Search   |   Back Issues   |   Author Index   |   Title Index   |   Contents



D-Lib Magazine
September 2006

Volume 12 Number 9

ISSN 1082-9873

Handle Records, Rights and Long Tail Economies


John Erickson
Hewlett-Packard Laboratories

Red Line

(This Opinion piece presents the opinions of the author. It does not necessarily reflect the views of D-Lib Magazine or the Corporation for National Research Initiatives.)

Sometime during the spring of 2006 I began to draw connections between Handles, metadata services, and small-value (or niche) rights transactions in light of Chris Anderson's notion of "Long Tail" economies [1], [2]. My ramblings on this topic actually date back more than a decade and include a keynote I once gave entitled "Frictionless eCommerce and the Future of Rights Management" [3].

As I understand it, Chris Anderson's Long Tail observations refer to a reduction of transactional "friction." He asserts that the Internet, as exemplified by services such as®, NetFlix® and eBay®, reduce transaction costs and therefore enable market niches of (in principle) one. His thesis can best be summarized by:

  1. Make everything available
  2. Help anyone find it

In the first approximation it seems that rights transactions and related services might be the perfect bases for Long Tail economies, but to date the total cost of most rights transactions remains too high. Based on my own experience this is primarily due to the cost of aggregating the metadata required to support individual fine-grained transactions, and secondarily due to cost of maintaining the services bound to items by that metadata, including the hosting of rights information and the transactions themselves. From the standpoint of a rights transaction, metadata is the very essence of the product; without metadata, not only can there be no transaction, but there can be no availability, no discovery, nothing. Without metadata, the "product" doesn't exist.

Someone once said, "Metadata is the lifeblood of eCommerce..." [4], [5]. eBay works in part because its participants are willing to invest effort to "wrap" their items in metadata: to virtually represent individual goods and services by their metadata. That the eBay community buys into this is a testament to the fact that eBay doesn't demand too much of the participants; users must invest enough effort to convey the essence of the product, but there isn't so much friction so as to discourage their efforts and lose the opportunity. As a result, it has been said that we can literally "find anything" on eBay.

Except that we can't, at least not niche rights for arbitrary pieces of content that we might come across in our daily travels. If, for example, a user wishes to transact the rights to use a photo that they find on Flickr™ [6], they can't do it; they wouldn't know where to begin. But imagine that every image posted on Flickr was somehow "wired" into a service (or services) by which anyone could (a) obtain descriptive information about the image, (b) obtain basic rights information about the image, and (c) initiate a rights transaction (including an open-ended or "free text" request) specific to that image and the intended use. These services are not available today, and thousands of opportunities are lost – especially if we apply the Long Tail model to these niche rights transactions.

This is not simply a question of enabling the requester to (e.g.) email the rightsholder, because most people, including advanced amateurs, happen to be clueless about the practical aspects of copyright. Requesters and owners alike need to be transported into a transactional context, based on metadata associated with the item, that will facilitate a dialog and ultimately a transaction. So furthering the Flickr example, when a photo is posted, it would have metadata elements (Flickr calls their metadata elements "tags") associated with it indicating that the photo is transactable, as well as metadata that would link to a transaction portal.

It must be noted that Flickr does make it easy for members to choose from a selection of Creative Commons licenses [8] to associate with their images; this is an important step forward, both in terms of rights awareness but also in terms of demonstrating how rights and other metadata-based services may be associated with objects posted to social networking sites.

In the current services-oriented era of the Web, we wouldn't expect Flickr to host rights transactions; instead, niche services such as these would be hosted by third-party services and fueled by the metadata persisted in Flickr along with additional "value-added" metadata essential to the transaction and supplied by the services that need it. Think of these services (if they existed) as Flickr's "copyright partners," just as Creative Commons is today. We can imagine a number of niche offerings, from amateur grade to professional.

The linchpin in all of this could be the individual item's Handle record, a "Point in Space" where the critical metadata for the item could be aggregated in the infrastructure. As the Handle record is established, the Point in Space is "borne," with only minimal metadata; as new services are added, that point is embellished or "decorated." Whether such a Point in Space is implemented within the infrastructure via the Handle System® [15] or within some equivalent but proprietary service, the fundamental notion is low-cost metadata aggregation – a "halo of metadata" – around a unique identifier, useable by a variety of applications and services.

Eleven years ago Esther Dyson wrote of her notion of "intellectual value," whereby the real value of a creative work was the unique capabilities of its creator, and not the work itself. Dyson's key assertion was that "content providers should manage their businesses as if it were free, and then figure out how to set up relationships or develop ancillary products and services that cover the costs of developing content." [7] By this reasoning we see that it is critically important to maintain the links between a creative work and its network of contributors, because without these links – the ability to discover, for example, who the editor of an audio clip was, or even the audio engineer – the essential value of the work is lost. Maintaining these links can be difficult and expensive, and cannot be accomplished without some investment.

In the past decade I have been involved in several attempts to achieve these goals; specifically, I've helped create technologies and services that:

  1. Let users discover directly from embedded objects various information about those objects, including their creators, etc., and
  2. Let users initiate rights transactions directly from the context of use, including their desktop, rendering applications or (especially) creativity tools

Over this period my colleagues at several companies and I have successfully demonstrated prototypes of technologies and pilots of services that accomplish these goals, but none exist in the infrastructure today. That said, we have seen the world move baby-steps closer toward these objectives; as I noted in the Flickr example, the Creative Commons' approach to specifying and linking to rights information goes a long way toward the sorts of "attached" rights declarations that my colleagues and I envisioned a decade ago, and we all merely stand on the shoulders of Henry Perritt and his "permissions headers" concept, circa 1993 [9]. We have also seen the advancement of technologies that will facilitate this vision, including rights metadata standards like <indecs> [10], identifier standards like the DOI [11], infrastructure like the Handle System, and core technologies like XML, RDF and web services. By way of example, the California Digital Library project provides an XML schema and a number of XML examples illustrating one approach to encoding rights metadata for different kinds of assets stored in digital libraries [12].

It might be useful to reflect on Guy Kawasaki's "cynic's checklist" for the implementation of Long Tail ideas, which he refers to as the "tactical items it takes to succeed" [13]. In the context of metadata services and rights transactions, we need to apply these concepts to the creation of rights-related metadata; the aggregation of metadata in repositories such as the Handle System; support for discovery and rights transactions; and the actual participation in those transactions. In retrospect, it should be stated that my previous efforts in this area directly violated several of these tenets and/or required our customers (especially metadata producers) to violate them:
  1. Low-cost production: The cost of creating useful metadata and "registering" the item in a supporting service should be minimal.
  2. Un-demanding, un-selfish, un-financially motivated, just-plain stupid, or just-plain smart producers: The objective of the operator is to engage as many participants as possible, and good things will happen.
  3. Near-zero inventory carrying costs: The reality is that metadata-based services do incur costs, including hardware, support, storage, and bandwidth. Reduce costs by leveraging metadata persisted "in the infrastructure" as much as possible.
  4. Near-zero selling and marketing costs: Licensing templates should be ready-made and easily associated with items.
  5. Near-zero support and training costs: No active "training" should be required to either offer items or transact rights. Let the community do the training via Wikis!
  6. Fast fulfillment: Rights should be easy to transact, and the results are immediately available, including "instant royalties" to the provider.
  7. Infinite selection: Any creative work can be the basis for a rights transaction.
  8. Singleness of purpose: Focus the metadata and/or the service on what the item is; avoid the "general purpose".
  9. Highly optimistic, if not delusional, personalities: When creators consider making their works available for transactions, they should have faith, because a certain amount of speculation is essential to the Long Tail working...

Finally, in his Long Tail blog Chris Anderson recommends the following approaches to "scaling down" or minimizing up-front and operating costs in order to achieve Long Tail success [14]. I think these points perfectly reflect the requirements for metadata-based services, whether for discovery and retrieval or for rights transactions:

  1. Self-service: Give customers all the tools they need to manage their own accounts.
  2. "Freemium" services: "Give your service away for free...acquire a lot of customers very efficiently through word of mouth...then offer premium priced value-added services or an enhanced version of your service to your customer base."
  3. No-frills products: Don't make consumers pay for what they don't need.
  4. Crowdsourcing: Let customers do the work of building the service; they will expand it far beyond what employees could do on their own.

So what, finally, is our call to action? What do these Long Tail observations tell us about finally bringing metadata-based rights services to fruition, after almost a decade [16] of speculation and failure? Fundamentally, these observations inform us of how much of a challenge it is to build a community around these sorts of services, and in the end community-building is what this should be about. Our primary, optimistic, speculative motivation should be to connect creators and contributors, because such connections are based on intellectual value, the very essence of a creative work. The collaborations between creators, fostered by these connections, is what will ultimately lead to revenue.

We have yet to see a web-based metadata aggregation service that enables users to easily and arbitrarily bind metadata to an infrastructure-unique identifier. What we need is a service that is in the infrastructure (and secure) like the that provided by the DOI, as easy to use for creators as TinyURLs [14], as easy to consume as RSS feeds, and that exposes web-based APIs that are as easy to "mashup" with [18] as Google's Search and Map APIs. Perhaps what we are suggesting is a "Web 2.0" [17] rebirth of the Handle System that enables lightweight, secure, easy-to-use aggregations of metadata in the infrastructure, or what I call Points in Space.


[1] Chris Anderson, "The Long Tail." Wired 12.10 (October 2004). See <>.

[2] Chris Anderson, The Long Tail: Why the Future of Business Is Selling Less of More. Hyperion, ISBN #1401302378 (2006).

[3] John S. Erickson, "Frictionless eCommerce and the Future of Rights Management." Keynote address, IQPC Digital Rights Management and Digital Publishing Conference, London (July 2000).

[4] John S. Erickson, "Metadata Initiatives and the DOI: Implications for Electronic Commerce and Copyright Management Services." Trialogue No. 8 (Summer 1998).

[5] John S. Erickson, "The Role of Metadata Supply Chains in DOI-Based, Value-Added Services." ICSTI Forum, Quarterly Newsletter of the International Council for Scientific and Technical Information, No. 30 (April 1999). See <>.

[6] See <>.

[7] Esther Dyson, "Intellectual Value." Wired 3.07 (July 1995). See <>

[8] The Creative Commons is a nonprofit organization that offers flexible copyright licenses for creative works. See <>.

[9] Henry J. Perritt, Jr., "Knowbots, Permissions Headers and Contract Law." IMA Conference on Technological Strategies for Protecting Intellectual Property in the Networked Multimedia Environment, Boston (April 2-3, 1993). See <>

[10] Godfrey Rust and Mark Bide, "The <indecs> Metadata Framework: Principles, model and data dictionary." (June 2, 2000). See <>.

[11] See <>.

[12] CDL Rights Management Group, "copyrightMD Schema, Version 0.9." California Digital Library Project (March 24, 2006). See <>

[13] Guy Kawasaki, "The Wrong Tale: A Checklist for Long-Tail Implementations." Signum Sine Tinnitu blog (July 18, 2006). See <>

[14] Chris Anderson, "Scaling up is good. Scaling *down* is even better." The Long Tail blog (June 14, 2006). See <>

[15] See <>.

[16] John S. Erickson, "Requirements for DOI-based Applications and Services." Prepared for the NISO DOI Working Group, Washington, DC (January 30, 1998). See <>

[17] Tim O'Reilly, "What is Web 2.0: Design Patterns and Business Models for the Next Generation of Software." Published at (Sep 30, 2005). See <>

[18] Programmable Web, "Web 2.0 Reference Center." (2006). See <>

Copyright © 2006 John Erickson

Top | Contents
Search | Author Index | Title Index | Back Issues
Editorial | Next article
Home | E-mail the Editor


D-Lib Magazine Access Terms and Conditions