Hiddinghausen talks, part 3: Pooling effort where free sharing fails

In HiddinghausenMy last talk in Hiddinghausen was a little talk given as a complement to the Commons Network idea. The Commons Network is based on the idea of free and unconditional sharing—that others share with you goods they have or produce because they like to do so (and the other way around). This leaves, inevitable, the question: what if there is nobody who likes to share what you need (and you can’t produce it for yourself)? What if people need some additional incentives to produce what you would like to have?

I have already given an answer to that question in my peerconomy book: people can join an explicit agreement to help each other to produce the goods each of them likes to have, sharing the necessary effort (the tasks to do) in some pre-agreed manner. The General Luxury Production System (GLuPS) is a slightly updated version of that idea. The word “luxury” in the name is meant to express the hope that the Commons Network will become sufficiently flexible and versatile to satisfy—at least—all of people’s essential needs (the “8 Essentials” discussed in my UPset talk).

This optimistic assumption stems from the generalization of Eben Moglen’s dictum

if you wrap the Internet around every person on the planet and spin the planet, software flows in the network. It’s an emergent property of connected human minds that they create things for one another’s pleasure and to conquer their uneasy sense of being too alone. (Anarchism Triumphant, emphasis added)

with which I concluded my Commons Network talk:

people like to do things that are useful for others and to be part of an active community.

→ Create suitable infrastructure (for sharing designs, for sharing goods, for organizing shared production, for spontaneous cooperation and stigmergic interaction), and production will follow.

The assumption (or rather, the insight) that people like to be creative and productive and that different people like to be creative and productive in wildly different ways implies that, in a society based on commons and peer production, production will organize itself far more “effortlessly” than we might have thought. Still, it certainly doesn’t guarantee that for every consumptive need somebody has there will be somebody else with a corresponding productive need waiting. In such cases, people can still get what they like to have by entering explicit effort sharing agreements. GLuPS can serve as a general framework for such cases, which will hopefully get rarer and rarer over time.

The rest of this article contains a handout of my GLuPS slides; they’re also available as an S5 slideshow.

General Luxury Production System (GLuPS)

Christian Siefkes
August 2008


What can people do if the open flatrate approach of the Commons Network doesn’t work? If nobody is ready to produce the goods you like to have, and you can’t produce them for yourself?

→ For such cases, peers can enter a joint agreement to help each other to produce what each of them likes to have, and to divide in a fair manner the effort that’s necessary to do so.

Open Flatrate vs. Luxury Production

(Optimistic) assumption: the open flatrate model of the Commons Network should be able to produce all the “normal” things and services that people need (the “8 Essentials”).

→ Explicit agreements about effort sharing (effort recovery) are therefore necessary only (if at all) for goods that most people don’t consider as relevant—for goods that aren’t part of the usual standard of living (“luxury goods”).

That’s why we call the complementary network for producing such goods the General Luxury Production System (GLuPS).

Effort Distribution

People help each other to produce the “luxury goods” they like to have, by contributing effort proportional to the effort necessary to produce the “luxury goods” they like to have.

In order to ensure that the different kinds of tasks (more or less popular ones) are picked up, a task weighting (weighted labor) approach as described in the peerconomy book can be used.

Possible modification to make formal task auctioning unnecessary:

  • Task niceness is self-estimated on a scale from -3 to 3 and averaged over everybody doing the task.
  • Task priority is measured on a scale from 1-8. Initial priority of all tasks is 1, but people can vote for tasks they would like to be done in order to increase their priority.
  • Time spent for a task is self-estimated.

Overall task weight is determined by combining task niceness (lower niceness → higher weight) with task priority (higher priority → higher weight). The production effort of a good is the sum of the task weight multiplied with the time spent on all tasks necessary for producing it (averaged over all goods of the same kind).

People give back the production effort spent on all the goods they take by contributing the same amount of production effort back to the system (helping to produce “luxury goods” that other people want).

  • Details of a possible implementation:
    • Convert niceness n to a niceness factor nf = 1 – 0.2n (range: 0.4-1.6)
    • Convert priority p to priority factor pf = 0.8 + 0.2p (range 1.0-2.4).
    • Calculate task weight by multiplying both: tw = nf * pt
    • → The lowest possible weight for nice+3/prio1 tasks is 0.4, the highest possible weight for nice-3/prio8 tasks is 3.84—people doing a very nice and unimportant task will have to work about ten times as much as people doing a very ugly and very important task.

Dealing with Cheating

  • Absolute and relative self-estimates of projects are visible.
  • Public user feedback (user trust?) ratings.
  • People decide with whom to cooperate and with whom not.

People Who Cannot Contribute

As described in the peerconomy book, people who cannot effort (e.g. because they are old, ill, or disabled) must not be excluded from consumption because of that.

→ There must be an effort redistribution mechanism in order to ensure that “luxury goods” are available to people who cannot contribute effort.

Entdecke mehr von keimform.de

Jetzt abonnieren, um weiterzulesen und auf das gesamte Archiv zuzugreifen.