Hint-Based Systems

From P2P Foundation
Jump to navigation Jump to search

= using Stigmergy for human organization


Christian Siefkes:

"A hinting system also serves as an informal mechanism for prioritizing tasks: the more people care for a task, the more likely it is to be picked up by somebody (since the corresponding hints tend to become more visible and explicit, and since people are more likely to pick up a task they wish to be done).

Hints are impersonal, they give people a chance to look around what is there to do and then to decide for themselves. In peer projects there are no "overseers" that can tell people what to do—people decide for themselves.

Francis Heylighen doesn't use "hint" in the narrow sense. Every missing feature of a program that you notice is a hint indicating how/where this software could be enhanced; every bug is a hintindicating where it needs to be improved. Whenever you discover and report abug, or whenever you discover it and submit a patch for it (which I have done quite often), you have followed a hint which the developers of the program left (though they didn't leave it intentionally)." (list-en list, April 2008)


Christian Siefkes:

“I'm on the mailing list of a medium-size free software project [1] and there are regularly (not frequently, but from time to time) mails from people asking "I like the software and I would like to contribute, what can I do?" (or "I'm using the software and would like to give something back, what can I do?") Then somebody points them to the task list [2], and, if they have the energy to follow this up, they self-select themselves for a task. Of course, there are also people who come up with their own ideas (with about the same frequency, I would guess), but even they typically refer to the agenda or the goals of the project.” (list-en list, April 2008)

Fictional Example

From a discussion of Red Star, an utopian s-f book of 1908, by Alexander Bogdanov. Reviewer is Demet Dinler.


"In this part of the book, there is a conversation between our hero and the socialist Martian engineer who tries to satisfy the curiosity of his visitor. Our hero reads the tables at the entrance of the factory. Table 1 says: “The machine building industry has a surplus of 968.757 man-hours daily, of which 11.325 hours are skilled labour. The surplus at this factory is 753 hours of which 29 hours are skilled labour. There is no labour shortage in the following industries: agriculture, chemicals, excavations, mining and so on in a long alphabetical list of various branches of industry.” Table 2 says: ”The clothing industry has a shortage of 392.685 man-hours daily, of which 21.380 hours require experienced repairmen for special machines and 7.852 hours require organization experts.” And this goes like that.

The visitor asks the meaning of the tables to the engineer who replies:

- The tables are meant to affect the distribution of labour. If they are to do that, everyone must be able to see where there is a labour shortage and just how big it is. Assuming that an individual has the same or an approximately equal aptitude for two vocations, he can then choose the one with the greater shortage. As to labour surpluses, exact data on them need to be indicated only where such a surplus actually exists, so that each worker in that branch can take into consideration both the size of the surplus and his own inclination to change vocations (emphasis is mine).

The visitor realizes that the figures change when he observes them. The engineer explains:

- The figures change every hour. In the course of an hour several thousand workers announce that they want to change jobs. The central statistical apparatus takes constant note of this, transmitting the data hourly to all branches of the industry.

But the visitor does not understand how all those figures are obtained by the central apparatus in the first place. The engineer explains:

- The Institute of Statistics has agencies everywhere which keep track of the flow of goods into and out of the stockpiles and monitor the productivity of all enterprises what and how much must be produced for any given period and the number of man-hours required for the task. The Institute then computes the difference between the existing and the desired situation for each vocational area and communicates the result to all places of employment. Equilibrium is soon established by a stream of volunteers.

Up to now, the narrative tells us roughly a solution to the problem of allocation of labour, which is a problem of both capitalist and future socialist society. The central mechanism here is not like state planning, which makes production decisions much in advance. It produces live data (as contemporary Stock Exchanges do for prices, for instance) and this data becomes performative (see Donald Mc Kenzie’s An Engine, Not A Camera for performativity of technical models in economics) of the allocation of labour (‘The tables are meant to affect distribution of labour’). The Institute of Statistics and its agencies in different localities work almost like a quasi-market institution, which sends signals about shortage and surplus of labour according to which individuals make their decisions.

The visitor is curious about whether consumption is controlled in this system. The engineer explains that everyone takes whatever is needed in desired quantities. This confuses the visitor because if consumption is not controlled there may arise fluctuations, which destroy statistical compilations. The Martian engineer responds:

- Not at all. A single individual may suddenly eat two or three times his normal portion of a given food or decide to have ten suits in ten days, but a society of billions of people is not subject to such fluctuations. In a population of that size deviations in any given direction are neutralized and averages change very slowly and with the strictest continuity.

Our hero then asks whether the statistics work almost automatically on the basis of pure calculations. The engineer replies:

- No, not really, for there are great difficulties involved in the process. The Institute of Statistics must be alert to new inventions and changes in environmental conditions, which may affect industry. The introduction of a new machine, for example, immediately requires a transfer of labour in the field in which it is employed, in the machine building industry and sometimes also in the production of materials for both branches. If a given one is exhausted or if new mineral fields are discovered there will again be a transfer of labour in a number of industries –mining, railroad construction and so on. All of these factors must be calculated from the very beginning, if not with absolute precision then at least an adequate degree of approximation. And until firsthand data become available, that is not easy task (emphasis is mine).

Here, the narrative is not clear how several factors can be calculated in advance (something which the state planning agencies faced under socialism). But still, what is telling is that the system has to respond to changes, which will effect the need for labour. Invention of new machinery, exploration of new mines are will decrease or increase demand for particular types of labour. In the mainstream political economy, such changes (technological….etc.) result in mass transfer of labour from old industries to new ones. But we know that this is a long-term equilibrium, does not happen automatically and happens with waves of unemployment as well. But in the Martian context, this re-allocation is enabled without those social problems. The agencies are responsible for watching over all possible factors effecting labour supply and demand from different localities, perhaps even responding to the Hayekian warning about the difficulty to centralize information.

The visitor wonders whether the worker who is working passionately and likes his job in the factory should cut down his labour and change his job, given that there is surplus in the machine building industry. The engineer’s answer explains how it is not the individual’s responsibility to think of the benefit of the system, but it is the system, which ensures efficient allocation of labour as an aggregate effect of the sum of individuals’ actions.

Of course not. Why should he just take it upon himself to restore the equilibrium? The statistics oblige no one to do that. Everyone takes these figures into consideration when making their own plans, but they cannot be guided by them alone. If you were to want to begin working at this factory you would probably find a job; the surplus figure in the central statistics would rise by one or two hours and that would be that. The statistics continually affect mass transfers of labour but each individual is free to do as he chooses.”

There is a very important point here: Individuals are choosing themselves whatever they like within the constraints of supply and demand of labour, which are effected by the external factors (technology…etc.). They do not have to take any responsibility in helping to close a gap. The system does it naturally and facilitates labour transfers at the mass level. Individuals, by just making their own preferences in available places for their labour, become parts of a system, which statistically regulates supply and demand. In other words, labour movements are nothing but what emerges from the choices and activities of individuals (individuals who choose which job they will perform, individuals who explore new mines, individuals who invent new machinery), with the mediation of an agency which constantly collect and process the information about those activities, which becomes performative in the functioning of labour allocation system. So individuals take existing demand for labour in different industries while making their choices, but there is no compulsion for them to work in specific industries. The statistical agencies are facilitating a process, which is, from a neoclassical point of view, completely natural. Yet in this system, the boundaries between artificial and natural in the capitalist system seem to be more blurred and even shifting: The engineer explains that it is the ‘natural’ need of the Martians to work without any compulsion (implying overtly that something must have changed in society to give way to a new understanding of what ‘natural’ is). But at the same time an ‘artificial’ agency of Statistics is facilitating a process in which people both make their choices for jobs but at the same time contribute to the functioning of the economy as a whole (making sure that production of certain goods and services continues thanks to relocation of labour)

The choices and activities of individuals would probably not be possible without other developments, which allowed individuals to get access to education possibilities (enabling them to perform different jobs in one life-time), which created resources to invest in technology and innovation…etc. In this part of the book, we do not know how those other factors were made possible but the text gives us some clues that the system evolved historically and there were constraints, which had to be overcome. For instance, one of the pre-conditions of the system is the constant surplus labour in diverse industries, which allow people to be mobile across industries and freedom to choose.

Two hundred years ago, when collective labour just barely managed to satisfy the needs of society, statistics had to be very exact, and labour could not be distributed with complete freedom. There was an obligatory working day, and within those bounds, it was not always or fully possible to take the vocational training of the workers into account. However although each new invention caused statistical problems, it is also contributed to solving the main difficulty, namely the transition to a system in which each individual is perfectly free to choose his own occupation. First the working day was shortened, and then when a surplus arose in all branches, the obligation was dropped altogether. Note that the labour shortages indicated for the various industries are almost negligible, amounting to mere thousands, tens or hundreds of thousands of man-hours out of the millions and tens of millions of hours presently expended by those industries.

Overall narrative shows that the Martian statistical system of labour allocation takes some components existing in the capitalist system: i) labour divorced from personal ties and free to shift across sectors ii) self-regulating dimension of supply and demand of labour which makes possible mass transfers of labour iii) unlimited consumption. Those elements are combined with new components (education system which allows everybody to work in different jobs, constructed agencies which give signals about shifting movements of supply and demand as live data; a new notion of ‘natural’ tendency to work; reduced working hours). All those components are and they are all tied to the solution of a socialist problematique: How to create a system which will coexistently produce all goods and services necessary for society AND will enable everybody to fulfill their potentials and talents without problems of inequality, unemployment? I am not saying that we have all the answers to this question in this book, but there is an amazing intellectual exercise, which goes beyond the limits of its own time.

There is also a sub-text in the emphasis on individual choice: Nobody has to sacrifice himself/herself for the good of society. As long as the individuals make their choices by taking as the reference point, changing figures on the tables, they contribute to the functioning of the economy and society. They do not need to work harder than others. Aggregate effects of individual activities contribute to the needs of the collectivity as a whole (one could even speculate that what Adam Smith said regarding the self-interest serving the interests of society is actually realized in socialism rather than capitalism). So individual satisfaction and collective need are not mutually exclusive in this system. On the contrary they are tied to each other.

There is one final point of interest in the narrative. Even though there is a consistent surplus of labour usually, some shortages still continue to exist in different sectors. The visitor speculates that those shortages must be covered by later surpluses. The engineer explains:

- Not only by later surpluses. In reality, necessary labour is computed by adding a certain quantity to the basic figures. In the most vital branches of industry, the production of food, clothing, buildings, machines and so on, this margin can be as high as 5 per cent whereas in less important areas it is about 1.2 percent. Thus generally speaking the figures in these tables indicating shortages express merely a relative deficiency, not an absolute one.

In this paragraph, what is interesting is that necessary labour can be computed. In the previous paragraphs we were told that the agencies collect information about the need for labour in different industries, but did not tell how this need was determined in the first place. In capitalism, socially necessary labour is not possible to be computed in advance (I am not sure here Bogdanov would use socially necessary labour though), because only after the sale of their products, capitalists will have a chance to really understand whether the value of the commodities they produced will be realised. If not, perhaps they will need to cut down labour costs, fire workers, reorganize labour process, implement new technique … etc. It seems that there is no such pressure for the Martians, because they are able to calculate in advance necessary labour to produce goods and services and it is those calculations, which govern supply and demand of labour. It seems that there is no imperative of the law of value, which reacts back and governs the organization of labour. How this was made possible is not known in the text. But the fact that labour is computed in advance and those computations enable allocation and re-allocation of labour constantly can be read as a clue )even though it can not be the only one) to understand that social practices in Martian world might no longer enact the historically specific pressure of value. Perhaps I am over-interpreting here, not very sure.

There are several questions, which can be raised and unclear points, but the text gives us a unique opportunity for thinking about the creative forms of organisation of economic and social life in a communist society without replicating traditional experiences of state-led planning." (http://nightsoflabour.wordpress.com/2010/03/06/mining-potentials-in-capitalism-thought-experiments-in-the-communist-laboratory/)