Stigmergic Collaboration: A Theoretical Framework for Mass Collaboration – PhD Examination Reports

This page contains examinations by the readers of Dr. Elliott’s PhD thesis.

 

 

Howard Rheingold


 

Summary

 

It is my strong opinion that “Stigmergic Collaboration: A Theoretical Framework for Mass Collaboration” by Mark Elliott, fulfills the requirements for the PhD degree thesis in every specified attribute – and more. This thesis ought to be a book. It would be an important book. I was impressed by, and learned a great deal from Mr. Elliott’s broad and insightful command of knowledge in a number of relevant fields. Relatively little is known about the deep structure of mass collaboration and the environments that catalyze mass collaboration, a field that has taken on accelerated importance with the growth of the World Wide Web. Mr. Elliott’s thesis constitutes a substantial contribution to the field. It should be widely read. … With masterful succinctness, Mr. Elliott elucidated the connections between and significance of a wide variety of relevant sources in what is almost certainly the first comprehensive investigation of stigmergic collaboration. The methods, techniques, results, conclusions and implications are stated and visualized with admirable lucidity. … I recommend that Mark Alan Elliott be awarded the degree of Doctor of Philosophy without further examination or amendment.

 

Full text

 

It is my strong opinion that “Stigmergic Collaboration: A Theoretical Framework for Mass Collaboration” by Mark Elliott, fulfills the requirements for the PhD degree thesis in every specified attribute – and more. This thesis ought to be a book. It would be an important book. I was impressed by, and learned a great deal from Mr. Elliott’s broad and insightful command of knowledge in a number of relevant fields. Relatively little is known about the deep structure of mass collaboration and the environments that catalyze mass collaboration, a field that has taken on accelerated importance with the growth of the World Wide Web. Mr. Elliott’s thesis constitutes a substantial contribution to the field. It should be widely read. Mr. Elliott’s approach is both incisively analytical, precisely parsing the functional differences and relationships between coordination, cooperation and collaboration, and comprehensively far-sighted, bringing into focus the emerging connections between findings in several heretofore unconnected fields, and establishing a framework for mapping the uncharted territory of mass collaboration. This thesis is a careful, rigorous and sustained piece of work that demonstrates that a research apprenticeship is complete. The community of scholars in the discipline has a great deal to gain by admitting Mr. Elliott.

 

With masterful succinctness, Mr. Elliott elucidated the connections between and significance of a wide variety of relevant sources in what is almost certainly the first comprehensive investigation of stigmergic collaboration. The methods, techniques, results, conclusions and implications are stated and visualized with admirable lucidity. His tests, meant to be the first and not the last such tests, show the way for systematic empirical study of stigmergic collaboration theory.

 

The literary quality and general presentation are good enough to convince me that Mr. Elliott ought to take it a step further and make it into a book, I would recommend only that if such further development takes place, Mr. Elliott should spend time looking at the work of Elinor Ostrom on institutions of collective action. I am not saying that Mr. Elliott left out anything important; I suspect, however, that Mr. Elliott’s thinking about stigmergic collaboration could be deepened by looking at the kind of collaboration required to overcome social dilemmas and create institutions for collective action. Another recommendation would be for Mr. Elliott to include a discussion of a book published since the submission of this thesis, David Weinberger’s “Everything is Miscellaneous.” These suggestions are by no means meant to be a criticism of the existing work; rather, I offer them as suggestions for consideration when he expands the present work, as I hope he does. Again, in the interests of expanding his understanding,

 

I recommend that Mark Alan Elliott be awarded the degree of Doctor of Philosophy without further examination or amendment.

 

 

Francis Heylighen


 

Summary

 

This thesis is an original and comprehensive contribution to the literature on a novel and very important subject: the emergence of systems of mass collaboration on the Internet. The originality of the work consists in the application of this problem of the concept of stigmergy, which was hitherto basically limited to the study of collaboration in social insects and software agents. Apart from myself, I know of only a handful of authors who have noted that the theory of stigmergy is directly applicable to explain collaborative web communities, such as Wikipedia and the Open Source movement. … That the paradigm of stigmergic web collaboration is bound to become very important is eloquently argued in the thesis, and is something that I and other authors have also demonstrated in different publications. This novelty in itself is an important achievement. … In conclusion, this thesis together with the associated creative works fulfills all the conditions for their author to receive a PhD degree. His contribution is timely, original and important. Therefore, I recommend that he be awarded the degree of Doctor of Philosophy without further examination or amendment.

 

Full text

Content of the thesis

 

This thesis is an original and comprehensive contribution to the literature on a novel and very important subject: the emergence of systems of mass collaboration on the Internet. The originality of the work consists in the application of this problem of the concept of stigmergy, which was hitherto basically limited to the study of collaboration in social insects and software agents. Apart from myself, I know of only a handful of authors who have noted that the theory of stigmergy is directly applicable to explain collaborative web communities, such as Wikipedia and the Open Source movement. Even then, references to this phenomenon are not older than a year or two, which is understandable given that:

  1. stigmergy as a theoretical framework has only been used outside the domain of social insects since little over a decade;
  2. mass collaboration on the web has originated about a decade ago with the first wiki systems (although older systems such as newsgroups and bulletin boards already implemented some of the functionality).

 

This means that the present thesis is very timely, catching the emerging new paradigm at its very beginning (or perhaps even slightly earlier, since the thesis appears to have been started before the first mention in the literature of stigmergy applied to web collaboration). That the paradigm of stigmergic web collaboration is bound to become very important is eloquently argued in the thesis, and is something that I and other authors have also demonstrated in different publications. This novelty in itself is an important achievement.

 

Another achievement of the thesis is the systematic treatment of the subject, where the author has collected and tried to integrate a variety of concepts and conceptual frameworks that help us to understand the process of collaboration, especially as supported by digital tools. Among his personal contributions is the distinction between coordination, cooperation and collaboration as increasingly more complex and productive ways of “working together”.

 

While I like the distinction between collaboration and cooperation, where the former adds the element of collective creation to the former, the distinction between coordination and cooperation is less clearly explained, and may give rise to confusion. For example, the author at a certain moment lists collaborative filtering under cooperation (noting correctly that the adjective “collaborative” here is not the best choice) and the Amazon feature “people who bought this book also bought these other books” under coordination, while it can be argued that the Amazon feature merely implements collaborative filtering. There is also no clear distinction between coordination and synergy, which is listed as a central feature of coordination. Perhaps the central problem is that the definition used, such as “harmony of proximal relations” remains rather vague and ambiguous. (What is “harmony” and what is not? What are proximal relations?).

 

If this were a PhD in science or philosophy, I might have insisted on more exact definitions, but given that the author is a musician making a theoretical PhD in art, I must admire the relatively high level of precision he has achieved in such a complex and confusing domain in which he lacks basic training and where very little formal theory exists. More generally, I am very happy with the high level of interdisciplinary integration that the author has achieved, systematically assembling and organizing material from half a dozen disciplines and approaches, from insect behavior, to law, economics, sociology and information technology. This “birds eye” perspective across a very broad terrain is something that we very much need in this era of information explosions and fragmentation. Without it, we will surely lose sight of the forest for the trees.

 

 

Presentation of the thesis

As another minor point of criticism, I was not always happy with the literary style or language, which tends to be overly heavy, sometimes difficult to read, and not always informative. To take a (perhaps extreme) example, the first sentence of section 2.4.3 reads “… situated action is comprised of a number of realms of inquiry which emphasise differing aspects of the theory”. If I try to reformulate this sentence in a more understandable language, I come to something like “situated action is an approach with difference aspects” which is not very informative, since most approaches look at different aspects of their domain.

 

If the author plans to publish this work (as I would recommend), I would suggest drastically shortening it, leaving out vague or redundant statements, and using a less formal-academic vocabulary. It is a common misconception among students that academic research needs to be written with bookish, “academic” sounding words. However, if these words do not add anything to the meaning, it is preferable to use simpler, more common alternatives, such as “begin” instead of “commence”. The present work has avoided that trap most of the time, but still it could be made much more readable, thus expanding the number of people likely to profit from it. It is not so much that potential readers would be unable to understand the text, but simply that it would require them more time and cognitive effort than they might be willing to invest given the present oversupply of interesting information. Therefore, I suggest condensing the thesis to a single paper covering the main novel ideas.

 

Accompanying Creative Works

The creative works accompanying the PhD (three collaborative websites) adequately fulfill their purpose: illustrating, implementing and testing the theoretical ideas proposed in the thesis. They illustrate both the potential power and typical obstacles associated with Internet stigmergy. As I know from my own experience (and that of numerous colleagues and collaborators), it is not sufficient to have a good idea that is well implemented and well received by its prospective audience: that audience also needs to do the effort to systematically use the system, thus developing the system’s content, and making it more attractive for others to use and contribute to. Only then can it reach the critical mass necessary to “take off”. As the author points out, the success of his three websites was mixed in this regard.

 

In the case of the Australian Bill of Rights a promising start was followed by a virtual standstill, requiring a reorganization of the initiative. I wish him well with the new approach, but I know from experience that intellectually difficult, involved tasks, such as formulating legal rights, demand more from the project participants than most are able or willing to give without external rewards or enticements. Wikipedia has avoided this problem by attracting such a massive audience that there will always be enough people ready to contribute to a specific problem. The ABRI, by its nature, will attract a much smaller audience, and therefore it is unclear whether the participants can be sufficiently motivated to put in the necessary effort. A similar problem was encountered to a less degree with the Metacollaboration wiki, which can in principle recruit worldwide (not just Australians), and is more open-ended in its requirements, but still its goals are a little too abstract to quickly attract a motivated group of collaborators.

 

The problem of motivation was easier to tackle with the Collaborative Contract environment, where the student participation could earn credits by contributing and where the requirement on the types of contributions were much looser. However, the risk in this kind of environment is that it becomes little more than a chat room venting divergent opinions, trivia, and mere practical arrangements. Still, it is clear that such an environment has much to offer to the students of this class, helping them to collaborate more constructively.

 

I am still looking myself for a reliable formula for a collaborative website to tackle complex issues, so I can hardly criticize the author for his limited successes so far. Apart from the hugely successful Wikipedia, I have until now mostly encountered examples of how not to do it. With the lessons I have learned, I can make two suggestions:

  1. provide a system of incentives to motivate contributors (such as the credits gathered by the students). How to do this is a topic that could perhaps have been developed in somewhat more detail in the thesis, as it remains one of the most poorly addressed issues in stigmergic design. Indeed, the best-known examples of stigmergically collaborating agents (insects, robots and software agents) have their motivation preprogrammed, unlike human agents.
  2. make contributing as simple and easy as possible. The more technical, cognitive, or other hurdles people have to overcome, the less likely they are to get involved. Happily, the wiki technology, as used by the author, is pretty user-friendly to start with, but users can be guided along even more by providing templates with typical attributes or questions, default values, and examples of contributions, while keeping the interface as clear and uncluttered as possible.

 

For example, in the case of the ABRI, examples could be provided of existing Bills of Rights, from other countries or the UN Declaration of Human Rights. This will immediately focus the attention of the contributors on the core topics and issues. Moreover, it is much easier to criticize, edit or improve an existing solution than to create one from scratch.

 

 

Overall recommendation

 

In conclusion, this thesis together with the associated creative works fulfills all the conditions for their author to receive a PhD degree. The author has demonstrated sufficient understanding of the relevant literature, issues, and methods in the domain, and has applied his understanding to comprehensively expose and develop the subject. His contribution is timely, original and important. Therefore, I recommend that he be awarded the degree of Doctor of Philosophy without further examination or amendment.