The Now Economy

February 19, 2005

CodeCon 2005

Several people from CommerceNet attended CodeCon last weekend for varying amounts of time. The Wheat project, which we've contributed to, presented on Sunday and got a very good audience reception; Walter Landry of ArX, a variant of the GNU arch revision control system, presented a fascinating table of comparisons between the different free-software decentralized revision-control systems.

Decentralized source-code revision-control systems are at an interesting intersection; they allow any person to modify a piece of software with total independence, while facilitating their cooperation with others as much as possible. Historically, distributed systems have often achieved cooperation at the expense of independence, and this has limited their size or resulted in unfortunate social problems. Rohit Khare, the Director of CommerceNet zLab, wrote his doctoral thesis on architectural styles that support this intersection of cooperation and independence, and that area has been the focus of zLab.

ArX, along with several other projects presented at CodeCon, are at the forefront of real-world advances in this area: ApacheCA, a CA that makes trust decisions based on the GnuPG/PGP web of trust; OTR, an extension for instant messaging clients that provides cryptographic privacy for conversations without depending on third-party privacy infrastructure; and i-brokers, an Identity Commons/2idi project to develop a decentralized naming system for people on the internet.

I really appreciated the opportunity to spend a weekend with the people who are already doing the things that CommerceNet, so far, only dreams about.

Posted by kragen at 05:31 PM | Comments (0)

January 13, 2005

Tor

Yaron Goland's thoughts on Tor suggest an interesting decentralized system...

You don't know what you have to hide and by the time you figure it out, it will likely be too late. This is where Tor comes in. It makes it much easier to hide. The reason to use Tor isn't so much because you have something to hide, the reason to use Tor is so when you find out you had something to hide you can rest a little easier knowing that your secret may be protected.

He has some interesting design notes about constructing a network of routers to serve this purpose.

Posted by adam at 09:56 AM | Comments (0)

January 12, 2005

Kenosis: P2P RPC that decentralized BitTorrent

Ben Sittler noticed that Kenosis got Slashdotted yesterday:

UnderScan writes "Eric Ries, writer/programmer/CTO, authored an
article 'Kenosis and the World Free Web' at Freshmeat [Owned by
Slashdot's Parent OSTG]. Kenosis is described as a 'fully-distributed
peer-to-peer RPC system built on top of XMLRPC.' He has combined his
Kenosis with BitTorrent & removed the need for a centralized tracker.
He states: 'To demonstrate Kenosis's suitability for these new
applications, we have used it to improve upon another peer-to-peer
filesharing application that Just Works: BitTorrent. BitTorrent does
one thing incredibly well. Using a centralized "tracker," BitTorrent
manages efficient distribution of data that is in high demand. We have
extended BitTorrent, using Kenosis, to eliminate this dependence on a
centralized tracker.' See also the Kenosis README for details on using
Kenosis-enabled BitTorrent."

From http://kenosis.sourceforge.net/
Kenosis is:

• a fully-distributed p2p RPC system built on top of XMLRPC.
  
• zero-defect software.
  
• highly compatible.
  

The inventor recently quoth:

Four years ago, I wrote an article for freshmeat called "The World
Free Web" in which I described a way to make Web content available in
a distributed and anonymous way via Freenet. Back then, I expected, as
did many others, that Freenet was on the verge of completion, and all
that remained was to think of interesting new applications to write on
this new platform.

Now, for the record, I still have high hopes for Freenet and am still
a contributor to the Freenet Foundation. But as it stands, Freenet
simply does not work, and it is not a suitable platform for the
development of new applications.

Two years ago, Malcolm Handley and I started the Dasein Software
Partnership in order to create new peer-to-peer tools and applications
for the Free Software world. We started writing applications for
Freenet, but grew frustrated with Freenet's lack of stability. Next,
we switched to The Circle, a distributed hashtable based on Chord.
Despite its maturity, it too is not stable or reliable enough to form
a suitable platform.

So we decided that we would need to create a new system, designed from
the ground up for simplicity, stability, and scalability. We call that
system Kenosis.

Kenosis is a fully-distributed peer-to-peer RPC system built on top of
XMLRPC. Nodes are automatically connected to each other via a
Kademlia-style network and can route RPC requests efficiently to any
online node. Kenosis does not rely on a central server; any Kenosis
node can effectively join the network ("bootstrap") from any connected
node.

Posted by rohit at 06:34 PM | Comments (0)

January 06, 2005

Paper on Tradesports' contingent political betting markets

Experimental Political Betting Markets and the 2004 Election (PDF)

Justin Wolfers, Wharton. U.Penn
Eric Zitzewitz, Stanford GSB

Abstract: Betting on elections has been of interest to economists and political scientists for some time. We recently persuaded TradeSports to run experimental contingent betting markets, in which one bets on whether President Bush will be re-elected, conditional on other specified events occurring. Early results suggest that market participants strongly believe that Osama bin Laden’s capture would have a substantial effect on President Bush’s electoral fortunes, and interestingly that the chance of his capture peaks just before the election. More generally, these markets suggest that issues outside the campaign -– like the state of the economy, and progress on the war on terror -– are the key factors in the forthcoming election.

Our idea is a twist on the usual election betting markets. In standard election betting markets, traders buy and sell securities whose payoffs are tied to the performance of a particular candidate. For example, since 1988 the Iowa Electronic Markets have traded securities that pay a penny for each percentage point of the two-party popular vote garnered by each of the major candidates, and these have proven to be accurate indicators. And as far back as Lincoln’s election, there were organized betting markets on the Presidency, and again, these markets were quite accurate.

Our contingent securities are linked to both the election outcome and to specific events that could influence the election. The three new contracts that we listed pay $100 if President Bush is re-elected, AND (respectively):

• Osama bin Laden is caught prior to the election;
• The unemployment rate falls to 5 percent or below;
  
• The terror alert level is at its highest level (red).
  

Bettors call this sort of combination bet a “parlay,” and they have long recognized that if the two events in a parlay are not independent, then the pricing of the bet needs to take their interrelationship into account. We can exploit this to learn about what market participants think about the correlation between any two events.

Osama’s Capture Would Hand the Election to Bush

The data on the first contract are the easiest to interpret. At the time of writing, the Osama-and-Bush contract is at $9, suggesting a 9 percent probability that both events will occur. (We are sampling the mid-point of the bid-ask spreads). By comparison, a contract paying $100 if Osama is captured by October31 (two days before the election) is trading for $9.90.

Comparing these prices suggests that if Osama is captured, the markets believe the likelihood of a Bush victory to be 91 percent. (Why so high? The difference between the two prices, which is small in this case, corresponds to the likelihood that Kerry would win despite Osama’s capture.)

By comparison, a contract simply tied to whether Bush is re-elected is trading at $66.60, suggesting that overall, Bush is a two-in-three chance to be reelected. We can also use the Bush and Bush-and-Osama securities to learn about Bush’s chances if Osama is not captured. By purchasing a Bush contract and selling a Bush-and-Osama contract, we get a portfolio that pays $100 only if Bush is re-elected and Osama is not captured. Paying the $66.60, and receiving the $9, yields a total cost of $57.60.

Comparing this with the probability of 90.1 percent that Osama is not captured, this suggests that the markets believe that if Osama is not captured by the end of October, the likelihood of a Bush victory falls to 64 percent. These odds are still good for Bush, but nothing like his 91 percent chance if Osama is captured. It seems to be no exaggeration to state that the election depends more on whether Osama is captured than on any amount of campaign strategy.

A Better Economy Would Help Bush, But Not As Much

...A similar exercise can be performed to compute the effects of a better performing economy on the election.

Unfortunately, the likelihood of an unemployment rate below 5 percent— the contingency our contract depends on –is currently so low that the relevant contract has generated little recent trade. However, if we look to earlier trading in June 2004, it is clear that markets believe that a stronger economy would have given President Bush at least an 80 percent chance to win re-election, well above his overall odds at the time.

...

Other issues arise if we want to use what we learn from these markets to inform decision making. While other prediction markets have proved surprisingly immune to attempted manipulation, once the stakes become more than academic, the incentive for manipulators will rise dramatically. Once security prices are used for decision making, other traders’ beliefs start affecting their payoffs, and so a trader’s goal can move from predicting what will happen toward predicting what others think will happen. Careful market design can help on both issues, but the issues are fairly complex, and the focus of current study.

We should also emphasize that these results are only experimental. They come from fairly thin, and quite novel, markets. The prices represent the beliefs of a small and possibly quite unrepresentative population of traders, and the financial incentives here are small. They also involve asking people to evaluate small probabilities, which studies suggest many have a hard time doing.

...

One thing we know about the participants in this market is that unlike political pundits on Sunday television, these participants are putting their money where their mouths are. That feature is what makes market opinions well worth watching.

Posted by rohit at 11:06 AM | Comments (0)

December 18, 2004

NYT Hightlights Decentralized Nature of Modern Filesharing

The Ninth Circuit had found Napster liable because the company itself maintained and controlled the servers that searched for the digital files its users wanted to download. Grokster and StreamCast, by contrast, operate decentralized systems that allow users to find each other over the Internet and then exchange files directly. Consequently, the appeals court said, the two services did not exercise the kind of control that could lead to legal liability for infringing uses.

Technology > Justices to Hear Case on Sharing of Music Files" href="http://www.nytimes.com/2004/12/11/technology/11copyright.html?ex=1260507600&en=6300c274ff92a111&ei=5090&partner=rssuserland">The New York Times > Technology > Justices to Hear Case on Sharing of Music Files

Justices to Hear Case on Sharing of Music Files
By LINDA GREENHOUSE

Published: December 11, 2004

WASHINGTON, Dec. 10 - The Supreme Court, accepting urgent pleas from the recording and film industries, agreed on Friday to decide whether the online services that enable copyrighted songs and movies to be shared freely over the Internet can be held liable themselves for aiding copyright infringement.

For the entertainment industry and for everyday consumers, the case is likely to produce the most important copyright decision since the Supreme Court ruled in 1984 that the makers of the videocassette recorder were not liable for violating the copyrights of movies that owners of the devices recorded at home.

The earlier decision, Sony Corporation of America v. Universal City Studios, ushered in one technological revolution. The new case, Metro-Goldwyn-Mayer Studios v. Grokster Ltd., No. 04-480, comes as another is already well under way. More than 85 million copyrighted songs and a smaller but rapidly growing number of movies are downloaded from the Internet every day by people using file-sharing services.

A lawsuit filed by the entertainment industry in 2000 eventually put the earliest file-sharing network, Napster, out of business. The industry then brought suits against individuals who shared copyrighted material over the Internet, but that proved inefficient. The defendants in the case the justices accepted today are two of the newer services, Grokster and StreamCast Networks, which offers peer-to-peer software called Morpheus.

Two lower federal courts in San Francisco, the federal district court there and the United States Court of Appeals for the Ninth Circuit, ruled that the technology of the new services is different from Napster's in a way that immunizes them from liability for aiding copyright infringement.

The Ninth Circuit had found Napster liable because the company itself maintained and controlled the servers that searched for the digital files its users wanted to download. Grokster and StreamCast, by contrast, operate decentralized systems that allow users to find each other over the Internet and then exchange files directly. Consequently, the appeals court said, the two services did not exercise the kind of control that could lead to legal liability for infringing uses.

In their Supreme Court appeal, the recording industry and the Hollywood studios are arguing that the Ninth Circuit erroneously relied on a distinction without a difference. "The Ninth Circuit has immunized Grokster and StreamCast from copyright liability for the millions of acts of copyright infringement that occur on their services every day, and that could not occur but for the existence of their services," the plaintiffs' brief told the court.

The large group of plaintiffs includes the Motion Picture Association of America, the Recording Industry Association of America, Warner Brothers Entertainment Inc. and a class of 27,000 music publishers and songwriters. The Supreme Court will hear the case in March and is expected to decide it by late June.

The plaintiffs' appeal included dire warnings about what would befall the entertainment industry and individual artists if the Supreme Court failed to intervene. The brief asserted that the Ninth Circuit's decision, issued in August, "poses a grave threat to the very foundations of the copyright law's incentive system for promoting the progress of science and the arts, and will profoundly reshape our nation's system of copyright in the digital era."

Friend-of-the-court briefs were filed on the industry's behalf by the attorneys general of 40 states; a group of 130 recording artists; and 15 national organizations, including professional sports leagues, that described themselves as dependent on "meaningful copyright protection."

For their part, Grokster and StreamCast described the plaintiffs' arguments as "hyperbolic" and told the court that there was no need to enter a dispute that Congress was able to resolve. Several bills that directly address file sharing are pending before Congress now, the brief said.

As part of their argument that the plaintiffs' cries of doom were exaggerated, the brief quoted a 1983 warning by Jack Valenti, then president of the Motion Picture Association of America, which was challenging the video cassette recorder as a copyright infringer, that "the VCR is to the American film producer and the American public as the Boston Strangler is to the woman home alone." Video sales turned out to be very profitable for the studios.

Grokster and StreamCast offered their own "friends of the court": the Computer and Communications Industry Association, representing software developers and Internet service providers, and the Internet Archive, a nonprofit "Internet library" of historical collections. They urged the court not to interfere with new technology forms and markets.

How to interpret and apply the 1984 Sony decision is likely to be central to the Supreme Court's decision. The court held there that because VCR's were capable of "substantial noninfringing use," like time-shifting by people who recorded programs for later viewing, the manufacturers should not be liable for any infringing uses. The Ninth Circuit applied that principle to the file-sharing context. But the plaintiffs argue in the new case that because it is undisputed that most users of the file-sharing services use the software in ways that violate copyright, the immunity the court gave the VCR manufacturers should not apply.

Posted by rohit at 08:07 AM | Comments (0)

December 12, 2004

Caltech/UCLA's jMarkets open source toolkit

There are some screenshots from their demo site.

"The main asset of the Caltech Laboratory for Experimental Finance (CLEF) is its markets software, called jMarkets. It allows us to run large-scale financial markets experiments reliably and flexibly over the web. jMarkets is pure-Java and J2EE compliant. It was developed from the beginning to become open-source, and a first release to the academic community is planned for 15 November 2004. We decided to make jMarkets open source, in order to promote experimental research on financial markets. Our research to date has demonstrated the potential of experiments, paving the way to investigating longstanding questions. But many more exciting questions exist than we can address on our own. jMarkets' features will make it accessible to other research groups, usable in a variety of locations and populations. It is to become a tool to which many research groups will have easy access and to which they will be able to contribute."

Welcome to jMarkets

jMarkets is meant to provide the infrastructure for running large-scale experiments. It is built around a specific theoretical framework, namely, General Equilibrium Theory (GE). This is the branch of Economics that studies large, competitive, interdependent systems.

...

Peter Bossaerts and William Zame are the scientific supervisors of the jMarkets project; Walter Yuan is the technical supervisor; Raj Advani is the lead programmer.

Posted by rohit at 06:49 PM | Comments (0)

A "Decentralized" A-life Visualizer

the breve simulation environment : home

breve is a free, open-source software package which makes it easy to build 3D simulations of decentralized systems and artificial life. Users define the behaviors of agents in a 3D world and observe how they interact. breve includes physical simulation and collision detection so you can simulate realistic creatures, and an OpenGL display engine so you can visualize your simulated worlds.

breve simulations are written in an easy to use language called steve. The language is object-oriented and borrows many features from languages such as C, Perl and Objective C, but even users without previous programming experience will find it easy to jump in. More information on the steve language can be found in the documentation section.

breve features an extensible plugin architecture which allows you to write your own plugins and interact with your own code. Writing plugins is simple and allows you to expand breve to work with existing projects. Plugins have been written in breve to generate MIDI music, download web pages, interact with a Lisp environment and interact with the "push" language. To develop your own plugins, you'll need to download the plugin SDK from the download section.

Klein, J. 2002. breve: a 3D simulation environment for the simulation of decentralized systems and artificial life. Proceedings of Artificial Life VIII, the 8th International Conference on the Simulation and Synthesis of Living Systems. The MIT Press.

[from a recommendation by Kai Mildenberger]

Posted by rohit at 05:25 PM | Comments (0)

December 11, 2004

Newswire: Cornell's decentralized news network ("zNN")

Newswire - Collaborative real-time content delivery

Why Newswire? Newswire is a peer-to-peer, fully decentralized system that brings news to your desktop, within seconds after it is published.  This technology gives the community the power to weave a collaborative infrastructure for the delivery of essential information to individuals in a robust, scalable and secure way. Newswire is a survivable system which will deliver news to subscribers even if large parts of the infrastructure are under attack or stress.

The development of Newswire followed discussions after the 9/11 attacks when flash-crowd effects made it impossible for many to reach essential news sites. Email and weblogs proved more effective technologies in that situation, although both still suffer from the centralized risks of overload and single-point-of-failure. A more robust approach is to use a peer-to-peer structuring of the system and provide strong incentives for people to collaborate on the delivery of information to everyone by giving up a bit of bandwidth and CPU cycles.

Newswire has the additional effect that it can significantly reduce the load on the websites carrying real-time news information by providing hints about when information has changed. This would change the frequent redundant poll behavior seen by many news sites into more effective visits.

The Newswire Technology What is the technology that makes Newswire so special? At the core is epidemic communication and state management to maintain distributed knowledge about  subscriptions, participant network capabilities and forwarding load. The epidemic technology was first developed by Alan Demers and friends at PARC, but has been completely revised at Cornell in the past years

The structuring into zones and virtual hierarchies is based on the small worlds phenomenon. When you organize participants into  small groups of local participants with some knowledge of other remote nodes you can construct very effective routes in a decentralized, autonomous manner.

These technologies are used to build a loosely coupled, dynamic overlay network, that continuously monitors network load and capacities to achieve a fair load.

Subscription information stored in Bloom Filters is aggregated such that simple forwarding decisions can be made anywhere in the network based on the location of the publication and the direction where possible subscriber are localed.

Vogels pointed to half a dozen really great papers in All Things Distributed: Epidemic Computing at Cornell

... that having to scale up your system is no longer a burden, but it becomes an advantage and you can deliver on the true promise of scale-out: more nodes means a more robust system.

Also, he noted: All Things Distributed: Wired on the Scalability of Feed Aggregators

We have had some legal issues with the deployment of the newswire beta, and a new core communication component is being developed. But I would be happy to break open the ideas in the system and work with others to see whether this can be a good experimentation platform.

Posted by rohit at 03:27 PM | Comments (0)

The Supremes Learn About Sharing

Technology > Justices to Hear Case on Sharing of Music Files" href="http://www.nytimes.com/2004/12/11/technology/11copyright.html?8br">[NYT] Justices to Hear Case on Sharing of Music Files

The Ninth Circuit had found Napster liable because the company itself maintained and controlled the servers that searched for the digital files its users wanted to download. Grokster and StreamCast, by contrast, operate decentralized systems that allow users to find each other over the Internet and then exchange files directly. Consequently, the appeals court said, the two services did not exercise the kind of control that could lead to legal liability for infringing uses.

Posted by rohit at 11:34 AM | Comments (0)

December 08, 2004

Decentralizing BitTorrent

Courtesy of zLab Associate Kevin Hughes' KevsNews:

Slashdot | Decentralizing Bittorrent
... "Exeem is a new file-sharing application being developed by the folks at SuprNova.org. Exeem is a decentralized BitTorrent network that basically makes everyone a Tracker. Individuals will share Torrents, and seed shared files to the network. At this time, details and the full potential of this project are being kept very quiet. However it appears this P2P application will completely replace SuprNova.org; no more web mirrors, no more bottle necks and no more slow downs. Exeem will marry the best features of a decentralized network, the easy searchability of an indexing server and the swarming powers of the BitTorrent network into one program. Currently, the network is in beta testing and already has 5,000 users (the beta testing is closed.) Once this program goes public, its potential is enormous. "

Posted by rohit at 10:40 AM | Comments (0)

December 01, 2004

Beyond Web of Trust: Enabling P2P e-Commerce

Beyond "web of trust": Enabling P2P E-commerce:

The long-term goal would be to design a fully functional decentralized system which resembles eBay without eBay's dedicated, centralized infrastructure. Since security (authenticity, non-repudiation, trust, etc.) is key to any e-commerce infrastructure, our envisioned P2P e-commerce platform has to address this adequately. As the first step in this direction we present an approach for a completely decentralized P2P public key infrastructure (PKI) which can serve as the basis for higher-level security service. In contrast to other systems in this area, such as PGP which uses a "web of trust" concept, we use a statistical approach which allows us to provide an analytical model with provable guarantees, and quantify the behavior and specific properties of the PKI. To justify our claims we provide a first-order analysis and discuss its resilience against various known threats and attack scenarios.

In support of our belief that C2C E-commerce is one of the potential killer applications of the emerging structured P2P systems, we provide a layered model for P2P E-commerce, demonstrating the dependencies of various security related issues that can be built on top of a decentralized PKI.

Posted by rohit at 01:55 PM | Comments (0)

November 29, 2004

MapReduce for decentralized computation

I just added a technical note to our Wiki extending the work of Dean and Ghemawat on MapReduce, a support library for programs that take advantage of large clusters such as those at Google. The fundamental problems of writing distributed systems like these --- latency, naming (or memory access), partial failures, and concurrency are toy versions of most of the key problems of decentralized systems. (Agency conflict is the one problem that distinguishes decentralized systems from distributed systems, and it dramatically worsens the other four.)

It occurred to me that MapReduce has the capability to handle some of the agency problems of decentralized systems, particularly those that affect reliability, in a remarkably simple way:

In the usual case, where these functions are deterministic, they can be executed on two administratively-independent servers, and the results (which, in the Google case, are merely files) can be compared. If they differ, the same results can be recomputed on more administratively-independent servers to see which ones were correct.

Posted by kragen at 12:53 PM | Comments (0)

November 28, 2004

Jim Youll's Thesis: The Atomic Market

The Atomic Market:

Today's electronic marketplaces are closed, centralized and inflexible.

We propose a new type of electronic marketplace, which we refer to as an "atomic market." Atomic markets differ from today's electronic marketplaces in that they are (1) open-ended, (2) decentralized and (3) component-based. The atomic market supports short-lived markets created around the individual components of everyday transactions. The traders in an atomic market are agents, software that acts as a proxy for an actual buyer and seller.

The atomic market allows expressive interactions among trading agents, leading to productive, automated agent-based transactions. The focus is on the technical infrastructure for atomic marketplaces, specifically the use of logic as a basis for the decomposition of transactions and the negotiations between the different agents.

The Atomic Market

S.M. Thesis, submitted August, 2001 to the Program in Media Arts and Sciences:
Peer to Peer Transactions in Agent-mediated Electronic Commerce (1.7mb PDF)

Posted by rohit at 10:32 PM | Comments (0)

November 26, 2004

Benchmarking Competition for Trading Agents

Found this linking from Vorobeychik, a student of the current sigecom chair, M. Wellman. He wrote a *great* survey 5-pager at http://www.eecs.umich.edu/~yvorobey/2002/YABackground.pdf

http://www.eecs.umich.edu/~yvorobey/professional.htm#projects

http://www.eecs.umich.edu/~yvorobey/

http://ai.eecs.umich.edu/people/wellman/research/group.html

PhD Graduates

(Chronological order)

name (defense date) [current affiliation]

thesis title

John Cheng (Jan 1998) [Capital Networks]

Essays on Designing Economic Mechanisms
co-advised with Carl Simon

Chao-Lin Liu (May 1998) [Taiwan Nat'l Chengchi University]

State-Space Abstraction Methods for Approximate Evaluation of Bayesian Networks

Tracy Mullen (Jan 1999) [Pennsylvania State University]

The Design of Computational Markets for Network Information Services

David Pynadath (Jan 1999) [USC Information Sciences Institute]

Probabilistic Grammars for Plan Recognition

Junling Hu (Jun 1999) [Talkai, Inc.]

Learning in Dynamic Noncooperative Multiagent Systems

Peter Wurman (Jul 1999) [North Carolina State University]

Market Structure and Multidimensional Auction Design for Computational Economies

David Pennock (Sep 1999) [Overture]

Aggregating Probabilistic Beliefs: Market Mechanisms and Graphical Representations

William Walsh (May 2001) [IBM Research]

Market Protocols for Decentralized Supply Chain Formation

Trading Agent Competition - TAC Classic - Game Description

In the TAC shopping game, each "agent" (an entrant to the competition) is a travel agent, with the goal of assembling travel packages (from TACtown to Tampa, during a notional 5-day period). Each agent is acting on behalf of eight clients, who express their preferences for various aspects of the trip. The objective of the travel agent is to maximize the total satisfaction of its clients (the sum of the client utilities).

Travel packages consist of the following:
A round-trip flight,

A hotel reservation, and

Tickets to some of the following entertainment events

Alligator wrestling

Amusement park

Museum

Illustration of the environment a TAC agent operates within. To the left are its eight clients and their preferences, in the middle all its competitors lined up (7 competitors/game), and to its right are all the auctions (28 simultaneous auctions of three different types).

There are obvious interdependencies, as the traveler needs a hotel for every night between arrival and departure of the flight, and can attend entertainment events only during that interval. In addition, the clients have individual preferences over which days they are in Tampa, the type of hotel, and which entertainment they want. All three types of goods (flights, hotels, entertainment) are traded in separate markets with different rules.

A run of the game is called an instance. Several instances of the game are played during each round of the competition in order to evaluate each agent's average performance and to smooth the variations in client preferences.

Posted by rohit at 02:27 PM | Comments (0)

November 16, 2004

Real Names

John Battelle points out that RealNames has relaunched. Founder Keith Teare writes:

The problem addressed by RealNames - that is the poverty of the DNS as a naming and navigation system for the world’s internet users - remains unresolved.

Posted by adam at 08:03 PM | Comments (0) | TrackBack

November 11, 2004

Market Based Resource Allocation

Fine is a former student of Ledyard's, who also Hanson's advisor.

Research at HP Labs : Information Dynamics Lab : Papers : Eliminating Public Knowledge Biases in Small Group Predictions

Kevin Lai, Bernardo A. Huberman, Leslie R. Fine

HP Laboratories
Palo Alto, CA 94304

Abstract

P2P clusters like the Grid and PlanetLab enable in principle the same statistical multiplexing efficiency gains for computing as the Internet provides for networking. The key unsolved problem is resource allocation. Existing solutions are not economically efficient and require high latency to acquire resources. We designed and implemented Tycoon, a market based distributed resource allocation system based on an Auction Share scheduling algorithm. Preliminary results show that Tycoon achieves low latency and high fairness while providing incentives for truth-telling on the part of strategic users.

Posted by rohit at 08:56 AM | Comments (0) | TrackBack

November 09, 2004

Great Oneliner From a Schmidt Talk

kwc blog: Forum: Google:

Google allows weirdos to advertise to weirdos (e.g. Monkey hats and Bird poop ads)

Posted by rohit at 10:32 AM | Comments (0) | TrackBack

November 08, 2004

Troutgirl In Da House

We welcome Joyce to CommerceNet Labs; in the coming months, she'll be working with us on projects such as zClassifieds and declassifieds and other fun things to do with ad networks...

Posted by adam at 11:19 AM | Comments (0) | TrackBack

November 01, 2004

zClassifieds

Dan Gillmor wrote in the 10/31/2004 SJ Mercury News:

Google will have all kinds of company in this expanding world of advertising. That will include, I would expect, many of the more traditional media companies that will see a chance to expand their advertising base beyond the equivalent of the blockbuster (expensive) model that now prevails.

The competitors will also include big companies that have already shown an appreciation of Net-based economics. Microsoft, Yahoo, eBay and at least a few others will certainly be among them.

Google will also find competitors, small ones, out at the edges. And some of those will be new entrants that are figuring out ways to create targeted advertising without massively centralized infrastructures. The principles of peer-to-peer file-sharing will come to the ad marketplace, too.

Google is unquestionably positioning itself in a smart way. The critical mass it's creating may even prove unbeatable, or turn into a new kind of monopoly that sucks up an astonishing portion of all advertising dollars into its corporate coffers. (That would be a dangerous dominance if it happened.)

Today, eBay, online classified-ad sites and traditional media are the marketplace of choice for the single-item seller. Ultimately, Google and others could even go after that market.

How many dollars (and euros, yen, pesos, renmimbi, etc.) will there turn out to be in the low-end advertising market? It's a big, big number.

Google may not own it. But it's going to get a share, a large one. I wouldn't touch the company's stock at today's prices, but there's plenty of room for growth in its primary revenue base. Nothing grows to the moon, as the saying goes, but there's a fair amount of sky left.

A month ago we gave the name declassifieds to the concept of decentralizing ads. Now we give the name zClassifieds to an internal CommerceNet project to work on "targeted advertising without massively centralized infrastructures". Will post more as we learn more.

Posted by adam at 03:59 PM | Comments (0) | TrackBack

October 31, 2004

P2P Traffic Measurement

After our decentralized filesharing post we discovered this item from ACM News Service, "Is P2P Dying or Just Hiding?"

High-order bits excerpted from the paper itself:

• In our traces, P2P traffic volume has not dropped since 2003. Our datasets are inconsistent with claims of significant P2P traffic decline.
  
• We present a methodology for identifying P2P traffic originating from several different P2P protocols. Our heuristics exploit common conventions of P2P protocols, such as the packet format.
  
• We illustrate that over the last few years, P2P applications evolved to use arbitrary ports for communication.
  
• We claim that accurate measurements are bound to remain difficult since P2P users promptly switch to new more sophisticated protocols, e.g., BitTorrent.

More bits:

CAIDA monitors capture 44 bytes 2 of each
packet (see section III), which leaves 4 bytes of TCP packets to
be examined (TCP headers are typically 40 bytes for packets that
have no options). While our payload heuristics would be capable
of effectively identifying all P2P packets if the whole payload
was available, this 4-byte payload restriction limits the number
of heuristics that can undoubtedly pinpoint P2P flows. For example,
BitTorrent string “GET /torrents/” requires 15 bytes of
payload for complete matching. Our 4-byte view of “GET ”
could potentially indicate a non-P2P web HTTP request.

The ACM News Service summary...

"Is P2P Dying or Just Hiding?"
CAIDA.org (10/04); Karagiannis, Thomas; Broido, Andre; Brownlee, Nevil

UC Riverside's Thomas Karagiannis, the Cooperative Association for Internet Data Analysis' (CAIDA) Andrew Broido, et al. dispute popular media reports that peer-to-peer (P2P) file-sharing has declined precipitously in the last year, and contend that the reverse is actually the case. The authors attempted to measure P2P traffic at the link level more accurately by gauging traffic of all known popular P2P protocols, reverse engineering the protocols, and labeling distinctive payload strings. The results support the conclusion that 2004 P2P traffic is at least comparable to 2003 levels, while rigid adherence to conventional P2P traffic measurement techniques leads to miscalculations. The percentage of P2P traffic was found to have increased by about 5 percent relative to traffic volume. Furthermore, comparisons between older and current P2P clients revealed that the use of arbitrary port numbers was elective in older clients, while current clients randomize the port number upon installment without the need for user action. Meanwhile, P2P population studies found that the ranks of IPs grew by about 60,000 in the last year, and the number of ASes participating in P2P flows expanded by roughly 70 percent. These findings outline several trends, including evolving tension between P2P users and the entertainment sector; increasing demand for home broadband links; plans to directly induce P2P applications into profitable traffic configurations; and a significant transformation in supply and demand in edge and access networks, provided that P2P traffic maintains its growth and legal entanglements are eliminated.

The full article (pdf) is contains many more bits for the interested reader...

Posted by rohit at 12:10 PM | Comments (0) | TrackBack

October 27, 2004

Plott's Cool Experiment

Science News: Best guess: economists explore betting markets as prediction tools

The research that led to future-predicting markets stems from the 1960s and 1970s, when Vernon Smith and Charles Plott, now of George Mason University and the California Institute of Technology in Pasadena, respectively, began using laboratory experiments to study different market designs. In the early 1980s, Plott and Shyam Sunder, now of Yale University, tested how well markets aggregate information by designing a set of virtual markets in which they carefully controlled what information each trader had.

In one experiment, Plott and Sunder permitted about a dozen study participants to trade a security, telling them only that it was worth one of three possible amounts--say, $1, $3, or $8--depending on which number was picked by chance. Plott and Sunder then gave two of the participants inside information by telling them which amount had been selected. Traders couldn't communicate with each other; they could only buy and sell on the market.

"The question was, Would the market as a whole learn what the informed people knew?" Plott says. "It turned out that it would happen lightning fast and very accurately. Everyone would watch the movements of the market price, and within seconds, everyone was acting as if they were insiders."

In another experiment, Plott and Sunder gave the inside traders less-complete information. For instance, if the outcome of the random pick were $3, they would tell some traders that it was not $1, and others that it was not $8. In these cases, the market sometimes failed to figure out the true value of the security.

However, if Plott and Sunder created separate securities for each of the three possible outcomes of the random pick instead of using one security worth three possible amounts, the market in which some traders had incomplete tips succeeded in aggregating the information.

The studies established that, at least in these simple cases, markets indeed can pull together strands of information and that different setups affect how well they do so.

This type of experiment gave researchers a "wind tunnel" in which to test different market designs, says John Ledyard, a Cal-tech economist who chairs the board of Net Exchange. "With experiments, we're starting to zero in on what really works," he says.

Posted by rohit at 08:39 AM | Comments (0) | TrackBack

October 26, 2004

Negroponte on p2p ecash

Future Of Transactions > Peer-To-Peer Payoff > October 18, 2004" href="http://www.informationweek.com/story/showArticle.jhtml?articleID=49901099">InformationWeek > Future Of Transactions > Peer-To-Peer Payoff > October 18, 2004

Some people, including myself, believe the next step is for some of those bits to have value. That is to say, consider a string of bits to be like a virtual cow or shell. In order to distinguish these bits (like telling the difference between a beautiful seashell and a piece of coal), they would need an agreed identity. To avoid forgery, they would need a unique and secure ID. And to stop multiple spending of the same bits, there would need to be a clearing process or a means to reveal the identity of anybody who tries to double-spend. All of these requirements are easily achieved in both traceable and anonymous systems of E-cash. In these cases, the money does move. The bits are money. The more you have, the richer you are. This is the future, though maybe only in part.

A parallel and more intriguing form of trade in the future will be barter. Swapping is a very attractive form of exchange because each party uses a devalued currency, in some cases one that would otherwise be wasted. Many of us are too embarrassed to run yard sales or too lazy to suffer the inconvenience and indignity of eBay. But imagine if you weren't. The unused things in your basement can be converted into something you need or want. Likewise, the person with whom you're swapping is giving something of value to you which is less so to them. With minimal computation, three-way, four-way, and n-way swaps can emerge, thereby removing the need for any common currency.

Swapping is extended easily to baby-sitting for a ride to New York, a mansion for a two-hundred-foot yacht, or leftover food for a good laugh. In some cases, people will swap for monetary or nonmonetary currencies. Without question, we'll see new forms of market-making and auctions. But the most stunning change will be peer-to-peer, and peer-to-peer-to-peer- ... transaction of goods and services. If you fish and want your teeth cleaned, you need to find a dentist who needs fish, which is so unlikely that money works much better. But if a chauffeur wanted fish and the dentist wanted a driver, the loop is closed. While this is nearly impossible to do in the physical world, it's trivial in cyberspace. Add the fact that some goods and services themselves can be in digital form, and it gets easier and more likely. An interesting side benefit will be the value of one's reputation for delivering on your promises -- thus, identities will have real value and not be something to hide.

The point can be generalized beyond money. Peer-to-peer is a much deeper concept than we understand today. We're limited by assumptions rooted in and derived from the physical world. Information technology over the next 25 years will change those limits through force of new habits. Let me cite just one: I think nothing of moving millions of bits from one laptop to another (inches away) by using the Internet and transferring those bits through a server 10,000 miles away. Imagine telling that to somebody just 25 years ago.

Nicholas Negroponte is the founding chairman of MIT's Media Laboratory and the author of the seminal work on the digital revolution, "Being Digital" (Knopf, 1995).

Posted by rohit at 03:35 PM | Comments (0) | TrackBack

October 21, 2004

Decentralization Defined

We've spent a little time working on the Decentralization page in the CommerceNet Labs wiki. Here's a snapshot of what we have so far:

Decentralization in Commerce means the freedom to do commerce the way you want, rather than the way your software wants.

We believe that to build software that works the way society works, that software design must reflect the principles of decentralization.

An agency is an organization with a single trust boundary. One way to think about decentralization is that it allows multiple
agencies to have different values for a variable.

See also:

• Wikipedia - Decentralization
  
• The Now Economy - Decentralization

Update. Allan Schiffman notes that Webster's has a fine definition of decentralization as well: "the dispersion or distribution of functions and powers". I'll add that to the wiki as well...

Posted by adam at 03:55 PM | Comments (2) | TrackBack

October 19, 2004

zSearch and CommerceNet's Neighborhood

I want SupplyFX, Webify, and Bonsai Development to be in CommerceNet's Neighborhood. I also want Rob Rodin, Allan Schiffman, Kevin Hughes, and Marty Tenenbaum in CommerceNet's Neighborhood.

And by linking to those pages like I just did, I just did. Such is zSearch.

Posted by adam at 01:21 PM | Comments (0) | TrackBack

October 18, 2004

"Fluffy Bunny" is a WinSock-puppet

We love Google's new Desktop Search. We've been arguing about something like this for a year or more. The idea of searching everything you've seen — not just your hard drive, but everything hyperlinked to it (such as your surfing history) — is so intriguing we've built something similar for ourselves.

We've modified Nutch to search not just CommerceNet's website, weblog, and wiki, but also everything we link to. Go ahead and try our index of CommerceNet's Neighborhood. If you query for Nutch you won't just see pages from our sites, but Nutch's home page and even an application of it at CreativeCommons...

We're also exploring a bunch of other ideas that GDS and other desktop indexing projects like StuffIveSeen haven't tackled yet. Foremost is ranking -- just like the AltaVista engine that Google itself dethroned, GDS doesn't have anything like a PageRank for the gigabytes of information on your disk. Like many researchers, we suspect that a users's social network is the key to discerning which hits are likely to be most useful. After all, if the Web is drowning in infoglut on any given query term, a user should be such an expert on the terms of his or her art that there ought to be even more hits to rank on localhost. One cure may be collaborative filtering with your friends...

Secondary aspects of the problem include tackling the fact that many of us have multiple computers and identities on the Internet, so we'd need networks of personal search engines. Or that a local-proxy-server approach might be better at capturing the ''dynamics'' of our interaction (how often we re-read the same email over IMAP, say).

But rather than rattling off a longer list of half-baked hypotheses, I'd like to cite GDS for at least one idea that never occurred to us: integrating it seamlessly with the public site. Sure, we thought AdWords-like ads were the key to a better revenue model for the Fisher category of PersonalWeb products.

No, what's cool is that Google's ordinary results pages from the public website automatically include hits from your hard drive. How'd they ''do'' that?! Read on...

CommerceNet Labs Wiki : FluffyBunnyBurrowsIntoWinSock

...we found that Google Desktop Server actually hooks into Windows' TCP/IP stack to directly modify incoming traffic from Google's websites to splice its local results in. Once you install GDS, there's a bit of Google's code running inside every Windows application that talks to the Internet.

It's done using a long-established hook in WinSock2, its Layered Transport Service Provider Interface (SPI)...

The Winsock LSP is mostly only used by spyware and censorware; it's a surprise to see a positive use for it. Spyware detectors like HijackThis consequently detect it.

[An aside: why is Rifkin's GLAT posting more relevant on the query "rifkin fisher" than Rifkin's actual Fisher posting? I think it's Battelle's fault, for increasing the GLAT's PageRank! :-]

Posted by rohit at 09:03 PM | Comments (0) | TrackBack

October 03, 2004

Blogs Are Decentralization Incarnate

From Biz Stone's excellent article, The Wisdom of Blogs:

Bloggers are a wise crowd.

1. Diversity of opinion - That's a no-brainer. Bloggers publish hundreds of thousands of posts daily, each one charged with its author's unique opinion.
   
2. Independence of members - Except for your friends saying "You've got to blog about that!" bloggers are not controlled by anyone else.
   
3. Decentralization - There is no central authority in the blogosphere; publish your blog anywhere you want with any tool you want.
   
4. A method for aggregating opinions - Blog feeds make aggregation a snap and there is no shortage of services that take advantage of that fact.

The article goes on to talk about how MIT Media Lab project Blogdex (one of the longest-operating and most-visited opinion aggregators) is like a hive mind of the blogosphere, collectively creating a modern Oracle with no single opinion about anything. Excellent.

Posted by adam at 03:38 PM | Comments (0) | TrackBack

October 02, 2004

Decentralized Filesharing Is Huge

Cachelogic Research paints an interesting picture of decentralized filesharing.

The most astonishing item is that global Internet traffic analysis in June 2004 revealed that in the United States peer-to-peer represents roughly two-thirds of traffic volumes, and in Asia peer-to-peer represents more than four-fifths of traffic volumes. By comparison, HTTP is less than a tenth of the traffic in Asia and less than a sixth of the traffic in the United States. CacheLogic calls peer-to-peer the killer application for broadband with a global reach and a global user base.

Perusing the architectures and protocols section of CacheLogic's site we find a table comparing the characteristics of web traffic (HTTP) with those of common peer-to-peer protocols. They point out that first generation p2p systems were centralized like Napster; second generation p2p systems were decentralized like Gnutella; and now

The third generation architecture is a hybrid of the first two, combining the efficiency and resilience of a centralized network with the stealth characteristics of distributed/decentralised network. This hybrid architecture deploys a hierarchical structure by establishing a backbone network of SuperNodes (or UltraPeers) that take on the characteristics of a central index server. When a client logs on to the network, it makes a direct connection to a single SuperNode which gathers and stores information about peer and content available for sharing.

Recent developments in peer-to-peer include dynamic port selection and bidirectional streaming of download traffic in the most popular peer-to-peer applications in 2004, BitTorrent (more useful thanks to many available BitTorrent clients and DV Guide) and eDonkey (and eMule). BitTorrent is by traffic the most popular peer-to-peer application:

BitTorrent's dominance is likely to be attributed to two factors: the rise in popularity of downloading television programmes, movies and software; and the size of these files - a MP3 maybe 3-5Mb while a BitTorrent often sees files in excess of 500Mb being shared across the Peer-to-Peer network.

The high usage of eDonkey in Europe can be attributed to the fact that the eDonkey interface is available in a number of different languages - French, German, Spanish, etc.

So even though the hype machine has stopped pumping p2p, the quieter revolution of the last few years has shown that peer-to-peer traffic has steadily grown to a majority of the Internet traffic worldwide.

Posted by adam at 01:37 PM | Comments (0) | TrackBack

September 24, 2004

Wikipedia Turns a Million

Joi Ito points out that Wikipedia just passed one million articles: "Wikipedia is in more than 100 languages with 14 currently having over 10,000 articles... At the current rate of growth, Wikipedia will double in size again by next spring." Wikipedia itself points to the power of a massive, decentralized content authoring effort.

Ross Mayfield adds, "To put this in perspective, if each article took 1 person week to produce, getting the next million would take 40,000 full-time equivalent resources to get it done in the same amount of predicted time. Co-incidentally Wikipedia has about the same amount of registered users, but they have day jobs too."

Even more impressive, "Wikipedia is a volunteer effort supported by the non-profit Wikimedia Foundation." When I clicked on their fundraising effort, I discovered that they're looking to raise fifty thousand dollars -- a tiny amount by corporate standards. It speaks to the fact that a massive, decentralized effort need not cost a tremendous amount to have a huge impact.

Posted by adam at 04:20 AM | Comments (0) | TrackBack

September 16, 2004

Declassifieds

It's taken a few weeks for this to sink in: John Battelle's post on Sell Side Advertisting (inspired by Ross Mayfield's post on Cost Per Influence).

Ross wrote: "An important facet of this format is the amount of user choice. Users decide what feeds to subscribe to and ads to block. Bloggers should be able to choose what ads to both host and pass through."

John's addition:

Instead of advertisers buying either PPC networks or specific publishers/sites, they simply release their ads to the net, perhaps on specified servers where they can easily be found, or on their own sites, and/or through seed buys on one or two exemplar sites. These ads are tagged with information supplied by the advertiser, for example, who they are attempting to reach, what kind of environments they want to be in (and environments they expressly forbid, like porn sites or affiliate sites), and how much money they are willing to spend on the ad.

Once the ads are let loose, here's the cool catch - ANYONE who sees those ads can cut and paste them, just like a link, into their own sites (providing their sites conform to the guidelines the ad explicates in its tags). The ads track their own progress, and through feeds they "talk" to their "owner" - the advertiser (or their agent/agency). These feeds report back on who has pasted the ad into what sites, how many clicks that publisher has delivered, and how much juice is left in the ad's bank account. The ad propagates until it runs out of money, then it... disappears! If the ad is working, the advertiser can fill up the tank with more money and let it ride.

This concept of decentralizing ads (instead of "classified ads", they're "declassifieds") empowers multiple agencies -- not just advertisers and ad networks but publishers, too -- to determine which ads propogate.

Taking this a step further to create a truly decentralized advertising network requires asking oneself the question of who is empowered to determine what goes in the square inch of real estate used by web browsers and feed readers to display ads? Not just advertisers, ad networks, and publishers -- but the software writers and the actual people reading the web and feeds as well.

Posted by adam at 01:52 AM | Comments (0) | TrackBack

September 13, 2004

Feedmesh: Decentralized Web Notifications

Jeremy Zawodny talks about the inevitability of search results as RSS that can be subscribed to, quoting Tim Bray:

They’ve also done something way cool with their Google appliance; one of the bright geeks there has set up a thing where you can subscribe to a search and get an RSS feed. Well, duh. Anyone could fix up one of those using the Google API, I wonder why Google isn’t supporting this already?

This in turn reminds us of Jeff Barr's real-time headline view (more thoughts), which he talked about this weekend at Foo Camp, also attended by Sam Ruby, who talked about FeedMesh, a working group to establish a "peering network" for decentralized web(site|log) update notifications and content distribution. This is the start of something potentially wonderful...

Posted by adam at 04:23 AM | Comments (1) | TrackBack

September 09, 2004

Plott on Decentralized Pasture Management

I came across this preparing to visit Prof. Plott at my alma mater next week... here's some snippets from one of his very latest papers; he's been interested in information aggregation of late.

DECENTRALIZED MANAGEMENT OF COMMON PROPERTY RESOURCES:

For several centuries, villages in the Italian Alps employed a
special system for managing the common properties. The
experiments and analysis of this paper are motivated by an attempt
to understand why that particular system might have been successful
in comparisons with other systems that have similar institutional
features. The heart of the system was a special monitoring device
that allowed individual users to inspect other users at their own cost
and impose a predetermined sanction (a fine) when a free rider was
discovered. The fine was paid to the user who found a violator.
In addition to the replication of the results of others, the paper
finds three classes of results. First, in comparison with a classical
model of identical, selfish agents, the data can best be captured by a
model with heterogeneous and other-regarding preferences where
altruism and especially spite play an important role. Second, the
model with heterogeneous agents suggests that the success of the
institution is related to its ability to turn these individual differences
to socially useful purposes. Third, the model also explains important
paradoxes that can be found in the existing literature.

...

The success of the Carte di Regola system appears to be related to its ability to use the
heterogeneity of preferences to socially advantageous ends. The system also appears to have
a type of robustness against institutional and parameter changes. Notice first that the Carte di
Regola channels attitudes that might normally be considered as socially dysfunctional, such as
spiteful preferences, into socially useful purposes. People with spiteful preferences choose
to monitor and sanction at a monetary loss. But when their preferences are considered as
part of system efficiency, they are the ones who can perform the function most efficiently
and are channeled into the activity for which they have a comparative advantage.

One might think that the Carte di Regola is similar to a system of vigilantes but there are
important differences. In the model, spiteful people do not care who they hurt, they just
enjoy hurting others, so it is important to direct and constrain them. The Carte di Regola
directs them by reserving the judgment of guilt for the courts, as opposed to the vigilantes,
who would be happy to judge anyone guilty. The court convicts a person only when the guilt
is consistent with social purposes. The magnitude of punishment is also reserved for the
courts in the Carte di Regola system, while in a vigilante system the inspector is allowed to
judge and determine punishment. So, the Carte di Regola constrains what the spiteful can do
to the guilty. Thus, there are important differences (OWG, 1992).

The Carte di Regola also channels arbitrary or random behavior toward useful ends. Such
behavior might ordinarily be regarded as dysfunctional from the point of view of economic
efficiency. Mistaken inspections or impulsively random inspections are costly to the
inspector and thus involve efficiency losses, but the fact that inspections take place has
consequences for those who are excessive users of the common pool resource by increasing
the likelihood that a sanction is imposed. Thus random inspection behavior that would
appear irrational helps preserve the commons.

Posted by rohit at 01:40 PM | Comments (0) | TrackBack

September 08, 2004

Decentralized Authoring Can Be Self-Healing

BoingBoing in Wikipedia proves its amazing self-healing powers pointed us to The Isuzu Experiment, which goes like this:

Joi Ito points to an ongoing discussion regarding the authority of wikipedia as a source of information and knowledge. The discussion was prompted by an article in the Syracuse Post-Standard that suggests, in part, that wikipedia “take[s] the idea of open source one step too far” by allowing the user to make corrections.

The article has been correctly ridiculed by many, including Mike at Techdirt. In a later posting, he suggests an experiment: why not go to a certain page, insert something provably incorrect, and see how long it lasts.

No matter which side of the debate you find yourself on, this sounds like an interesting experiment. So, I have made not one, but 13 changes to the wikipedia site. I will leave them there for a bit (probably two weeks) to see how quickly they get cleaned up. I’ll report the results here, and repair any damage I’ve done after the period is complete. My hypothesis is that most of the errors will remain intact.

Does that invalidate Wikipedia? Certainly not! If anything, the general correctness and extent of Wikipedia is a tribute to humankind. It suggests the Kropotkin may be right: that the “survival of the fittest” requires that the fittest cooperate. It means that there are very few Vandals like me who are interfering with its mission.

Terrible experiment, but it demonstrates how decentralized authoring can be self-healing. Wrote Phil in the Boing Boing summary,

Remember Al Fasoldt, the journalist who disparaged Wikipedia? He was challenged by a Techdirt writer to change an item and see if his change was found. While Fasoldt dismissed the idea, Alex Halavais thought it was an interesting idea. He made 13 changes to 13 different Wikipedia pages, ranging from obvious to subtle. He figured he'd give them a couple of weeks and then fix the ones that weren't caught. Every single change was found and changed within hours.

It's a terrible idea to vandalize Wikipedia like this. But it's a wonderful thing how quickly self-healing Wikipedia is to such attacks.

Posted by adam at 12:35 PM | Comments (0) | TrackBack

September 04, 2004

Decentralizing Akamai

A step in the direction of "decentralizing Akamai" -- but still uses the "centralized DNS" to create an interesting distributed web caching network -- is the Coral Content Distribution Network.

Mike Dierken talks about the Coral CDN by quoting Gordon Mohr quoting Michael J. Freedman's post to the p2p-hackers list:

To take advantage of CoralCDN, a content publisher, user, or some third party posting to a high-traffic portal, simply appends .nyud.net:8090 to the hostname in a URL. For example:

http://news.google.com/ --> http://news.google.com.nyud.net:8090/

Through DNS redirection, oblivious clients with unmodified web browsers are transparently redirected to nearby Coral web caches. These caches cooperate to transfer data from nearby peers whenever possible, minimizing the load on the origin web server and possibly reducing client latency.

Mike writes: "DNS/HTTP based P2P -- Wicked cool and finally a REST based scalable p2p network. I wonder how I could use that at Amazon..."

Rohit asks if this technique could help Slashdot alleviate "The Slashdot Effect." According to the Slashdot post on Coral, apparently not.

Posted by adam at 06:12 PM | Comments (0) | TrackBack

September 01, 2004

Markets in Deep Space

We've recently written a new position paper titled
Agoric Architectural Styless for Decentralized Space Exploration. It's been submitted to the
2004 Workshop on Self-Managing Systems (WOSS'04) to be held at FSE-12 in Newport Beach. it was originally based on some notices of intent (NOIs) for a NASA Broad Agency Announcement (BAA) on innovative Human & Robotic Technologies (H&RT) for future space exploration missions.

Abstract: This position paper discusses an architectural approach to managing decentralized space exploration missions. Developing control applications in this domain is complicated by more than just the challenging computing and communication constraints of space-based mission elements; future exploration missions will depend on ad-hoc cooperation between independent space agencies’ elements. Currently, the frontier of interoperability is providing communication relays, as shown in by recent Mars missions, where NASA rovers relayed data via ESA satellites.

Future mission planning envisions more extensive autonomy and integration. Examples include: taking advantage of excess storage capacity at another node, multicasting messages along several paths through deep space, or even scheduling concurrent observations of an object using several instruments at different locations. An architectural style for developing mission control applications that does not depend on positive ground control from Earth could provide (a) increased margins for space-based computing systems, (b) increased reusability by an effective build-it-for-autonomy-first strategy, and (c) avoid the single-point of failure bias in standard distributed system design approaches.

In particular, we propose combining an architectural style for decentralized applications based on the Web (ARRESTED) with agoric computing to apply market discipline for allocating resources dynamically among coalitions of mission elements in space. Similar approaches may have applicability in other domains, such as crisis management or battle management.

Posted by rohit at 11:25 AM | Comments (0) | TrackBack

August 31, 2004

How can society build software that lasts decades?

The following are only a few lines of excerpts from an extremely important argument about the "culture of design" surrounding software. It is a critical aspect of any effort to design "software that works the way society works," to cite the credo of the decentralized software architecture crowd.

It may have an important impact on how CommerceNet Labs refines its own mission, too...

Software That Lasts 200 Years

In many human endeavors, we create infrastructure to support our lives which we then rely upon for a long period of time...

By contrast, software has historically been built assuming that it will be replaced in the near future (remember the Y2K problem). Most developers observe the constant upgrading and replacement of software written before them and follow in those footsteps with their creations...

In accounting, common depreciation terms for software are 3 to 5 years; 10 at most. Contrast this to residential rental property which is depreciated over 27.5 years and water mains and brick walls which are depreciated over 60 years or more... I can go to city hall and find out the details of ownership relating to my house going back to when it was built in the late 1800's.

[Dan Bricklin] will call this software that forms a basis on which society and individuals build and run their lives "Societal Infrastructure Software". This is the software that keeps our societal records, controls and monitors our physical infrastructure (from traffic lights to generating plants), and directly provides necessary non-physical aspects of society such as connectivity.

What is needed is some hybrid combination of custom and prepackaged development that better meets the requirements of societal infrastructure software.

How should such development look? What is the "ecosystem" of entities that are needed to support it? Here are some thoughts:

* Funding for initial development should come from the users...

* The projects need to be viewed as for more than one customer... Funding or cost-sharing "cooperatives" need to exist.

* The requirements for the project must be set by the users, not the developers. The long-term aspects of the life of the results must be very explicit...

* ... Impediments such as intellectual property restrictions and "digital rights management" chokepoints must be avoided...

* The actual development may be done by business entities which are built around implementing such projects, and not around long-term upgrade revenue...

* The attributes of open source software need to be exploited. This includes the transparency of the source code and the availability for modification and customization... The availability of the source code, as well as the multi-customer targeting and other aspects, enables a market for the various services needed for support, maintenance, and training as well as connected and adjunct products.

* The development may be done in-house if that is appropriate, but in many cases there are legal advantages as well as structural for using independent entities..

* Unlike much of the discussion about open source, serendipitous volunteer labor must not be a major required element. A very purposeful ecosystem of workers, doing their normal scheduled work, needs to be established to ensure quality, compatibility, modifications, testing, security, etc... The health of the applications being performed by the software must not be dependent upon the hope that someone will be interested in it; like garbage collecting, sewer cleaning, and probate court judging, people must be paid.

The ecosystem of software development this envisions is different than that most common today. The details must be worked out. Certain entities that do not now exist need to be bootstrapped and perhaps subsidized. There must be a complete ecosystem, and as many aspects of a market economy as possible must be present.

Posted by rohit at 10:22 AM | Comments (0) | TrackBack

August 17, 2004

Agoric Systems Reading

This is all very old school - 1988! - but it's always refresing to take another look at the basics. While a survey paper from U. Florida says that the concept can be traced to Ivan Sutherland auctioning timesharing slots in 1968, the likely origin of the term "agoric systems" (from the greek agora, or market) is Mark S. Miller and K. Eric Drexler's chapter in The Ecology of Computation.

Like all systems involving goals, resources, and actions, computation can be viewed in economic terms. Computer science has moved from centralized toward increasingly decentralized models of control and action; use of market mechanisms would be a natural extension of this development. The ability of trade and price mechanisms to combine local decisions by diverse parties into globally effective behavior suggests their value for organizing computation in large systems.

This paper examines markets as a model for computation and proposes a framework-agoric systems-for applying the power of market mechanisms to the software domain. It then explores the consequences of this model at a variety of levels. Initial market strategies are outlined which, if used by objects locally, lead to distributed resource allocation algorithms that encourage adaptive modification based on local knowledge. If used as the basis for large, distributed systems, open to the human market, agoric systems can serve as a software publishing and distribution marketplace providing strong incentives for the development of reusable software components. It is argued that such a system should give rise to increasingly intelligent behavior as an emergent property of interactions among software entities and people.

Posted by rohit at 10:03 AM | Comments (0) | TrackBack

August 16, 2004

Best Practices Trump Decentralization

Jon Udell wrote in Infoworld this week:

In the end, scalability isn’t an inherent property of programming languages, application servers, or even databases. It arises from the artful combination of ingredients into an effective solution. There’s no single recipe. No matter how mighty your database, for example, it can become a bottleneck when used inappropriately.

It’s tempting to conclude that the decentralized, loosely coupled Web architecture is intrinsically scalable.

Not so. We’ve simply learned — and are still learning — how to mix those ingredients properly. Formats and protocols that people can read and write enhance scalability along the human axis. Caching and load-balancing techniques help us with bandwidth and availability.

But some kinds of problems will always require a different mix of ingredients. Microsoft has consolidated its internal business applications, for example, onto a single instance of SAP. In this case, the successful architecture is centralized and tightly coupled.

For any technology, the statement “X doesn’t scale” is a myth. The reality is that there are ways X can be made to scale and ways to screw up trying. Understanding the possibilities and avoiding the pitfalls requires experience that doesn’t (yet) come in a box.

Decentralization is not a hammer that can hit every nail. Experience is still the mother of good judgment.

Posted by adam at 02:38 PM | Comments (0) | TrackBack

August 13, 2004

Amazon CTO: "We've Just Scratched The Surface"

Information Week:

Sometime in the next few weeks, Amazon.com is scheduled to release Amazon Web Services 4.0, the next version of the electronic retailer's toolset for developing applications that tie into its Web site. It's the next step in Amazon's strategy is to create a "programmable" Web site. Over the past two years, 50,000 developers have downloaded earlier versions of Amazon Web Services. InformationWeek senior editor-at-large John Foley asked Al Vermeulen (pictured right), Amazon's chief technology officer, about how the model works.

The whole interview is great. I love this quote from Al Vermeulen:

There's a host of [services], probably 15 or 20, maybe more, and they range from components like personalization and search applications, to fulfillment applications, supply-chain services, and on and on. The basis is, let's think about everything we do at Amazon.com and about how we break it up into individual pieces, smaller pieces. What we try to do is break apart a piece of the business. From a technology point of view, that becomes a service. From an organizational point of view, it becomes an autonomous team with their own mission, and then we work on defining the interface to get to that service. We try to solidify that interface and make it permanent and robust. It's kind of a bottom-up, decentralized way of building your technology, as opposed to a top-down way where you try to make all the technology look like one piece.

It's very exciting to see where Amazon is going with this...

Posted by adam at 03:06 PM | Comments (0) | TrackBack

August 10, 2004

More on that Toyota fire

Toyota Manages Quick Recovery from Fire

Wall Street Journal
8 May 1997
Page A-1
by Valerie Reitman staff reporter

Toyota Motor Shows Its Mettle
After Fire Destroys Parts Plant

KARIYA, Japan -- No one knows what sparked the fire that roared through Aisin Seiki Co.'s Factory No. 1 here before dawn on Saturday, Feb. 1, leveling the huge auto-parts plant. But one thing is clear: The crisis-control efforts that followed it dramatically illustrate one reason Toyota Motor Corp. is among the world's most admired and feared manufacturers.

Wall Street Journal
8 May 1997
Page A-1
by Valerie Reitman staff reporter

Toyota Motor Shows Its Mettle
After Fire Destroys Parts Plant

KARIYA, Japan -- No one knows what sparked the fire that roared through Aisin Seiki Co.'s Factory No. 1 here before dawn on Saturday, Feb. 1, leveling the huge auto-parts plant. But one thing is clear: The crisis-control efforts that followed it dramatically illustrate one reason Toyota Motor Corp. is among the world's most admired and feared manufacturers.

The fire incinerated the main source of a crucial brake valve that Toyota buys from Aisin and uses in most of its cars. Most Toyota plants kept only a four-hour supply of the $5 valve; without it, Toyota had to shut down its 20 auto plants in Japan, which build 14,000 cars a day. Some experts thought Toyota couldn't recover for weeks.

But five days after the fire, its car factories started up again. Only recently have Toyota and its suppliers revealed how they did it.

Model of Cooperation

The secret lay in Toyota's close-knit family of parts suppliers. In the corporate equivalent of an Amish barn-raising, suppliers and local companies rushed to the rescue. Within hours, they had begun taking blueprints for the valve, improvising tooling systems and setting up makeshift production lines.

By the following Thursday, the 36 suppliers, aided by more than 150 other subcontractors, had nearly 50 separate lines producing small batches of the brake valve. In one case, a sewing-machine maker that had never made car parts spent about 500 man-hours refitting a milling machine to make just 40 valves a day.

"Toyota's quick recovery," says Yoshio Yunokawa, general manager of Toyoda Machine Works Ltd., a Toyota-group maker of machine tools and steering systems, "is attributable to the power of the group, which handled it without thinking about money or business contracts."

That is a common approach in Japan's keiretsu, the almost tribal groups of companies that supply and support behemoths such as Toyota. Japanese auto makers' blood pacts with their suppliers largely explain how they can slash their costs to the bone and stay globally competitive.

No Foreign Help Required

And the fealty the parts makers paid to Toyota during its crisis helps indicate why Japan's auto companies return the loyalty -- often to the detriment of U.S. and other foreign parts makers seeking market share here. Toyota and Aisin didn't bother to approach any foreign companies during the crisis, a Toyota spokesman says, because "there were no foreign suppliers in a position to help us."

Aisin (pronounced "eye-sheen") is an archetypal supplier in Toyota's group. Founded during World War II to make aircraft engines, it is based in Kariya, an industrial warren occupied by many other big Toyota suppliers. Toyota holds 23% of Aisin's stock. Aisin's president is Kanshiro Toyoda, scion of the Toyoda family that founded the auto maker.

Long a supplier of engine and brake parts, 80% of which it sells to Toyota, Aisin has won almost all of Toyota's contracts for brake-fluid-proportioning valves -- "P-valves" in industry parlance. The fist-sized valves control pressure on rear brakes and help prevent skidding.

For most parts, Toyota has at least two suppliers. But over the years, it turned to Aisin to produce all but 1% of its P-valves because of Aisin's high quality and low cost. The supplier shipped parts to Toyota plants under a just-in-time inventory system: several times a day, just enough valves for a few hours of production.

Calculated Risk

Depending on a single source and holding essentially no inventory is a calculated risk, concedes Kiyoshi Kinoshita, Toyota's general manager of production control. But it also is what keeps Toyota's production lean; Aisin gets major economies of scale that it passes on to Toyota in lower prices.

Toyota acknowledges that it didn't figure in the risk of fire. Aisin executives speculate that sparks from a broken drill bit may have ignited wooden platforms. Just after 4 that morning, flames swept through an air duct and ignited the roof. Graveyard-shift workers escaped as 36 fire engines arrived, mostly from nearby Toyota-group companies.

Even as the fire burned, Aisin officials organized a committee to assess the damage, notify customers and labor unions and, following Japanese custom, visit neighbors to apologize. A subcommittee ordered 320 cellular phones, 230 extra phone lines and several dozen sleeping bags for executives who were expected to live at headquarters in the coming days.

At 8 a.m., Aisin asked Toyota to help. Kosuke Ikebuchi, a Toyota senior managing director, was tracked down at a golf-course clubhouse; he left his wife there and rushed to Toyota headquarters to help set up a "war room" to direct the damage-control operation. Toyota quickly sent more than 400 engineers to Aisin. In reacting to such a crisis, Mr. Kinoshita says, "we're like the U.S. military."

When the last embers died just before 9 a.m., the damage at Factory No. 1 began to grow clear -- along with an apparent Achilles' heel in Toyota's lean corporate physique. Most of the factory's 506 highly specialized machines, which make other brake parts as well as P-valves, were charred and useless. Toyota estimated that more than two weeks would be needed just to restore a few milling machines to partial production, and six months to order new machines. That was too long: Auto plants were on overdrive to meet strong domestic demand and serve the brisk-selling U.S. market.

Moreover, a Toyota shutdown would damage local economies. Firms supplying the 20,000 parts in the average Toyota, along with hundreds of businesses such as utilities and trucking companies, would be hurt without Toyota orders. Each day Toyota is down, a state agency calculates, cuts Japan's annual industrial output by 0.1 percentage point.

Prospects for a quick comeback seemed to dim as the Saturday wore on. Toyota production officials were dismayed to learn they needed 200 P-valve variations. And chances that anyone else would quickly take up production looked distant: The valves have many complex tapered orifices that require highly customized jigs and drills. The production department told Toyota President Hiroshi Okuda, who was in the Middle East, that they would close most plants from Monday until at least Thursday -- the longest suspension in company history.

Hectic Scene

On Saturday afternoon, Toyota and Aisin summoned officials from some of the major parts suppliers to a second war room, at Aisin headquarters. It quickly became a hectic scene, with officials shouting out for copies of the blueprints of different P-valves while Toyota executives divvied up valve-making assignments, recalls Osamu Natsume, sales division head at Toyoda Machine Works. "It was chaos," he says.

Then the parts makers had to assemble the tools, dies, drills and other fixtures for machining systems that would normally take several months to perfect. "We had to work, no matter how hard," says Tetsuro Yamakage, production-engineering manager at Toyoda Machine. He immediately raced to find the 30 kinds of cutters, knives and special drills needed to make the valves.

But there still weren't enough suppliers. So Toyota purchasing officials called more parts makers to a Sunday-afternoon meeting. These officials, like those that had met on Saturday, were like family-people who work closely with Toyota from the start of a car's design. "It was crucial because we knew each other, we knew the face of the people," Mr. Ikebuchi says.

One familiar face was Masakazu Ishikawa, a former Toyota manager whose division had designed Toyota P-valves and who now is executive vice president of Somic Ishikawa Inc., a supplier of brake parts and suspension ball joints. Mr. Ishikawa called Somic's top production engineers from his cellular phone on the two-hour ride home from Aisin and asked them to meet at the office at 8 p.m. Sunday. They stayed there until after midnight to plot strategy: They would farm out some of their current factory work to free up machines to make the Toyota parts.

A New Undertaking

At 6 a.m. Monday, Somic's four designers began an eight-stage design process. "We'd literally never done anything like this before," says Isoo Suzuki, production-engineering director. Staying up 40 hours, Mr. Suzuki and his engineers designed jigs. Then they called in some chits from Somic's chain of suppliers. For example, Somic got a machine-tool maker, Meiko Machinery Co., to turn down other orders and put 30 workers on round-the-clock shifts to make the jigs it needed.

Somic drafted technical and administrative staffers to help man the machines. On Feb. 6, right on schedule, it delivered its first P-valves to Toyota.

So many suppliers were rushing to please Toyota that they sparked an unofficial race. Taiho Kogyo Co. would have been first, says Nobuo Fukuma, a vice president of the bearing maker, if it hadn't had to search nationwide for special machine tools. Taiho was forced to alter imported tools, he complains, because "the toolmakers didn't give us priority" and shipped to bigger parts makers instead. Although Taiho's first batch of 500 P-valves was ready on Thursday, less than a day behind two bigger Toyota affiliates, Mr. Fukuma was despondent. "We had wanted to be first," he says.

Adapting a Machine

Others were delighted just to have been called on. Brother Industries Ltd., which usually makes things like sewing and fax machines, got calls Sunday night from Aisin and Toyota. A sense of panic at Toyota had inspired it to go far afield, and Toyota knew Brother had a small machine-tools unit.

Brother cobbled together a P-valve production line by adapting a computerized milling machine that usually makes sewing-machine and typewriter parts. Each valve took Brother 10 minutes to make and five minutes to inspect -- hardly a cost-effective use of its resources. The firm helped out, says Yoshihiko Tsuzuki, a general manager, "because it was an emergency."

Early in the week after the fire, even Toyota's Mr. Ikebuchi had doubts about the goal of resuming production in all plants by Friday. But the supplier group came through. Trucks bearing the first 1,000 usable P-valves rolled in late Wednesday. On Thursday, 3,000 more arrived, and on Friday, 5,000. Slowly, Toyota's assembly lines started up again.

All told, Toyota lost production of 72,000 vehicles. But with overtime and extra shifts, Toyota officials say, it has already nearly recouped the lost output.

The Right System

The fire and its aftermath have left Toyota executives convinced that they have the right balance of efficiency and risk. "Many people say you might need to scatter production to different suppliers and plants, but then you have to think of the costs" of setting up expensive milling machines at each site, Mr. Ikebuchi says. "We relearned that our system works."

In fact, the fire may have made the system even more efficient. Nisshin Kogyo Co., which was making the other 1% of Toyota's P-valves, says that during the crisis it raised production efficiency 30% by speeding up production.

Mr. Kinoshita says the fire spurred Toyota to begin an effort to trim the number of its parts variations, a project that should eventually cut costs. And sole-source suppliers are moving quickly to build fail-safe mechanisms. Somic, which makes all of Toyota's steering linkages, is revamping its system so it can easily shift to another site if disaster strikes.

Suppliers never asked Toyota or Aisin what they would be paid for rushing out the valves, says Somic's Mr. Ishikawa. "We trusted them."

Indeed, as the first valves arrived at Toyota factories, Aisin told the suppliers it would pay for everything, from drills and overtime pay to lost revenue and depreciation. And Toyota promised the suppliers a bonus totaling about $100 million "as a token of our appreciation," says Mr. Okuda, its president. He adds that the auto maker will certainly remember the companies that pitched in during its crisis.

Posted by admin at 09:45 AM | Comments (0) | TrackBack

August 09, 2004

Slate on 'Decentralized Intelligence'

Decentralized Intelligence - What Toyota can teach the 9/11 commission about intelligence gathering. By Duncan Watts

I found this a captivating read, though it's about risk management and problem solving, not decentralization of power per se. The vivid lessons excerpted in the full entry are a reminder that complete risk analysis is impossible, so the only sure strategy is containment.

We don't do that often enough in software -- tracing how errors propagate as ever-longer chains of legacy software are automated together...

Decentralized Intelligence - What Toyota can teach the 9/11 commission about intelligence gathering. By Duncan Watts

When organizations fail, our first reaction is typically to fall into "control mode": One person, or at most a small, coherent group of people, should decide what the current goals of the organization are, and everyone else should then efficiently and effectively execute those goals. Intuitively, control mode sounds like nothing so much as common sense. It fits perfectly with our deeply rooted notions of cause and effect ("I order, you deliver"), so it feels good philosophically. It also satisfies our desire to have someone made accountable for everything that happens, so it feels good morally as well.

But when a failure is one of imagination, creativity, or coordination—all major shortcomings of the various intelligence branches in recent years—introducing additional control, whether by tightening protocols or adding new layers of oversight, can serve only to make the problem worse...

In 1997, the Toyota group suffered what seemed like a catastrophic failure in its production system when a key factory—the sole source of a particular kind of valve essential to the braking systems of all Toyota vehicles—burned to the ground overnight...

How does one rapidly regenerate large quantities of a complex component, in several different varieties, without any specialized tools, gauges, and manufacturing lines (almost all of which were lost), with barely any relevant experience (the company that made them was highly specialized), with very little direction from the original company (which was quickly overwhelmed), and without compromising any of their other production tasks?...

the response was a bewildering display of truly decentralized problem solving: More than 200 companies reorganized themselves and each other to develop at least six entirely different production processes, each using different tools, different engineering approaches, and different organizational arrangements. Virtually every aspect of the recovery effort had to be designed and executed on the fly, with engineers and managers sharing their successes and failures alike across departmental boundaries, and even between firms that in normal times would be direct competitors.

Within three days, production of the critical valves was in full swing, and within a week, production levels had regained their pre-disaster levels. The kind of coordination this activity required had not been consciously designed, nor could it have been developed in the drastically short time frame required. The surprising fact was that it was already there, lying dormant in the network of informal relations that had been built up between the firms through years of cooperation and information sharing over routine problem-solving tasks. No one could have predicted precisely how this network would come in handy for this particular problem, but they didn't need to—by giving individual workers fast access to information and resources as they discovered their need for them, the network did its job anyway.

Perhaps the most striking example of informal knowledge helping to solve what would appear to be a purely technical problem occurred in a particular company that lost all its personnel associated with maintaining its data storage systems. The data itself had been preserved in remote backup servers but could not be retrieved because not one person who knew the passwords had survived. The solution to this potentially devastating (and completely unforeseeable) combination of circumstances was astonishing, not because it required any technical wizardry or imposing leadership, but because it did not. To access the database, a group of the remaining employees gathered together, and in what must have been an unbearably wrenching session, recalled everything they knew about their colleagues: the names of their children; where they went on holidays; what foods they liked; even their personal idiosyncrasies. And they managed to guess the passwords. The knowledge of seemingly trivial factoids about a co-worker, gleaned from company picnics or around the water cooler, is not the sort of data one can feed into a risk-management algorithm, or even collate into a database—in fact, it is so banal that no one would have thought to record it, even if they could. Yet it turned out to be the most critical component in that firm's stunning return to trading only three days after the towers fell.

So, how does one make this kind of magic happen? Unfortunately, no one is quite sure.

Posted by admin at 07:17 PM | Comments (0) | TrackBack

August 05, 2004

Some Decentralization papers for the day

I know it's a little rude to just post links and little more, but I at least wanted to note four papers I came across today. The first is pretty fanciful, but I think it actually speaks to a real set of problems for programming "mass quantities" of computing devices: Spray Computers: Frontiers of Self-Organization for Pervasive Computing Marco Mamei, Franco Zambonelli Dipartimento di Scienze...

Abstract: We envision a future in which clouds of microcomputers can be sprayed in an environment to provide, by spontaneously networking with each other, an endlessly range of futuristic applications. However, beside the vision, spraying may also act as a powerful metaphor for a range of other scenarios that are already under formation, from ad-hoc networks of embedded and mobile devices to worldwide distributed computing.

The 2003 Blackout - Reference and Analysis from the Kennedy School of Government, particularly pointing to the US-Canada Power System Outage Task Force. Final Report on the August 14, 2003 Blackout in the United States and Canada: Causes and Recommendations.
April 2004. 238 pages.

These last two seem like important theoretical contributions, regarding heirarchical composability of policies in coalitions. I really ought to come back to this and write more soon...

Flexible Regulation of Distributed Coalitions - Ao, Minsky

Developing Multiagent Systems: The Gaia Methodology - Zambonelli, Jennings, Wooldridge

Posted by admin at 09:37 PM | Comments (0) | TrackBack