Today after considerable delay I released a copy of the "metacommons manifesto" on PlanetMath, on the PlanetMath mailing list, and to several people who seem interested in supporting PlanetMath. The version I released is http://planetmath.org/~jcorneli/everything/PM/manifesto-Nov15.pdf (or .tex). I am putting a copy here too: feel free to add in-line comments or make changes. (We are revising this for future publication and to make it a more effective networking tool.) Please see also PlanetMath Development Agenda (a companion document).
--jcorneli Nov 15, 2006
For general-purpose (not inline) discussion, please use Metacommons Manifesto - Discussion.
--jcorneli Dec 13, 2006
See metacommons for general discussion of the metacommons idea.
--jcorneli April 22, 2008
New stuff more focused on developing a specifically "math-focused" metacommons is at Surveying the Math Metacommons. Focusing on math could make some parts of the project more feasible; other parts may continue to bleed out and be general. Both approaches can be pursued in parallel.
--jcorneli August 28, 2009
by: Joseph Corneli, Aaron Krowne, Raymond Puzio
This is a document about producing tools and practices for doing commons-based peer production. We will talk about how this process can be improved through the development of a metacommons. In order to make clear what we mean by this, we begin by defining the key terms.
Commons-based peer production (or CBPP) is a phrase first used by Yale law professor Yochai Benkler to describe decentralized, Internet-based collaborative projects . In order to keep the discussion simple, we will understand the term in somewhat greater generality, and include within its scope all decentralized efforts from which individuals derive some benefits from their participation, and for which there is some organizing principle that facilitates cooperation.
A key feature of CBPP is that it accumulates value associated with positive externalities coming from the mediated interactions of individuals.
We describe as meta-activities those activities associated with managing a commons. Meta-activities complement and support the activities that take place within the commons itself. Meta-activities may themselves have peer-production components (e.g. peer review can be considered to be a meta-activity), or they may be based on other sorts of organizational arrangements (e.g. firms). In particular, meta-activities may or may not be carried out by the same group of people who use the commons. The precise nature of meta-activities will, in general, depend on the nature of the associated commons.
Commons-based peer production has shown itself to be important in many areas, from business, to science, to entertainment, and beyond. In this document we will talk about how the idea of a metacommons – a commons for fostering the development of CBPP – can be used to make CBPP a even more effective means of producing value.
Our goals in writing this document are to open a channel of communication with possible supporters; to talk about why improved tools for doing CBPP constitute an important social good; and ultimately to assemble a productive team to work on CBPP issues. We hope to help motivate and facilitate the involvement of our readers in contributing to existing commons-based projects, in starting new ones, and in changing norms, laws and society at large in ways which facilitate the development of CBPP. This manifesto indicates and advocates an ongoing movement towards better understanding of how to manage vital common resources of all kinds.
Our views are grounded, particularly, in the experiences we've had with the wide array of meta-activities associated with running PlanetMath.org, a popular collaboratively-produced mathematics website. In order to develop a well-rounded exposition of a somewhat broad topic, our opinions will be contextualized herein by our experiences behind the scenes at PlanetMath; our experiences with a loose confederation of related projects; some discussion of the relationships between these projects and other organizations; and finally, our best attempt at conveying a broad social perspective on the matter at hand.
Throughout history, human knowledge has grown to fill vast libraries, and technology has become greatly refined. And yet, past progress merely sets the stage for future advancement. Improvements to CBPP are, fundamentally, improvements to the process of social progress itself.
We want to make knowledge more available, expand knowledge, give more people access to both content and to the process of knowledge creation. We want to effect social transformation towards the participatory and empowered, fun, free, responsible, and sustainable. In order to do this, we must understand knowledge production and commons-based peer production from both social and technological points of view. We should build and support improved infrastructure for knowledge production and communication. We should work together with others who share these goals, since it is by working together that we will achieve the most.
PlanetMath specifies our goals to the domain of mathematics. With this project, we aim to get a comprehensive and accessible source of mathematical information online. We are developing increasingly powerful domain-specific tools which enable anyone to be involved in mathematics as both a consumer and producer, and in important hybrid roles, where, for example, a student's questions can become part of future curricula.
PlanetMath, and the software platform called Noosphere which supports it, are not just valuable for their successes in the mathematics domain – they also constitute an ongoing experiment in CBPP. Indeed, improving our CBPP framework is one of the core goals of the PlanetMath organization, since this will help us serve the needs of all of our users better.
At the same time, it is clear to us that PlanetMath cannot reach its goals using a "pure CBPP" mode of operation. In particular, we will need a sustainable centralized organization to manage some parts of the overall effort; a contemporary example would be development of the Noosphere codebase (although we hope that in the future this development effort can be made more decentralized).
As a rough comparison, while markets operate in a fundamentally decentralized fashion, they nevertheless rely on the existence of a central trading floor, which had to be put together at some cost. The PlanetMath.org nonprofit plays this central coordinating role, and also works to address the various needs of PlanetMath users which are not currently met through CBPP. Accordingly, we have made developing the PlanetMath nonprofit an important high-level goal.
PlanetMath has become a beacon for people interested in "free math" and CBPP for the mathematical sciences, and it is often the first place that like-minded persons turn for help. As a result, a number of philosophically related projects are currently clustered around PlanetMath's organizational shell.
The Hyperreal Dictionary of Mathematics project is concerned with building artificial intelligence for mathematics. The project's main goal is to use this software to get all known mathematics online, without violating copyright law.
Metamath is a long-running project in formalized mathematics whose developers have released a large quantity of formal mathematical proofs to the public domain. This project promises to have a synergistic relationship with HDM.
Various other "planets" (e.g. PlanetPhysics) have been starting up, to support knowledge communities working in various areas of inquiry. We intend for Noosphere to facilitate content-sharing and other collaborations between these several projects.
A theme common to a several next-generation collaboration tools we have designed and started work on for PlanetMath and HDM is that they should facilitate naturally-overlapping communication between the members of related knowledge communities.
As one example application, we're interested in using CBPP ideas as the basis of a tool to help support the development of free/open source software. We call this project the "Code Market", after Jordi Carrasco-Munoz . In fact, the same ideas that apply to this project could be used as a framework for producing other public goods.
Collaboration is important on an inter-project scale as well as on the intra-project scale. We might well hope to be significantly better at collaboration than traditional (centralized, non-CBPP, top-down) efforts, which still fail to be collaborative in any deep sense of the word, resulting in considerable duplication of effort, and in many potentially fruitful synergies getting ruled out out-of-hand. A healthy metacommons should facilitate fruitful exchanges and ongoing collaborative relationships between various commons-based projects and stakeholders.
Any time society is confronted with a problem that requires an organized effort (as all important problems do), there is utility in having the following tools available: infrastructure that can support large-scale, multi-domain collaborations; and infrastructure for maintaining massively multi-dimensional information artifacts.
In other words, we should be able to build and work with information artifacts that coherently incorporate many different facts, methods of interpreting these facts, opinions, artistic displays, and computer programs – and so on – and which lend themselves to the simultaneous and fluid involvement of many participants.
Improved tools and theoretical models for CBPP are going to be of widespread appeal. However, research and development in this area is associated with high initial cost, particularly since participation from many different kinds of experts is called for.
On the other hand, many different kinds of organizations are already interested in CBPP (although some will know it by another name or description). Intuitively, CBPP development would be more efficient if these groups could manage to work well together than if they do not.
We plan to organize each of our projects (PlanetMath, HDM, Code Market, etc.), and the collection of these projects as a whole, into a coherent and fruitful commons.
The overall structure is of course going to be highly decentralized. We plan to be working at suitable meta-levels, guiding the system as a whole towards our stated goals.
We plan to provide improved support across the board for different interest groups operating within PlanetMath's commons.
In particular, we plan to add places to collaborate on and discuss current research; and, similarly, to improve our support for basic education by providing collaboration and learning tools for teachers and students.
In order to achieve these development goals, our core organization is going to need an income stream, as well as an influx of other resources, which we will have to obtain as a combination of arrangements with other organizations and contributions from individuals.
So far, PlanetMath users have not provided a sufficient source of funds or other resources that would support our meta-level efforts. This is because most users have (at best) exactly one kind of resource to offer: volunteer effort directly related to their work in mathematics. We have therefore broadened our search for financial support and peer producers interested in working at the meta-level.
Where it proves to be possible, we will expand our range of services to include value-added items that financially-empowered groups of users will be interested in paying for – which, notably, does not imply that the outputs have to be exclusively limited to this group, as free/open source software nicely illustrates. More abstractly, we are looking for ways to better understand and communicate to various groups of users the practical benefits they will obtain from contributing to our efforts.
We are currently in the phase of seeking out and negotiating the relevant relationships, and from this basis, formulating a suitable development plan.
It is typically useful for any group to have one central place to turn for information that is relevant to everyone in the group. For example, Google famously plays this role for Internet search.
We believe we can help create a set of central services which will be effective at serving the specific needs of the groups whose work is aligned with our goals. In particular, we hope to be able to develop a system for administering discussion fora with integrated subchanneling, and other novel CBPP communication tools, to share with everyone involved in our (thereby) coordinated effort.
In addition to generally facilitating better communication within and between projects, we plan to give special emphasis to the development of tools for sharing content. The idea is for projects with different foci to create diverse forms of content, which can then be reused creatively throughout a broader information commons.
Of course, this already happens to some degree. But we want to push for the development of a social and technological environment in which it happens better.
Although the efforts we have mentioned are ambitious, they are widely relevant, and we feel we can reasonably hope to garner support to continue and expand them. The NSF, the Department of Education, the Library of Congress, the United Nations, Wikimedia, First Monday, GNU, Google, various professional associations, foreign governments, for-profit companies, and philosophical and philanthropic groups are all likely to show some degree of support for the notions outlined in this document, and to have some synergies with us moving forward. It might therefore make sense for some of these organizations to become financial supporters of our projects, or directly aid in our work by providing other vital resources, or otherwise collaborate with us in the development of the metacommons.
Participants in practically any social organization, from city government to large-scale social movement, will find their experience enhanced by improved communication, access to relevant facts, and means for considering and weighing alternative courses of action. Understanding how to do these things well is what CBPP is all about. Put another way, the key to CBPP is to create an integrative platform that makes it easy for people to usefully share their work with others without going too far out of their usual practice.
The relationship between PlanetMath and the wiki AsteroidMeta, which it shares with our other projects, illustrates how a successful project can begin to take on the role of "project incubator." AsteroidMeta also shows how related projects can collaborate, exchanging ideas and arranging to share resources as appropriate. The several projects collaborating through AsteroidMeta have laid the foundations for a successful "free math" movement, and are the beginning of a promising CBPP development programme.
Over the course of its 5-year existence online, PlanetMath has shown that internet-based CBPP can be a success. On the other hand, we've seen that there are some things required for further growth that PlanetMath's commons does not provide. We've started a 501(c)3 nonprofit to support PlanetMath's development, and the strengths of this organization should help balance the shortcomings of "pure CBPP."
PlanetMath has all the characteristics you would expect of a fledgling non-profit. Through our CBPP approach, we have all but solved the sustainability problem for content (i.e., the content is fully maintained by the community). However, we're left with sustainability problems related to project development and societal impact.
Specifically, we would like to ensuring that the PlanetMath project will continue through time and maximize its impact, without running against shortages of resources that the project will need to in order to bring this about. The critical need is to balance volunteer effort and the CBPP mode of production with reliable support coming from a centralized organization devoted primarily to development and outreach.
We believe that solving these problems for PlanetMath would provide many instructive lessons and, likely, tools, which could be applied in other digital libraries and online community settings. More generally, we think it is likely that such solutions would be applicable to cooperative organizations and constructive social action in general.
We have been engaged with strategic planning meant to address these points. We have also been networking with other organizations that are likely to be able to help us in these areas.
As a collaboration tool, our shared wiki AsteroidMeta has a number of shortcomings, for instance, a site-wide recent-changes page that diminishes its value to readers who are not interested in reading updates on all of the projects using the wiki. However, it has the strength of being usable for any however-tangentially-related philosophical discussion connected to any of the projects or ideas involved, and, indeed, it is out of these sorts of discussions that a number of the projects have had their genesis (e.g. our Code Market). In light of this, the wiki can certainly be considered to be a preliminary success. Improving AsteroidMeta will be part of our ongoing work to develop improved collaboration tools.
PlanetMath has become part of other broader efforts, including the NSDL, CITIDEL, Wikipedia, the joint NSDL/MAA Math Gateway project, and Google's 2006 Summer of Code program. We have also established a broad community of interest in our work, both among "math people" and "free culture people."
We have furthermore identified items of potentially broad interest within our other projects (and in cases, we have found people who want to collaborate with us in developing these items).
Universities and the Internet are two of the institutions that have helped to established a broad cultural commons. Why do we want a metacommons when these great institutions already exist? Indeed, existing markets and laws could presumably be relied upon to coordinate CBPP development. However, the ansatz driving this manifesto is that we can be still more efficient if we apply the strategy of CBPP development to the development of CBPP itself.
Historical precedent for improvements in CBPP bringing about improved efficiency in cultural production is found in, e.g., Mersenne. Research on CBPP can show the way to further breakthroughs in efficiency.
Sustainability entails building sustainable relationships between various entities. One needs to know who to turn to for what. In any given organization the conditions for sustainability will have both external and internal components.
Building sustainable relationships is currently our main focus. We have seen many great ideas, but unless an effort can last long enough to achieve results, all the good ideas in the world don't amount to much.
In order to be successful, we will need ongoing investments of work effort, money, attention, knowledge, etc.; and of course a work product which is worthy of these investments.
We have sufficient funds in our budget to begin to offer a small amount of money to consultants (whether professional or amateur) to help with our organization's internal structuring and strategic positioning. We hope as well to attain gratis or quid-pro-quo help with organizational development from persons and organizations in positions of influence or ability. Indeed, this sort of thing that we hope that a metacommons could help provide.
Among other organizational issues that PlanetMath will have to address are the role of members, exactly what sort of staff we will be able to support, and how much and what kinds of help we will be able to offer to other projects.
There are many people who are working on things that are related to the things we are working on who nevertheless we are not collaborating with. This is something we would like to change; even relatively weak collaborations (e.g. sharing research reports) would be a big improvement. We expect that other researchers share our sentiments.
The metacommons begins with simple networking. As we work to improve our network, we will need to think strategically about our relationships with other groups. Sometimes we may not know immediately who we will be able to count on for a given resource, and other times we may not be taken seriously on our own. In either of these cases, a third party may help us succeed. Sometimes we may need to involve fourth or fifth parties in order to secure a good position!
It might appear that the largest problems facing society will require immense new investment if they are to be solved. Whether or not this is the case, one should not imagine that the world's largest problems will be solved de novo. No matter what the level of new investment is, people will have to figure out how to capitalize on existing and preceding efforts. Another basic fact is that costs can be reduced by find new ways to exploit widespread enduring effects, for example, by creating forms of entertainment which have the property that small increments of value are returned from the consumer's participation.
If you value the goals we mentioned above, then you should find a way to coordinate with other system agents (e.g. developers) some exchange or division of labor that will secure us all a better position.
PlanetMath can become a key site on the Internet not only for free math, but also for CBPP. You may be able to help us get there. At this point what we need most is to be able to engage a few developers.
Many of our projects are in very early stages of development; in fact, even those project we can safely say to be well-begun are in a relatively early stage, compared to what we expect from them eventually. All of our projects will need further investments if they are going to do new and exciting things. You can help us by funneling appropriate investments to these projects.
By the same token, if we have not been clear on what we're bringing to the table, please ask us questions! As the network of groups and individuals supporting our projects grows, managing the associated information about who is doing what will become a large-scale collaborative project in its own right. Indeed, that is the purpose of the metacommons.
Figuring out what role or roles are best for you to take relative to the commons is likely to be highly non-trivial. Luckily, we can all make use of the strategy of progressive improvements.
Here is a link to the latest release candidate. http://planetmath.org/~jcorneli/manifesto-rcb.pdf (substitute ".tex" to get the source code).
Please feel free to comment here. The plan is for this to be published (to its main audience) soon.
--jcorneli Sun Jul 9, 2006