ecological designs, minimal electricity usage, etc — and that house be ‘owned’ by the collective? The current system for enabling ha-ha’s is not strong enough to support this kind of initiative because we can’t track SEA. However, one day the system will be more rigorous, and if participants put £100 per month into these initiatives rather than a mortgage, the world may look rather different in a decade or so.

second critical week of co-design of ha-ha’s

We had our second meeting on Monday, framing what needs to be done this week, namely working out the ‘Invitational-Protocol’, how we invite people to the ha-ha’s and what the tech required to enable a smooth player experience.

I’ve timestamped the first hour so far. There are a few framing bits to start with, framing piece, and then two conversational bits of about 10-15 mins between Doug, Mani and me which are very interesting.
I get the impression when reviewing this, the complexity of the space we are creating within. Amazing to hear Doug’s and Mani’s perspectives. Very revealing, of themselves and their trajectories, but more importantly the space. Doug has a remarkable understanding of the current corporate ecology, Mani is focussing on designing the player experience, and I am attempting to introduce ecosquared financial protocols. If I wasn’t locked into the objective of 1000 people by end of september, this is the kind of conversation we ‘should’ have had in phase 1. I believe our objective would have been different.

benefits of light-touch co-design of a specific instance

I still think we can pursue multiple strategies, which is the benefit of working as lightly as we are. By creating a prototype, by fulfilling our commitment:
    • we have something to show eg reddit or quora which is along Doug’s lines
    • we’ll be able to tighten up an experiential tube to fulfil Mani’s method
    • we have experienced a way of working which is light, and yet fulfils our objective

That is, we have something concrete to take forwards, as well as the experience of trying to integrate our thinking in a specific instance. A second iteration may emerge as a consequential result, either directly through the ha-ha’s if they turn out to be successful, or as a separate trajectory if not.

this week’s co-design focus

So, let’s apply ourselves to the co-design for this week: creating the ‘Invitational-Protocol’. The default is running much like a crowd-funding site. Can we include a loop that binds people through trust? Can we include a financial movement when we invite friends? If we manage it, this increases the chance that the ha-ha’s will work. By binding value, trust, and moneyflow at the dyadic relationship, in the ‘Invitational-Protocol’, we create an ecological economics equivalent to ‘marketing’ and ‘sales’. Check out the MTTP animation, since the IP is a version of MTTP.

Remember, though, it is not sales or marketing, since what we are inviting people to is ‘not-known’. It is a genuine invitation to co-create. Amazing potential — let’s see if we can apply ourselves this week and crack it!

 

happening-hangouts, a player’s perspective

Having engaged a few people regarding these happening-hangouts, I am encouraged to think about how the actual hangout will occur. However, I do not want to fix on the details since I believe this should form from the process of co-creation once we enter into phase 2. At phase 1, we are simple scouting, conducting a feasibility check, checking out the lay of the land, specing the opportunity space. Thus, what follows are ‘visioning’ experiments, imaginary projections of how the experience of ha-ha may feel to prospective players.

encounter film channel

the_encounter

It’s 2020 and there are ha-ha’s on TV, a kind of eco-political reality-TV show. The most powerful is hosted by Jason Silva, where 10,000,000 participants have a couple of hours to decide what to do with $100 million, every week.

Purdesh doesn’t go for that scale of game-show. She thinks it is fake, and what are the chances that your idea will ever end up getting the votes? Nope, she is happy with playing on the net, through various small-scale specialist piped through Youtube channels. This is the real home of ha-ha’s, not TV. She’s particularly interested in film, and she’s party to a longer term sequence of ha-ha’s in the production of a full-length indy feature film. Each week, everything is being decided, from the script up. The direction of the film can be altered from week to week. Like the script of Lost, she’s been told, but that’s ancient history. Because they are rolling it out over a two months, and they are in week 3, there’s a lot to determine. There are other channels which provide more flexibility in terms of changing directors, actors and virtually every parameter of the film production, but the quality suffers in terms of production, actors and so on. This film, working title ‘Encounter’, has been promised TV airtime; one of the facilitators is a si-fi Channel Commissioner, another is a professional producer, and they are joined by the director or writer or main actor.

Purdesh’s contributions have been noted: not only have her votes ended up in the ‘winning’ results, but one of her ideas regarding a character actually won in week 2. She is effectively responsible for a character, a girl called Reyka, and she has been invited to co-write the script as an ‘advisor’. If that character remains, which means further developing the plot, she will remain engaged. She’s written herself into the script.

There are forty-thousand people playing this ‘Encounter’ ha-ha, distributed through three threads, writers, production, and sets (both real and virtual), which is $400,000 a week. Because she is contributing to the script thread, Purdesh has the opportunity to take out some money, but she thinks the experience is worth more, connecting with ‘professionals’, getting her script-writing skills an airing. It’s a start. There are plenty of people who are contributing without taking any money out, so that most of the money goes into getting the best graphics and sets in the sets thread. After all, success for the production depends on it looking good and not just to the 40,000 regular participants. It’s got to have professional standards for a TV airing.

awesomest parties

Screen Shot 2013-07-31 at 23.49.39

It’s 2015 and ha-ha’s are as ubiquitous as kick-starter. There are several sites, but the most popular is happening-hangout.com.

Dirk is plays two ha-ha’s a week. A regular tech-orientated ha-ha hosted by Hermione Way from Techfluff.TV. There’s around 120,000 regulars to the show, and they decide what new prototypes to take to production. Dirk really liked the start-up that came second the previous week, a watch-sized phone, and he’s funded it on the slower kick-starter track. Techfluff ha-ha is one of the most popular fast-track channels. Dirk plugs in to the hangout, votes start-ups he likes on a complementary site, a special instance of the stack-overflow system, and watches Hermione Way and her guests talk about the options. There are usually a few prototypes that rise up initially and these end up escaping from the rest, and the comperes draw Dirk’s attention to those which are lost in the shallows of the long-tail distribution. Several times Dirk has switched throughout the hour when a new item appears.

Dirk particularly enjoys watching how a contender shifts from pure voting to when people commit to making it happen. Teams form as players offer their expertise. Because it is all tracked, reputation is everything. Gone are the days when players would say they would code this bit of the app another design the interface, and then nothing would happen. Big let down. With reputation systems tracking player’s contributions, such gaffs weren’t allowed to happen. Dirk himself had contributed to a few start-ups, producing some short music patches on time, and his reputation stats were building gradually. This made him cautious about offering his skills to the more obvious early winners, contributing his skills to those lower down hoping that they get a final rush. Dirk is currently contributing to several start-ups which didn’t make it to the cut but got enough airing to attract his and other’s attention.

The beautiful thing about these ha-ha’s was that you just didn’t know what was going to arise that night. Techfluff-TV had a policy that once a prototype made the cut, the top three four or five main contenders, they weren’t allowed to be entered again. Others ha-ha’s didn’t have this imposition, and though you had the chance to vote and enable your favourite start-up week-in week-out, it got a little political. Which is why Dirk liked Techfluff. It was light enough and fun enough.

mf_ddp_large

The other ha-ha Dirk participated in was a local one he was heading up himself. He hosted it most weeks and brought in guest facilitators. They had a regular viewing of around 200 players. So they were making weekly decisions for around $2000. It was a local network, mostly his school friends, but there was a mix with friends, students, parents even. They had upgraded a local community centre, put new tablets at the local primary school, but most of the winners involved organising some awesome parties. It was becoming so regular, it was starting to get boring, which is why Dirk was inviting this guy Girvin from the coast on the show this evening. Girvin was a player in a local group in his home town, and he had been funded to come to the big city and drum up interested in a big party, more like a festival.

Dirk didn’t really care where the money went. A lot of the suggestions didn’t involve much money, and he’d roll it over to the following week. He was never very interested in where his $10 went, even though it was all transparent. As a host of his own ha-ha, he realised what a hassle it was to redirect the money to those who needed it to fulfil the chosen activity. It was a head-ache deciding who should get what when the activity was chosen. He’d like to switch to the other mode of money-flow which didn’t centralise with him, where players simply directed their entrance fee directly to those who made a claim for it. But his friends were lazy bastards and they didn’t want the hassle. Looked like other ha-ha’s managed it, and most of them were older. They could deal with the higher level of trust and the responsibility to ensure the money got to the right people. It was just a higher level of granularity of respect, something which Dirk wasn’t interested in himself.

purplebeach happenings

2013, and ha-ha are being trialled for a few weeks. If they don’t hit 1,000 regular players, whether in one ha-ha or across multiple ha-ha on the happening-hangout.com site, the experiment would be over.

Screen Shot 2013-07-31 at 23.25.33

Annamie was asked to host a ha-ha with her network. She put out the invitation to her friends and colleagues. The challenge was simple: $10 and one hour to make a decision about what they could do collectively with $1000. A friend was interested in the process and agreed to guarantee the first show of $1000; regardless of how many people turned up, the amount to be decided about was capped at $1k. No pressure for Annemie. If people didn’t pay their ticket, it didn’t matter, so she felt relieved that she wasn’t ‘selling’ anything to her network.

The people who wanted to take part, booked a ticket, and they were informed when the happening-hangout was going to take place, and their requirement to have a quora.com membership. Most players turned up 5 to 10 minutes early, though the hangout was open 30 mins before it began. Various technical hitches were overcome before Annamie, Stephan and David were engaging on the hangout, and it was broadcast live on air. They would splice the youtube recording with views of the quorum site for viewers who wanted to watch the show afterwards.

During the show itself, the three facilitators were kept on their toes by the various suggestions offered by the players. At first the contributions were rather slow, but Annamie suggested some alternatives and Stephan himself wrote up his idea which got a lot of traction. Half-way through a suggestion triggered off a flurry of activity and the facilitators had a hard time simply reading through the offerings. It was like a chain-reaction, and though it flustered Annamie, David was laughing. It didn’t matter what they thought, since the system would take care of itself.

They had three main contenders towards the end of the hour, and the tables were turned as David pleaded with everyone to vote only for one of them. The true power of the system was that they reached consensus. They could always try another activity the following week. What was important was that they fully realised their power as a collective, aligning to one target. This wasn’t politics. He didn’t want anybody to feel that they were ‘losers’ because they had just been pipped by a few votes. It required self-discipline. The vote swung suddenly in the last minute when Stephan made a startlingly simple observation about one of the activities, how one naturally led to the other.

They money from ticket-sales, $860 was put forward to the next ha-ha which was to be hosted by another facilitator and populated by another network. There was an open invitation to the Purplebeach participants; how many of them would return for the following week’s ha-ha? A decision was made to stick to the ‘gift-it-forwards’ system initiated by Annamie’s friend: ticket-sales paid for the following event, not this one. Thus, it was up to the participants to invite new players in order to increase the amount they would play for in the following week, thus taking the edge off the necessity to invite people for this coming ha-ha. It felt healthier.

provisional mock-up site

Screen Shot 2013-08-01 at 16.13.42

This is what the mock-up looks like. The real one enables participants to host their own ha-ha’s, as demonstrated by the above ‘visioning’ examples. We have a open-hangouts planned for each Monday through August, starting on Monday the 5th at 7pm BST.

experiential iterations and social niches

The word ‘iteration’ has been used to describe the different attempts to put the financial protocols into practice, the fourth of which is the happening-hangout. Let’s reformulate this language.

iterative experiences relative to self

I, David Pinto, have been using the term iteration in relation to my own personal experience. After the first attempt, the birthing of the economic entity in 2012, the team wanted to continue with regular meetings at the Hub. I did not see much future in these, but attended a few. I am not into community building just for the sake of community building. I see this as one of the errors of how we self-organise; before you know it, we are supporting the continuance of social entities that have long-ago surpassed their use-by-date. Most of our social institutions, and indeed cities, suffer from this.

From birthing, I went ahead alone by approaching advertising agencies, and when personal issues intervened later in the year, I delimited my efforts to writing a book, which I have termed the third iteration. As you can tell, this is a personally relative sequence. However, what becomes clear is that different scales of social niche emerge, and it is these that may constitute our collective self-organised strategy.

social niches

The following modes of social experience, or social niches, may help map out the self-organised strategy involved:

  • information: this website, blog, explanatory videos and prezi’s, the book
  • synchronous pairing: tango, textango, MTTP, gems
  • dynamic network alignment: action-cycle, happening-hangouts, SEA

There is a tendency for all initial business engagements to collapse back into an information-exchange. As we know, MTTP departs from the notion of exchange at source and is primarily based on setting up a scalable, synchronous, social experience: money-flow and co-creation first, then evaluation, information etc based on shared experience. MTTP is a financial protocol which embodies the new methodology.

1-s2.0-S000510980900435X-gr10

The new methodology is based on social risk, on play, on not-knowing, on trust. By aligning to a future goal, even if it is as short as an hour or a week, we have something concrete to talk about: to iterate experientially, socially. Otherwise, we fall back into the attractor of what we know, what we can reason, what we think is possible, and so our current system of social paralysis continues.

alone, paired, and collectively

We will go into some detail regarding these three in subsequent posts, suffice to delineate the various scaled ‘projections’ in our ecological economy, as they map to the ‘social niches’ described above, from individual, paired, to small teams:

  • writing fictional narratives (to embody ecosquared financial protocols and their underlying methodology based on giving)
  • engaging the front-line of business (gifting companies, advertising or otherwise, with ‘gems’)
  • opportunistic and receptive meetings (the ‘water-cooler’ effect expanded into all social spheres, to realise social serendipity)
  • birthing the ecosquared entity (attracting a seed fund of around £10k per person in order to play at the ‘organisational’ level)
  • hosting happening-hangouts (monetising social media by actuating the latent potential in our networked relationships)
  • conducting action-cycles (to increase permeability across businesses, government, educational boundaries)

 

explanation-exploration for the proposed happening-hangout

Our current system favours the talker, the entertainer, the wilful person, the persuader, the extrovert and so on. Although we have come a long way, I am fairly convinced the amount of money that flows to those we really value is not high enough.

I have not heard anyone say anything that is outwith the current economic paradigm that isn’t literally about giving things away for free, eg freecycle, various share movements, and all sparked (at least in the last few decades) by programmers sharing code. But the current system can go on quite happily, money laundering its way through the degradation of the natural world, while a bunch of people “buck” the system. It tolerates sharing within the family, for example; it tolerates a certain amount of political and even economic dissension.

An alternative economics, one based on giving — and yet still using money we use nowadays — will produce a different social effect, which may enabling us to be generous, for example, and to value that which we hold dear to us the most: our loved ones, the environment; listeners, contemplators, quiet, loving, careful, open-minded, equanimous people.

designing the happening-hangout

If such an alternative economy exists, the way to prove it, is to test it. That’s the point of the ‘happening hangout’. So, if we set up the happening hangout to reflect us well, and the moneyflow ends up with those we individually and uniquely value; that is, the facilitators get ‘paid’, and more people want to take part. We need to get a result in one hour. I am going to give the experiment a test run of four happening-hangouts, with the end-goal that we have 1,000 people who are wanting to ‘play’ for £10-hour. That means, we are making a collective decision about £10,000 on a weekly basis.

Is this possible? And if so, how?

The question that may come to my mind is, which way does the money go? Does this sound like an odd question to you? It may be, but can we answer it anyway: does the money go outward or inward? Does it go out to the people who are joining it last, or inwards to the people who host, who have done this before?

Traditionally, money goes to the centre, it centralises to kings, governments, banks, companies. Does this suggest that in our alternative economy the money actually goes outwards to the periphery, branching to more and more people, that it becomes diluted…? If this moneyflow was to ever coallesce again, then it would be done by peripheral individuals redirecting the money to those people who they thought deserved it, people they valued. This, actually, is a variation of the Invitational-Protocol.

I am not sure at what stage we are at in the world, whether we can initiate such an IP with the first happening-hangouts. It may be more sensible to think of it as a trust game where participants are inviting the facilitators to play the game, like inviting a referee to steward a football game. Once an activity is decided on, the moneyflow goes to where it has been decided. The fact that it does indicates the trust we have in the facilitators. It is important that all MTTP flows onwards to other people, and the facilitators never take any out for themselves. Their payment must rely on the consequence of the social result of the event.

parameters of experimentation for ha-ha’s

Imagining a state where happening-hangouts are actively springing up with thousands of players, what will distinguish ha-ha’s will be the different flavours of experimentation, whether they are geo-local, what protocols the facilitators use in their engagement, the overal timing of the hangout, and the external apps used like quora, the direction of money-flow, the ‘objective’ of the ha-ha, and so on.

When inviting a bunch of people to a meal, we want to cater for different tastes. My brother had a rule-of-thumb, relying on a few stable dishes he knew how to cook, and add an experimental dish. A level of experimentation is encouraged for every hangout.

If there are enough of us playing these games, we will hone in on few distinct systems that enable different results in the real world. The important thing is that the feedback loop to determine what works is determined not by thinking, but by social impact. In this way, we do no try to ‘isolate parts’, specific people, specific protocols, specific numbers of people. It is very much a ‘wholeness’, and what solutions we evolve are systemic, not at the level of any individual part.

But this is jumping the gun. We definitely need to produce the first set of ha-ha’s that simply work — as good as, if not better than ‘working’ in a company structure.

phase development of ha^2 as interlocking s-curves

phases as interlocking s-curves

Phase zero is coming up with the idea.

Phase one is sharing it, and seeing what people say, who is interested, feedback, gentle awareness. Conversational. This is the stage we are at now. This attracts potential designers, facilitators and contributors, and is on-going in subsequent phases since the entire process is open.

Phase two is inviting the self-selecting team to design the basic DNA of the first ha^2. Who are into open design? Who independently see value in their vision of what the ha-ha could be? What is the basic DNA of the first ha^2? It’s dealing with the mechanics, the software, as well as provisional design of hangout engagement between facilitators, and so on. This will continue as an iterative cycle of evolution in subsequent phases, but we need to have a provisional first game fit for play.

Phase three is the actual embodied experiment. This will require the designers and the facilitators to commit to one or all of the first run of happening-hangouts. Provisionally, it will be a run of four ha-ha’s.  We will need some people who are connected enough within a network that by their invitation, we get plenty of people to trial the system. We can not enter into this phase unless we have all the necessary players.

Phase four, what to do if it works? In order for us to enable this, the first website design must have a provisional plan for growth. That is, for example, the ability for participants to start hosting their own ha-ha’s and allow their networks to crowd-fund them. This may even involve ‘companies’ at some stage using this platform.

time-scale?

The first phase arose over a few weeks, engagement with people like John Kellden, Gregory Esau, Joris Claeys, Bert-Ola Bergstrand. The second was sparked inadvertently a week ago with Michael Maranda which has led to an open, though not particularly well-crafted, invitation. Some incredible engagements with people, like Alex Gagnon, Alexandre Enkerli, Willi Schroll, Anna Blume, Joe BreskinGrizwald Grim, Susan Cox, and Michael Layne Hartsell. As usual, this is piece-meal. Those who happen upon it, who recommend, who are interested. I will give this period a couple of more weeks. I will be calling for phase two mid-July, and the actual trial to start in August. However, this may all slide since it is summer, so that the experiment is conducted in September. We’ll see, depends on whether there are enough open-designers (which reminds me of meta-designer John Wood) who catch the vibe.

fourth iteration — a viable hangout

Coming to the end of the book, and I am fairly certain it is not going to be a big page-turner. I have started to pull together ideas for generating a hangout game show which crosses crowd-sourcing with social impact. Like a social version of kickstarter.

the niche

Currently, someone comes up with an idea or a prototype and they find partners, form a company, seek funding from start-up capital or angel investors or crowd-funding. This fits the competitive model of development.

Many people who are into world change use various online services to gather and share information, from dedicated website, newsletter posts and social media sites. A rising phenomena are ‘conversations’, g+ communities or fb groups curated by a suitably enlightened instigator. Some contributors use the space to connect people, to learn, and then use this information in their offline work, such as consultants. They value these communities, most in the same way we had the ‘water-cooler’ effect in businesses. There is no moneyflow, and yet the space is considered very valuable.

How about producing a hangout show that does both? Here’s a mock-up. An engagement so rich that participants not only reach consensus but commitment to produce a social impact within a short period of time. Social impact is the means by which the hangout is evaluated, and money follows. It allows a network of strangers to make decisions as if they were in charge of a company. It is active use of trust.

the hangout game

Screen Shot 2013-06-28 at 18.23.53

Appropriate a crowd-funding WP site, like the one above. By investing say £10, participants are invited to a hangout game. There are a number of games run at different times and facilitated by different people with different flavours of activity.

The hangout game itself consists of a few facilitators and everyone else watching live. Imagine this is running successfully with 100 people. The name of the game is answering the following question (which is similar to the question in action cycles):

What is the greatest social impact that up to 100 of us do with up to £1000 in the following week?

Participants use quora.com to put in their submissions, and everyone votes up and down the best options. The facilitator’s job is draw attention to those suggestions that may be added late or simply deserve more attention; namely to avoid that phenomenon where the top few listings end up escalating simply because they are highlighted (you know, top videos or articles put at the front of a website like youtube, evident in comments in youtube for example). The trick of the game is to reach consensus and commitment within the hour.

For example, someone may have an idea that could be converted into a kickstarter with a few people’s help. Perhaps within that week a handful of people help turn an idea, or a prototype, into a kick-starter programme and everyone else of the 100 people are the seed funders and recommenders of the project. Perhaps at the end of the following week, the project is funded $80,000. Some of this money is redistributed back to the ‘marketing’ team, the 100 people who participated, say $100 each. That’s a conversion of £10 for entry to $100 return.

Not all projects will have a financial return, but it depends very much on the ‘flavour’ of the hangout game show. Some will be more oriented to getting a financial return, but what matters is how people respond to suggestions that catch their attention enough to add to it, offering their skills, and helping it become a more realistic venture. The point is, we are realising the power of the collective to overcome what halts most of our open source projects — the difficulty of finding followers and securing marketing. Collectively, we have the resources to produce incredible results socially on a regular basis, if we are freed from the usual constraints imposed by company boundaries. Networks working with money flow.

deepening hangouts and ecosquared protocols

Not all the hangouts are in the form of this game. There are others which have a flavour of deepening engagement, learning how to reach consensus faster, how to confront differences and arrive at consilience. These hangouts are run along similar principles, where people who are interested fund the hangout and watch and perhaps participate, from one-to-ones to one-to-many’s, and ideally we will work out protocols to have many-to-many. The objective of these deepening hangouts is to learn what are the protocols which we need to adopt to make best use of our time when we meet collectively on line. Think intense and meaningful. Think of the equivalent to global summits, when potentates meet and need to make decisions about the direction of their governments, or CEOs making decisions as to the direction of their company.

As the fourth iteration of ecosquared, this is an attempt to get real results in the real world. With all our virtuality, our obsession with information and opinion, we do tend to lose the substance. The intention of these hangouts is to bring to the fore what we really value. The ecosquared protocols set up a game where money becomes ‘well-behaved’ mathematically speaking (MTTP), while opening up an economy that is based on subjective enumeration (SEA).

The direction of money is interesting. It appears that participants are paying for a ticket to be able to play the game and the money is going towards the centre, collecting as a ‘capital’. It would be nice to have a gift-it-forward in action, where the money generated from one hangout-game actually is carried over to another. One option is to initiate the Invitation-Protocol where the direction of the moneyflow is outwards. People are invited to play. People are showing trust by inviting people they know to the game-show, and even when they suggest activities to perform over the week, the decision actually hangs on those they have invited. That is, the centre is financially powerless, and the periphery is the collective decision-maker. A good thing for communities.

This may constitute the minimal set-up for us to test the protocols, including SEA, once the system gets up and running. The money that is attracted to the value the network produce, $10k in the example, is distributed by the values that people give one another. Those who contribute more actively are recognised by their peers collectively, and thus end up with more moneyflow. In our example, the team of which produced the video and wrote up the copy for the prototype pitch will naturally end up with more money-flow.

invitation for partners

I am looking for partners to set up the technical system, which looks relatively easy since there are templates for crowd-funding already available, and more importantly seed the game-show with their communities. That is, I am looking for people who would like to host one of these shows.

As a team, our collective input will produce a comprehensively appealing design, as we cover for each other’s blindspots. A team effort will produce a better invitation than this.

Auditions are now open 🙂