PCP-news digest 1998-2000

News - Sep/Oct 1998

After the board meeting in August, the last two months have basically been used to start implementing the many activities planned during that meeting.

We are still working on an extensive overhaul of the layout of PCP-web. This has been facilitated by the creation of a layout template approach, where the placeholders for items such as "author", "date", "title", are automatically replaced by the appropriate fields in the database that holds all PCP documents. Thus, we only need to edit the single template file in order to change the layout of the hundreds of PCP pages all at once. As announced previously, the database fields have also been separated more clearly in the HTML documents, using fieldtext tags. A new "synopsys" field has been created to hold a 1 sentence summary of the document (see http://cleamc11.vub.ac.be/chaos.html for an example).

Several proposed layouts have been circulating, including some proposed by subscribers to this mailing list. Most usefully, Katie Lucas suggested to include a complete hierarchical pathname on each web page, e.g.

Home : Metasystem Transition Theory : Epistemology : Evolutionary Epistemology

This should alleviate the feeling of "lost in hyperspace" which many people experience when browsing through such a large and complex site as PCP Web. If you see immediately how the present page fits into the overall hierarchy, you will find other pages more easily. It is our intention to integrate this hierarchical path, which lists the documents that are"above" the present document in the hierarchy, with the present list of "child nodes", the documents that are "below" the present document. With a better graphical layout, we hope to convey this hierarchical up-down structure more intuitively than by the present list of "parent" and "child" nodes.

Since many people have asked us whether they can download the PCP-web as a whole, we now have provided a zip-compressed version of all the main documents. It can be downloaded as a single file, and then be uncompressed to produce a local version of PCP web on your hard disk. The file is available at http://cleamc11.vub.ac.be/PCP-Web.zip

Getting this single file is much to be preferred to the presently ever more frequent procedure of using a robot to download the whole site page by page. This procedure results in a heavy load on the server, slowing down its response to other users. This is particularly true when the robot activates all kinds of cgi scripts, that are used e.g. for annotations or search, but that don't produce any new information. So, please, stop using robots to suck in the whole site! (or at least program them to be polite, that is, collect files slowly and ignore files that end in ".cgi" or ".acgi").

We have also produced an experimental HTML file which contains the main documents as a single *printable* document, rather than a collection of hundreds of cross-linked pages. This is meant for users who would like to have a printout of all the basic material, so that they can read at leisure. This "book-like" version of PCP-web is available at http://cleamc11.vub.ac.be/PCP-book.html

NEWS IN BRIEF

The first steps have been taken to create a mirror of PCP web on the Los Alamos site.

Ashby's book "Introduction to Cybernetics" has been completely scanned in. We just need to finish some formatting details (tables, formulas, etc.) and then will make the book available as a single PDF file on the web. We plan to stick as closely as possible to the original layout of the book.

As announced in the previous newsletter, Liane Gabora, a memetics researcher, has joined the PCP team at the Center "Leo Apostel" in Brussels. Her newly created home page can be consulted at http://cleamc11.vub.ac.be/gabora/

Inversely, PCP researcher Johan Bollen, whose new home page is at http://pespmc2.vub.ac.be/, is preparing to move from Brussels to join the PCP group in Los Alamos.

A summary of the panel discussion which concluded the Memetics Symposium, co-organized by PCP, is now available at http://www.cpm.mmu.ac.uk/jom-emit/1998/vol2/panel_discussion.html

Principia Cybernetica News - Nov/Dec 1998

GENERAL NEWS

NEWS IN BRIEF

The other assistant editor, Johan Bollen, will be moving in February from Brussels to join the PCP group in Los Alamos.

The newest issue 2:2 of the Journal of Memetics: Evolutionary Models of Information Transmission, the most extensive one yet, has been published at http://www.cpm.mmu.ac.uk/jom-emit/1998/vol2/

Robert Rosen, a systems theorist and theoretical biologist, whose ideas on complexity, anticipatory systems, the definition of life and the modelling relation have inspired many debates on PRNCYB-L, unfortunately has died on Dec. 31.

After a rather quiet period, as its list of subscribers grows, the global brain mailing list is becoming more and more active. Its archive of messages can be consulted at http://www.cpm.mmu.ac.uk/~majordom/gbrain/thread.html

Principia Cybernetica News - Jan/Feb 1999

NEWS IN BRIEF

Johan Bollen has moved from Brussels to Los Alamos. His new coordinates are: E-mail: jbollen[at]lanl.gov, URL: http://ivanhoe.lanl.gov/johan/

The grant money that was being used to pay Johan while he worked in Brussels will now be used to support Alex Riegler, another PCP assistant editor, and Didier Durlinger, who supports the computer network at CLEA.

CLEA has now become the main center where the interdisciplinary journal "Foundations of Science" (http://www.vub.ac.be/CLEA/FOS/), published by Kluwer Academic, is prepared. Editor-in-chief is CLEA's director Dirk Aerts, assistant editors are CLEA members Liane Gabora and Jan Broeckaert. Although not officially associated with PCP, this journals covers many of the same fundamental issues, and it is certainly worth your consideration for publishing papers or special issues on PCP related themes. PCP assistant editor Alex Riegler has already started to prepare a special issue on constructivism.

Another, new, journal worth mentioning is "Entropy: An International and Interdisciplinary Journal of Entropy and Information Studies". PCP editor Francis Heylighen has joined its editorial board.

The two research proposals mentioned in the previous newsletter: "Collective Knowledge Development: elaboration of a theoretical model with applications on the World-Wide Web" and "Distributed Knowledge Structuring Systems" have been submitted, respectively to the Belgian National Fund for Scientific Research and DARPA. The latter proposal is also being adapted to be submitted to the NSF. If these proposals are accepted, they will create two or three new positions for researchers at CLEA.

The mirror of PCP-Web at LANL is nearing completion, as is the web edition of Ashby's "Introduction to Cybernetics" (it always takes much more time than expected ;-). Some other cybernetics and systems books are being considered for electronic publication on PCP-Web. Thus, we hope to create a real Principia Cybernetica "library" on the web!

Principia Cybernetica News - March/April 1999

GENERAL NEWS

The possibility will be explored whether the Foundation for the Future http://www.futurefoundation.com, which organized the Humanity 3000 Seminar, would be willing to organize and sponsor a workshop with invited specialists on the "global brain" idea. The Foundation shares PCP's interest in the past and future evolution of the universe and humanity, but lacks our conceptual framework inspired by cybernetics. (they compensate the lack of theoretical underpinning by near boundless financial resources, though: the Foundations plans to continue sponsoring conferences, research and other activities for the next 1000 years!).

The PCP-Web server has been replaced by a machine that is about 4 times faster. The improved speed should be especially noticeable during periods of high activity and for text searches. Of course, this does not solve the bottleneck of the network connection. This connection can be quite slow during the busy hours, especially for connections to North and South America. To resolve this problem, we have been setting up a mirror server at the Los Alamos National Laboratory in New Mexico. Unfortunately, the recent creation of a firewall (to protect internal information from outside abuse) around the lab has delayed the implementation.

The PCP editorial board is planning their yearly board meeting in Brussels in the beginning of July. The main purpose is to reorganize the structure of PCP-web, identify gaps, make a list of nodes to be written, and invite authors to write these texts.

One of the research proposals we submitted about intelligent webs was rejected by DARPA. Although he proposal was given top marks in technical and scientific merit and in the capabilities of the proposing team, it was considered not to be sufficiently closely connected to existing DARPA programs. A similar proposal is now being submitted to the National Science Foundation's program on "Knowledge and Distributed Intelligence". A preliminary evaluation of the proposal outline was quite positive.

Principia Cybernetica News - May/June 1999

NEWS IN BRIEF

Leor Grundlinger , an Israeli computer scientist, has visited the Brussels group, and decided to stay there in order to make a PhD thesis, under the supervision of Francis Heylighen. Although the precise theme of his research still must be specified, Leor would like to study the analogies between transactions between people in society and the communication between neurons in the brain. He is particularly interested in economics, market mechanisms, collective behavior and neurophysiology.

Johan Bollen, PCP editorial assistant on sabbatical in Los Alamos, has used the server log of PCP web to determine a matrix of association strengths between nodes by counting the numbers of users that have travelled from one node to another. This is equivalent to a non-interactive simulation of our learning web approach. This matrix can then be used for spreading activation to retrieve the nodes most strongly associated with a particular query.

The project we submitted on "Collective Knowledge Development" has not been retained for funding by the Belgian Fund for Scientific Research (because of lack of money, basically), but still remains in the running for funding by the university.

In spite of many delays along the road, the new layout for PCP web and the mirror server at Los Alamos are nearing completion.

The PCP editors, F. Heylighen, C. Joslyn and V. Turchin, will have their yearly meeting in Brussels on July 9-16, to discuss the general management of the project, and the specific organization and content of PCP web.

A new analysis of the server log has allowed us to create an updated "hit parade" of the most popular documents on PCP-web: http://cleamc11.vub.ac.be/HITPARAD.html The results are not always what you would expect, with some relatively minor pages that are deeply buried in the PCP hierarchy getting an unexpectedly large number of hits. This is probably due to links leading from popular servers directly into these pages (e.g. the PCP node on "Infinity" is directly mentioned in the Yahoo subject index), and to the fact that pages about popular keywords (e.g. "non-verbal communication" in our Web dictionary) are returned frequently by search engines.

Generally speaking, PCP web is quite well represented in the external links on other websites, as can be seen from the high "authority" or "PageRank" that PCP pages get in the new Google search engine (http://www.google.com). Searching for typical PCP subjects (such as cybernetics, memetics, global brain, or self-organization) on Google will produce many more PCP pages in the top ranks than the same search on another search engine. This means that PCP pages are considered to be authoritative by many other websites (see http://cleamc11.vub.ac.be/WEBCONAN.html for more details on determining the "authority" of web pages).

PRINCIPIA CYBERNETICA ELECTRONIC LIBRARY

This announcement was greeted with several enthusiastic reactions, and a flurry of downloads from our server. The announcement will be further distributed on different mailing lists and newsgroups. We hope that PCP has in this way contributed to the wider publication of cybernetics ideas, and made it easier for interested people to study this domain on their own. This is important especially given the few institutions where cybernetics and systems courses can be taken at present.

Subject: Principia Cybernetica Web has undergone a face-lift!

First, during their last meeting the PCP board of editors have agreed about a reorganization in the hierarchical structure of the web, in order to make it more transparent and coherent. This will be especially noticeable on the top level (the home page: http://cleamc11.vub.ac.be/), where two new high level categories have been created, "Navigation" and "Reference Material", which group important material that was otherwise scattered in many different places. This also implied the creation of several new nodes (pages), some of which must still be completed.

Then, the visual appeance of the web was redesigned, using a table layout. It includes a graphical header, with a new version of the logo and an incorporated menu bar that gives access to the most useful functions (search, what's new, etc.). There is also a new vertical side bar, which includes the author and date fields, the parents (and more generally ancestors) and children of the present page. The graphical metaphor uses arrows for moving up (parent, grand-parent, great-grand parent), down (children), left (previous page) and right (next page). We hope this will be more intuitively clear to the users than the old parent and child fields at the top, respectively bottom of the page.

The annotations have been moved to the bottom of the side bar under the header "Discussion", in order to emphasize that these pages are not refereed PCP nodes, but a forum for various arguments and counterarguments from our readers. The "annotate" function has been moved to its logical place below the list of "Discussion" links, and is now called "Add Comment...". The layout of the annotations themselves must still be redone, again to emphasize the difference between their discussion character and the more "authoritative" character of the standard nodes.

The server software has been upgraded, which should lead to a two times faster response and greater overall stability. Together with the replacement of the hardware by a 4 times faster computer a few months ago, this should result in much greater server capacity and shorter waiting times for users. However, the main bottleneck is always the Internet connection, rather than the server response, and therefore we have created a second server in a different location.

The menu bar includes a possibility to choose the server, either the original server in Belgium (Europe: EU), or the new mirror server in New Mexico (US). The mirror server is at present still experimental: it is not yet being updated automatically, and some of the links are not yet working properly. But it is expected that this mirror at Los Alamos National Laboratory will provide much faster and more reliable access at least for users in the Americas.

You can of course bookmark this mirror for quick access, but we would ask you to still use the old http://cleamc11.vub.ac.be/ address for making links from your own pages to PCP web, since this address is well-established on thousands of other websites and search engines, and we would like to avoid confusion because of people referring to Principia Cybernetica via two different addresses. We also cannot as yet guarantee that the mirror will remain available as such.

The rewritten "Help" page, reachable via the menu bar, explains the basic new and old functions.

Finally, some finetuning and updating has been done in several pages and functions, such as the Table of Contents (reachable via the "Outline" option in the menu) and the old Alphabetical Index (http://cleamc11.vub.ac.be/ALPINDEX.html).

We would ask you to try out these various new functions, and tell us how you like the new layout. We would be grateful for any comment, suggestion or criticism you care to make. This may help us to further improve Principia Cybernetica Web, so that it can maintain its status as most authoritative site in the broad domain of cybernetics and systems, and attract a growing number of users to explore its growing number of pages.

Principia Cybernetica News - July/August 1999

BOARD MEETING

We also agreed that Principia Cybernetica would participate in the 2000 World Congress on the Systems Sciences (http://www.isss.org/2000meet/2000anno.htm) organized by the ISSS in Toronto, by organizing a session on "Distributed Knowledge Systems and Social Evolution". The session will be chaired by Cliff Joslyn, but further practical details still need to be discussed

We finally agreed to use the data gathered on the usage of PCP web to test out some of the web learning algorithms that Johan Bollen is implementing in Los Alamos. Because PCP web is well structured, it is easy to compare various measures of how the web is organized (e.g. number of incoming and outgoing links per page, length of text, depth in the hierarchy, duration since last update) with the factual use of the web (e.g. how many times a particular page is consulted, how long that page is read on average, which other pages are reached from that page). This will allow us to test various algorithms to improve the web structure, e.g. by adding a list of "relevant links" to each page which is automatically generated on the basis of how users navigate through that page.

J

OURNAL OF MEMETICS

Principia Cybernetica News - Sept/Oct 1999

NEWS IN BRIEF

We hope to have a demonstration version of these adaptive recommendations available on PCP web in the near future. The basic idea is that the navigation bar would not only list the parent and child nodes of a particular page, but also offer a list of "suggested links" that the system has discovered by analysing the way users have browsed the system. These suggested links represent pages that have been used intensively by users that also used the present page, and that are therefore assumed to be relevant to the present page.

In addition to these general recommendations attached to a particular page, the system would also provide personal recommendations, based on the pages which a user has consulted previously. Depending on which pages you visited earlier, and how long you spent reading those pages, the system would calculate which other pages you are likely to find interesting, and offer you these as a list of suggestions. Johan Bollen in Los Alamos is already developing a program to generate such recommendations, as part of the final work on his PhD thesis.

Principia Cybernetica News - Nov/Dec 1999

NEWS IN BRIEF

PCP editor Francis Heylighen, his graduate student Leor Grundlinger, and PCP contributor Bruce Edmonds have started writing a proposal for the Future and Emerging Technologies programme of the European Union. The idea is to demonstrate the capacities of the algorithms we have developed for selforganizing websites, which adapt their links and recommmendations to their users. We are still looking for possible further European partners to participate in this proposal. More info at: http://www.cpm.mmu.ac.uk/~bruce/bsi/

A similar, but more limited, proposal has already been written and will be submitted to the Belgian Fund for Scientific Research-Flanders. The project will be led by Francis Heylighen and Alex Riegler, together with two local colleagues in Brussels.

THE PCP MIRROR

Now that hardware and software have been upgraded, both sites should be quite fast, but this of course depends on the connections between you and the site. If your access seems slow, try out the other site by clicking either "US" or "EU" in the menu bar above every PCP page. If you use Principia Cybernetica Web regularly, it is worth bookmarking both sites, so that if one of them is unreachable, for whatever reason, you can still consult the other one.

WARNING: although we encourage you to use and bookmark the Los Alamos mirror, we would ask you when you make links to PCP pages on your own web pages to always link to the original site: http://cleamc11.vub.ac.be/ This will minimize confusion, and will maintain PCP's high scores in various search engines. Since search engines determine the importance of a page by the number of links that point to it, dispersing links between different web addresses, even if they carry the same content, will make it less likely that other people will find the page. So please, always use the cleamc11 address for PCP in your links (you can add a second link with the lanl address, if you wish)!

ACTIVITIES RELATED TO PCP

The New England Complex Systems Institute has created a mailing list on complex systems. Since its creation a little over two months ago, it has been very active, discussing many of the same topics that have been discussed on our PRNCYB mailing list (and often by the same people), although the emphasis is less on philosophy and somewhat more on the "hard" scientific modelling. More info about the list, including its archives and how to subscribe can be found at http://necsi.org/discuss/discuss.html

Gottfried Mayer-Kress, an old acquaintance of our little PCP community and a member of the Global Brain Group, has started up a most useful service: the Complexity Digest. This is a weekly "newsletter-like" publication which summarizes interesting articles about issues related to complex systems, that have appeared in various scientific journals and magazines, such as Nature, Scientific American, Non-linearity etc., providing a link to the full article if available. The complexity digest is available on the web: http://www.comdig.org/ or via email. Let us all hope that Gottfried will find sufficient time and support to continue this excellent initiative.

The "Revue Internationale de Systemique", the French systems science journal, has changed name, publisher, and focus to become the "European Journal of Economic and Social Systems". It aims to apply various concepts related to systems, complexity, self-organization, evolution, autopoiesis, etc. to the understanding of our present society and economy. More info at http://www.edpsciences.com /docinfos/EJESS/

USER ANNOTATIONS

Moreover, thanks to the new layout (where annotations are now grouped in the side bar under the header "discussion") and some additional explanation of what annotations really are (see http://cleamc11.vub.ac.be/MAKANNOT.html), the use of annotations seems to have become more transparent to new users. The result is a flurry of often quite interesting and well-developed comments in various places of our web.

News - Jan/Feb 2000

Although the sudden popularity of PCP as a center for doing graduate studies pleases us, we must note that our resources are limited, both in time for supervising work, and in funding (no scholarships are available at the moment). Therefore, candidates must understand that they will have to arrange most practical support themselves, although we will of course help them with general guidance.

Because of general demand, we have also created a web page listing study programs in cybernetics, systems and complexity around the world (http://cleamc11.vub.ac.be/CSSTUDY.html). Since this list is far from complete, we would ask you to send us the URLs of any additional programs you may be aware of.

Allison DiazForte, a librarian from Australia, through volunteer effort, has helped us to produce a PDF version of Valentin Turchin's 1977 book "The Phenomenon of Science", a general introduction to the PCP philosophy. This makes it easy to print out the book as a whole, instead of having to read it chapter by chapter on the web. The PDF file is available at http://cleamc11.vub.ac.be/POS/TurPOS.pdf

The PCP editorial board has decided to introduce the new term of "evolutionary cybernetics", as a general description of PCP's scientific approach. Evolutionary cybernetics is a synthesis of the theories of self-organization and evolution, on the one hand, and cybernetics and systems theory, on the other hand. Its aim is to understand how organization and goal-directedness can emerge and evolve, in nature, mind, society and technology. MSTT, the theory of MetaSystem Transitions, is an essential part of evolutionary cybernetics. We have started to work on a paper that will describe the concepts and principles of evolutionary cybernetics in more detail. A page on it will soon be available on PCP web.

Because of a number of practical constraints, we are still not sure whether PCP will be able to organize a discussion session at the World Congress on Systems Science in Toronto (http://WWW.ISSS.org/2000meet/2000toc.htm), in July. If we organize such a session, its aim will be to introduce evolutionary cybernetics to a broad audience of systems researchers.

A new draft paper by Francis Heylighen, "The Global Superorganism: an evolutionary-cybernetic model of the emerging network society", is now available at http://cleamc11.vub.ac.be/Papers/Superorganism.pdf . This is an extensive review paper about the notion of a global organism/global brain, including a number of present and future developments in society and economy. All comments or criticisms are welcome!

TEMPORARY SHUTDOWN OF PRNCYB-L

However, because of Cliff's increasingly distant relationship with the Binghamton system, the list had intermittent interruptions and new subscribers were added with often long delays. We have now found out that the SUNY Binghamton computer system has completely stopped supporting LISTSERV mailing lists, and that PRNCYB-L has therefore been shut down definitively, without us even receiving a warning. We will try to set up a new PRNCYB-L mailing list as soon as possible on a different computer system (probably the one of Los Alamos National Laboratory).

If you would like to subscribe to the new PRNCYB-L, please send in your subscription form (http://cleamc11.vub.ac.be/PRNCSUB.html ) to the list administrator, Cliff Joslyn . If you were subscribed to PRNCYB-L before, but did not get any message recently announcing the temporary shutdown of the list, then it may be that the subscription address we had for you was out of date. In that case, please send your new address to Cliff.

News -Mar/Apr 2000

Assistant Editor Johan Bollen is busy writing his PhD thesis on "Application of Associative Network Models to Web Linking and Retrieval", and hopes to defend it at the end of summer. Michael Brooks, a journalist from the "New Scientist" weekly, has been interviewing several PCP people, and plans to write a big feature article on our ideas for a global brain/intelligent web. In the meantime, Johan has started to implement some of our algorithms to recommend web pages and create new links according to the duration of user's visits. We hope to test these out on Principia Cybernetica Web in the near future. If the results fulfill our expectations, we plan to submit a paper about this approach to an important jounal such as Science or Nature. This, together with an eventual "New Scientist" feature, may produce a lot of publicity for our global brain ideas.

News - May/August 2000

In July, Francis Heylighen and Jan Bernheim have presented their research on progress (http://cleamc11.vub.ac.be/PROGRESS.html) and the measurement of happiness at the International Society for Quality of Life Studies (ISQOLS) Conference in Girona, Spain. The reactions were quite positive, and our general feeling was that the QOL community has reached a level were simple empirical measurements of things like life satisfaction and positive/negative feelings can be used to build robust scientific models of the values and forces that drive individual action and social development. The only things lacking is an evolutionary-cybernetic mechanism to understand these developments, and that is where our PCP approach comes in.

A rather spectacular example of the power of such models is the prediction by Michael Hagerty, who was present at the conference, that Gore would win the US presidential elections with 52% of the votes. This prediction is based on a statistical analysis of the correlation between increases in QOL in a region and voting for the incumbent party, see http://www.gsm.ucdavis.edu/~mrhagert/Pres2000.html We're curious to see how close to the mark this prediction will be. Together with our Dutch colleague, Ruut Veenhoven, Hagerty has also shown that average happiness has increased over the last two decades: http://www.gsm.ucdavis.edu/~mrhagert/Papers/easterlinreply8.pdf

Francis Heylighen then participated in the Humanity 3000 Symposium of the Foundation for the Future (http://www.futurefoundation.org/humanity3000/index.html), but was somewhat disappointed since the discussions basically reiterated the ones at the Humanity 3000 workshop last year, except that the group was larger, so that there was less time for participants to express their ideas. On the other hand, he used the occasion to get acquainted with some leading thinkers working on themes close to PCP, such as the memeticist Susan Blackmore, Gregory Stock, author of "Metaman", and Christian de Duve, Nobel Prize winner and author of "Vital Dust", on the origin and evolution of life.

THE PCP-DISCUSS MAILING LIST

To make things more intuitive, and to clarify the relation with the PCP-news list through which this newsletter is distributed, we renamed the list from PRNCYB-L to PCP-discuss. The address is pcp-discuss[ at ]lanl.gov, but note that this is a closed list that you can only receive by submitting a request to the list adminstrator. We encourage anyone interested to explore and join our discussions on all aspects of evolutionary cybernetics. Please see http://cleamc11.vub.ac.be/MAIL.html for details on how to join.

The list was restarted in May with the same people that had subscribed to the original PRNCYB-L. Since then some ten new members have joined. The mailing list has as yet not been as active as PRNCYB during its heyday, but that is not surprising given the summer period, and a general phenomenon of "email tiredness" that many people submerged in mail presently experience. As before, this newsletter will include a list of topics discussed in the list during the past period.

THE GLOBAL BRAIN IN THE MEDIA

A feature article on our work with the "Global Brain" has appeared in New Scientist magazine, 24 June 2000, p. 22. It is based on extensive interviews with PCP board members Heylighen, Bollen, and Joslyn, and our colleagues in the global brain mailing list, Norman Johnson and Ben Goertzel. Although this paper has created a lot of publicity for our work, the journalist, Michael Brooks, has made it rather sensationalist, in addition to including a few factual errors. It emphasizes the scary, "Big Brother"-like possibilities, while minimizing the in-built protections against such abuse. For a somewhat more balanced view, read the accompanying New Scientist editorial.

The article is available at: http://www.newscientist.com/features/features_224417.html and the editorial at: http://www.newscientist.com/editorial/editorial.jsp?id=ns224444

As we anticipated in the previous newsletter, the publication of this article seems to have suddenly aroused a flurry of interest in our work, resulting in lots of email reactions and in further interviews with journalists in Belgium, Holland, Chile and Canada. Having to answer the same questions again and again (and then see the same misunderstandings crop up once more) has stimulated us to finally prepare a "Global Brain FAQ". The as yet unfinished text is available at: http://cleamc11.vub.ac.be/GBRAIFAQ.html Comments about questions and answers are appreciated.

Francis Heylighen has written a long paper reviewing the concept of the "Global Superorganism", an extension of the Global Brain concept to model the evolution of society as a whole, and especially to understand its future development. The paper will be submitted to the Journal of Social and Evolutionary Systems. Comments are invited to the draft text, available at http://cleamc11.vub.ac.be/Papers/Superorganism.pdf

PCP CONTRIBUTORS

Joel de Rosnay, an associate of the Principia Cybernetica Project, has authored several wide-ranging and well-written books touching on the concepts of systems theory, evolution of complexity and the Global Brain. His most recent book, "The Symbiotic Man: A New Understanding of the Organization of Life and a Vision of the Future", has now been updated and translated into English. We would recommend it to anybody interested in understanding complex systems and the future evolution of society.

More info on the Amazon page: http://www.amazon.com/exec/obidos/ASIN/0071357440/

John E. Stewart, an Australian theorist, has written a book, "Evolution's Arrow: The direction of evolution and the future of humanity" (Chapman Press, Australia, 2000), in which he develops a view of evolution very close to the one of PCP. Its main argument is that evolution progresses in the direction of cooperative organisations of greater scale and evolvability, up to global society. It is available at http://www4.tpg.com.au/users/jes999/

We are now discussing with John about the differences between his approach and our theory of metasystem transitions, in the hope of coming to a better understanding of both. The chief novelty of John's approach is his suggestion that MST's may take place because of an agent taking control over a group for purely selfish purposes, but then being turned by selective presssures into an efficient "manager" that promotes synergy and cooperation between the members of the group. This mechanism can be applied from the level of DNA taking control over an autocatalytic cycle, up to human society with its kings and emperors. A review by F. Heylighen of John's book together with some related books on evolutionary transitions will appear in the journal "Complexity". The paper is available at: http://cleamc11.vub.ac.be/Papers/Review_Complexity.pdf

On both sides of the Atlantic Ocean we have had extensive discussions with Bryan Thompson, a researcher from Cognitive Technologies Inc. Bryan had previously been the driving force behind two research proposals (to DARPA and NSF) about collaborative cognition in which we participated. His way of thinking is very close to ours, and it is likely we will collaborate more intensively in the future. Bryan is exploring the creation of an interest group within the WWW consortium, to discuss standards for a "cognitive web", and suggested to get some kind of legal protection for our "global brain" label (although several others have been using this label, for various activities and technologies, Bryan suggested that we should register a "global brain initiative").

ASSOCIATIVE NETWORKS MODELS OF PCP WEB

You can try out Johan's "enhanced" search engine for PCP web at http://bighorn.lanl.gov:8077/jserv-bin/SpreadAct_PCP_loop It is based on "spreading activation": the engine first retrieves the PCP pages that have the keywords you entered in its title, and then uses a matrix of associations to retrieve additional pages that are associatively related to the ones found first. The association matrix is still based on our old learning rules applied to the web log, and is therefore likely to be less efficient than a planned one based on our new algorithms that take into account duration of user visits.

At present, Johan is measuring the quality of the recommendations (precision and recall) by comparing the recommendations of the system with expert estimates of the relevance of the recommendations for a number of typical queries. (The experts, of course, are us, members and associates of the PCP board.) This will allow him to fine-tune the parameters of the system.

WORK ON ONTOLOGIES AT LANL

Some of this has been prompted by some very recent advances in computational linguistics. In particular, we're tracking the work begun by Steven Pinker, and much more fully developed now within the linguistics community, towards the identification of a small, canonical, generative set of semantic relations related to each other in a loosely hierarchical multiple inheritance type lattice. While these are proposed to represent the semantics of natural language texts, we've hypothesized that they should also inform a sufficient set of link types for ontological networks.

Among the other things we're examining are Sowa's conceptual graphs and Visual Basic extensions to use Visio as a GUI platform. The other essential ingredient is a sufficient, presumably XML-based, ontology exchange markup language for full read-write compatibility. We're tracking the standards community moving in this direction (e.g. The WWW Consortium, DOM, XSchema, RDF, etc.).

Cliff was recently invited to SRI International in Menlo Park, CA, to address a loose consortium of Silicon Valley researchers and developers trying to develop a Dynamic Knowledge Repository (DKR) within an Open Hypertext Standard (OHS). This group is led by Doug Engelbart of SRI fame, and whose early work (invention of the mouse and other essential elements of the present computer interface) we know as being so important for us and everyone, and who is now with the Bootstrap Institute http://www.bootstrap.org along with SRI. Cliff found Engelbart to be a charming and insightful man, who is very appreciative of everything PCP is trying to do.

Cliff's talk spanned a number of issues, including PCP (technology and form and content,), Lab activities, and the ideas described above. You can look at the overheads at http://www.c3.lanl.gov/~joslyn/KS_Team/sri.pdf.