NETTING IT OUT

Until now, there has been no consistent way for publishers, aggregators, or librarians to consistently track access of published intellectual assets. The answer appears to be an emerging standard for tracking online usage--the COUNTER project, which is intended to provide online usage statistics that are credible, consistent, and compatible.

Although the initial focus of the COUNTER project has been on scholarly journals, books, and conference proceedings, the implications are profound for all business publishing, including software publishing. When you know who is actually accessing and reading your information, you can price it fairly and reap the proper rewards.

TRACKING USAGE

No matter what business model you use--free access, pay-per-view, group or individuals subscriptions, or membership(1)--everyone in the intellectual-property food chain wants to know who is accessing what information. Of course, there are also privacy considerations. You may have the legal right to access information, but you don't want to have your reading habits tracked for a variety of reasons.

In the world of scholarly research, librarians have been on the forefront of both protecting readers' privacy rights and of tracking usage.

Who Is Accessing Information?

The emerging de facto practice in the scholarly research world is that you have the rights to access certain information because you belong to an institution that has purchased those rights. Your individual identity may be known by the administrators at the institution, but it is not passed onto the publisher or the party that is hosting the intellectual property on behalf of the publisher or the institution who purchased access rights. Therefore, unless an individual chooses to identify him or herself in order to receive additional benefits (personalized services, and so on), he may have full access without disclosing his identity. This authentication and authorization is typically handled through IP-based authentication.

IP-BASED AUTHENTICATION. In the scholarly research world, most libraries prefer to use the institution's IP address as their single sign-on mechanism. If I am "coming from" a university's Internet address as a professor or a student, I have the same access rights to all the material to which that university subscribes. Additional authentication and permissions may be required to access certain materials, as well as to log onto the university's network at the outset.

This automatic authentication is a boon for the student or researcher. He or she no longer needs to bother remembering lots of log-ons or passwords. And librarians love this IP-based authentication because it takes them out of the administrative gatekeeper role.

Where Is the Information Being Accessed?

The vast majority of researchers who search and access journal articles electronically do so through their institution's library. But, rather than keeping an electronic copy of each journal locally, most libraries pass the reader through to one or more external e-journal portal(s). These are hosted by journal publishers, aggregators, or gateway providers. Some libraries keep local copies of frequently-accessed journals and/or cached copies of frequently-accessed articles. (This leads to the "appropriate copy problem," which we'll touch upon later.)

Each visitor is automatically authenticated based on the IP address from which they're coming. If they log onto an e-journal portal from home, they either pass through their institution's firewall to access the intellectual property the institution has licensed and/or they must logon with their own member or subscriber permissions.

What Information Is Being Accessed?

Although librarians are vigilant about protecting readers' privacy, they do want to know which of the information to which they've subscribed is being accessed and how often. Perhaps more important, librarians want to know when their readers are trying to gain access to information that they can't retrieve. For example, if you're reading a journal article that references another one, and you click on that reference, you want to gain access to the cited article, not be turned away because it isn't in the library's current collection.

Publishers also need to know what information is being accessed and how often in order to justify the subscription prices they're charging for the information.

THE USAGE TRACKING DILEMMA. Librarians don't feel they can trust the honor system of relying on the publishers or aggregators to tell them which of their material is being accessed. Moreover, the information they do receive from each of these parties is inconsistent and impossible to reconcile.

Librarians don't want to bear the burden of having to track usage across publishers' offerings because every publisher tracks different information and organizes it differently. As recently as October, 2001, the Association of Research Librarians (ARL) issued a report stating that "we conclude that it is largely impossible to compare data across vendors...we believe that the comprehensive standardization of usage statistics cannot be easily achieved in the short term"(2).

WHY A SOLUTION HAD TO BE FOUND. According to Peter Shepherd, Project Director of the COUNTER project, both librarians and publishers recognized the need for an objective way to monitor the usage of networked electronic resources. "Both publishers and purchasers need an objective way to track the usage of their information," Shepherd explained in his presentation at the STM Innovations conference.

"Libraries need online usage statistics in order to:

* Assess the value of different online products/services
* Make better-informed purchasing decisions
* Plan infrastructure

Publishers need online usage statistics in order to:

* Experiment with new pricing models

* Assess the relative importance of the different channels by which information reaches the market

* Provide editorial support

* Understand what features/products reach the market.

* Plan infrastructure, portal, and site design"

The answer for both communities appears to be an emerging standard for tracking online usage--the COUNTER project.

Counting Online Usage of Networked Resources (COUNTER)

The COUNTER project was launched in 2001 in response to librarians' and publishers' needs to reliably track information access. The project was endorsed by a number of influential organizations from both the librarian and the publishing community. (See Illustration 1). The goal of the project is to provide online usage statistics that are credible, consistent, and compatible. The not-for-profit COUNTER organization (www.projectcounter.org) is being funded by sponsors from the primary and secondary publishing communities. Basically, all of these publishers are throwing their hats in the ring and agreeing to develop and to comply with a common code of practice in monitoring usage of networked resources and in providing that information back to the purchasers of the information in a consistent and auditable manner.

Project COUNTER
Please download PDF to see the illustration.
Illustration 1. The goal of the COUNTER project is to provide online usage statistics that are credible, consistent, and compatible. Note that the COUNTER Code of Practice was released just a month ago.

Project COUNTER began with a survey of the paying customers. 650 librarians responded to the detailed survey, among them, 49 corporate librarians. Their needs were consistent and shaped the first version of the code of practice and implementation.

WHAT PUBLISHERS ARE REQUIRED TO TRACK & REPORT: Here's what the librarians asked the publishers to provide:

* Focus first on electronic journals and online databases

* Provide a small number of reliable reports

* Make these reports available (in Excel-downloadable format) on a password-controlled Web site, with email alerts when new data is posted

* Provide new reports at least monthly

* Data must be updated within two weeks of the end of the reporting period

* Supply all of last year's data and this year's data to-date

WHAT TYPES OF INFO NEEDS TO BE TRACKED? What information needs to be collected and reported? First, there's a standard definition of what is meant by "requests." For example, "all users' double clicks within 10 seconds on an http-link should be counted as only one request (allow 30 seconds for a PDF)." In addition, libraries want the publishers and/or aggregators to provide the following information for each request:

By page view:

* Bibliographic data (e.g., DOI, online ISSN, etc.)
* Page type, e.g., full-text article, citation, abstract
* Source of page, e.g., referred from an aggregator or gateway
* Authentication of user, e.g., IP address
* Access rights, e.g., access granted, turned away

By Session Data:
* Start time, end time

Market elements:
* Consortium member, association member, etc.

WHAT KINDS OF REPORTS DO LIBRARIANS REQUIRE? Here's the level of detail that librarians want to see in the first set of reports to be provided by electronic publishers (in 2003) in accordance with the COUNTER standards:

* The number of full-text articles requested by month and journal
* Turnaways by month and journal
* Number of item requests by month, journal, and page-type
* Total searches by month and collection

For databases, including abstract databases, the reports include:

* Total searches and sessions by month and database
* Turnaways by month and database
* Referrals by aggregator or gateway

The COUNTER project is now underway. By 2004, all participating publishers will be required to supply COUNTER-compliant information. The COUNTER standards are endorsed by the National Information Standards Organization (NISO), as well as several international bodies, notably, the Publishers Association (PA) and the STM. The complete list of COUNTER sponsors is as follows:

* Association of American Publishers/Professional & Scholarly Publishing Division http://www.pspcentral.org

* ALPSP, The Association of Learned & Professional Society Publishers http://www.alpsp.org

* ARL, Association of Research Libraries http://www.arl.org

* ASA, Association of Subscription Agents & Intermediaries http://www.subscription-agents.org

* Blackwell Publishing http://www.blackwellpublishing.co.uk

* EBSCO Information Services http://www.ebsco.com

* Elsevier Science http://www.elsevier.nl

* Ingenta http://www.ingenta.com

* Institute of Physics Publishing http://www.iop.org

* JISC, Joint Information Systems Committee http://www.jisc.ac.uk

* Lippincott, Williams & Wilkins http://www.lww.com

* Nature Publishing Group http://www.naturesj.com

* Oxford University Press http://www.oup.co.uk

* PA, The Publishers Association http://www.publishers.org.uk

* ProQuest http://www.il.proquest.com

* STM, Int. Assoc. of Scientific, Technical & Medical Publishers http://www.stm-assoc.org

* Taylor & Francis Group http://www.tandf.co.uk/

* UKSG, United Kingdom Serials Group http://www.uksg.org

Implications for Other B2B Publishers

Although the initial focus of the COUNTER project has been on scholarly journals, books, and conference proceedings, the organization expects to expand its scope to all other forms of information whose usage librarians (or other purchasers) might want to track across the network. What we like about the COUNTER project is that:

* It provides a set of standards and policies to which publishers can easily adhere

* It doesn't violate any reader's privacy

* Usage tracking information is provided in a way that makes it easy to analyze across publishers (e.g., an Excel spreadsheet with a common format)

* It tells you which assets people have accessed and which ones they tried to access but weren't able to access

The implications of COUNTER for the business publishing and library communities are certainly profound. The fact that many of the world's major B2B publishers are already conforming to these reporting standards bodes well for its continued success. Will this kind of standardized usage tracking and reporting be adopted in the software industry? We don't see why not.

How will standardized usage tracking impact the business models for publishers? Those who provide information that is used, even by a small community, will be rewarded. Those whose information is never accessed will not be able to charge for that information.

Will this kind of usage tracking extend into the consumer world? That's probably less likely, since it may be difficult to track usage without associating it with individuals in the consumer world.

*****ENDNOTES*****
1) For more information on business models, see " Understanding Digitization: Trends in Business Models "

2) ARL E-Metrics Phase II Report, October, 2001.
*****ENDNOTES*****