Table of Contents

Feature Article 1 
RLG DigiNews: Taking Stock at Five Years, by the Editorial Staff
          Link Analysis in RLG DigiNews, by Richard Entlich

The Final Cedars Workshop: a report from Manchester, UK, by Michael Day

                     print this article

Figure 1: The masthead has evolved to reflect the changing look of the Web

Publishing Milestones

• RLG DigiNews began as a quarterly newsletter, but became a bimonthly publication in its second year. The Council on Library and Information Resources provided initial support for the two extra issues each year.
• One FAQ and at least one Highlighted Web Site (HWS) have appeared in every issue. In recognition of the increasing popularity of FAQs, we moved this feature beginning with the August 1999 issue from its less obvious location in the midst of news and announcements to a more prominent place after the HWS.
• Technical Reviews by editorial staff appeared only in Volume 1. These were replaced by periodic Technical Features, for the most part written by external authors.
• The first Conference Report appeared in the December 1999 issue, reflecting the increasing importance and regularity of key meetings on digital imaging and preservation.
• Beginning in 2000, each issue included coverage of some aspect of digital preservation. Relevant articles and items were flagged by a new icon that incorporated the infinity symbol typically associated with preservation, e.g., denoting the use of permanent/durable paper.
• The Editor's Interview, introduced in August 2000, has provided an opportunity for focused discussions with key people on current hot topics.
• For the first issue, searching was limited to use of a Web browser's "Find" or "Search" function, but by the second issue, viewers could browse the tables of contents or use keyword searching. Author and title indexes were subsequently added.

Access and Use

The initial intended audience for RLG DigiNews included "managers of digital initiatives with a preservation component or rationale." Since then, the reader base has grown dramatically, with the number of hits more than tripling from 1997 to 2000, from just over 20,000 hits to over 70,000. RLG reports that this publication is one of its most popular electronic resources. Lars Aronsson's Telecom History Timeline mentions the founding of RLG DigiNews as a historic event in 1997. The publication meets the Americans with Disabilities Act requirements for accessibility.

Each issue attracts thousands of readers on five continents. Back issues have a long shelf life. Usage of many early issues has not diminished substantially over time, and in some cases has increased. The October 15, 1998 issue, for instance, had more hits in 2001 than it did in 1999. Some issues remain significantly more attractive to users than others. The two most popular issues of 1999 featured lead articles on digital imaging and preservation microfilm (February) and digitization costs (October), proof that both topics continue to spark interest.

A Google index search identified over 1,000 links from various Web sites to RLG DigiNews. The publication shows up frequently on resource pages of consultants and faculty for digital preservation, digital imaging, and library conservation. Library portals around the world, including those in Australia, Canada, China, Europe, Israel, New Zealand, South Africa, South America and the United Kingdom, link to the publication. Twenty-one features have been highlighted in Current Cites, which each month selects "only the best items to annotate" from the current literature in information technology in print and digital form. Preservation "safekeeping" arrangements for many RLG DigiNews articles have been made with Australia's PADI, a subject gateway to select digital preservation resources. (see below)

Figure 2: A map showing the world readership of RLG DigiNews. The majority of readers come from North America (74%) and Europe (19.5%). Click on the pin to see a larger map.

1. Organizational commitment

• What is RLG's commitment to maintain and continue the publication?
• What is the funding stream; for how long is it secure?
• Does RLG have a preservation strategy/plan in place?

2. Technical infrastructure

• Where do the bits reside?
• What kind of hardware/software/server is used?
• Is it backed up, 24/7 supported?

3. Data Fixity

• What means are in place to secure the files and protect them from unauthorized change and use, data corruption, etc?

4. Format stability, reliability, and complexity

• What formats are used? What versions?
• Do they adhere to common, open standards?
• Is the coding correct and the data validated (e.g., HTML validator, parser)?
• Does the journal rely on experimental approaches (e.g., technology that may be very short-lived)?
• Is format control exercised by the editorial staff (E.g., does the staff do the mark-up or do contributors, does the journal establish and maintain format requirements)?
• How complex are the formats in terms of variations, computation, volume?

5. Authenticity and Provenance

• What is the policy on correcting mistakes? Is the original version maintained or the changes noted?

6. Redundancy

• Is the publication mirrored? If so, where and in how many places?
• Is there a formal agreement for mirroring in the works?

7. Metadata

• Technical (are the technical approaches well documented, e.g. use of javascript, the guts of the technical application, dependency on external programs and scripts, documentation on changes):

• Navigation (nature and extent of descriptive and structural metadata, e.g., SGML, Table of contents, consistency of approach, etc.):

• Resource discovery: How can people find the journal?
• What search engines and abstracting services pick them up?
• What practices promote/inhibit resource recovery (e.g., use of metatags)?
• Does the journal provide indexing/searching features itself?

8. External dependencies

• Does the publication use consistent/persistent link naming?
• What's the nature and extent of dead links?
• Are links really dead or just moved?
• What policy does the publication follow, if any, when including external links, e.g., linking priorities, extent of monitoring and updating?

9. External, third party archives

• What external archives cover the journal and how complete is the coverage; to what extent could the journal be recreated from these archives?

Robin Dale: The Internet Archive can capture pages and sites that might not be saved otherwise and may be a piece of a retention program but should not be viewed as a substitute for a digital preservation program for RLG DigiNews. The Internet Archive holds copies of all issues of RLG DigiNews with the exception of the last, Volume 6, Number 1, 15 February 2002. This is because the most recent crawl of the site was on 7 February 2002. Since crawls of the site tend to take place every 8 to 9 weeks or sometimes longer, I'd imagine that this "missing issue" will be covered soon. Other than that, the journal could be recreated.

Cornell Staff Note: The Internet Archive recently launched the "Wayback Machine," an online tool to search the archives's vast holdings. When we used the Wayback Machine to search for RLG DigiNews we discovered some interesting results, as indicated by the table below. Note that the first copy listed for the April 1997 issue was from May of that year, while the first copy obtained for the RLG DigiNews home page was from December 1998, a year-and-a-half after the journal was first published.

Table 1. The Wayback Machine capture dates for RLG DigiNews issues

Besides the Internet Archive, the National Library of Australia's Preserving Access to Digital Information (PADI) initiative has established the Safekept program. Being selected for the Safekept program provides an opportunity for organizations to establish, review and/or enhance their Web site preservation programs. The SafeKept program identifies a nine-step program for insuring the preservation of the selected Web sites, to which contributing organizations must adhere. The Safekept program has many, but not all issues of RLG DigiNews marked within its databases. This "minimalist" approach was borne out of an agreement with PADI that RLG would provide for/maintain all of RLG DigiNews in many ways and formats and therefore it wasn't absolutely necessary for NLA to do the same.

10. Look and Feel

• Is the old design and any functionality of early issues maintained (e.g., interface changes)?

11. Virtual content

• How much of the content is virtual (e.g., created at the point of access, or generated on-the-fly)? How well can this virtual content be maintained?

12. Ability to retain extended (added-value) services of the journal

• Is preserving its function as a ready reference database or other information services supported?

We were pleased to discover that 87 percent of respondents claimed that the content of RLG DigiNews is "just about right" in its usual level of technical detail. We found it interesting that two-thirds of respondents favor digital imaging features over other parts of the journal. However, respondents were split on the type of content they prefer; 39 percent favor policy recommendations, while 28 percent prefer information on technical standards and best practices, and 27 percent favor equipment reviews. Preferences may be changing, as we found that the number of new readers (those who have discovered the journal within the last two years) were slightly more numerous than responses from long-time readers.

We were also interested to learn that 60 percent of respondents discovered RLG DigiNews through listservs, while only 23 percent learned about it from colleagues and just 10 percent through other publications and Web sites. This is no doubt due to posting announcements of the release of new issues on 20 listservs worldwide.

In sum, as we begin our sixth year of publication, the state of RLG DigiNews appears healthy. Although we can't forecast technological advances and readers' interests for 2008, we can expect at least as many changes as the past five years have witnessed. The editorial staff will maintain flexibility, will periodically take the pulse of our readers, and will look forward to writing the 10th anniversary article!

Link Analysis in RLG DigiNews Richard Entlich Once an issue of RLG DigiNews goes "to press" its content is considered fixed. Subsequent changes are only made to correct significant errors, and those are always documented within the issue. Although such a policy guarantees the editorial integrity of the publication, it also means that, over time, the links to external Web sites will gradually degrade. How bad a problem is "link rot?" Pretty bad, as any regular Web surfer will quickly tell you. Not only does Web site content change and move around a lot, but domain names lapse and are reassigned on a regular basis. We conducted an analysis of links to external Web sites from the first five years of RLG DigiNews issues. The results are shown below. When all the issues for each year of publication are averaged together, there is a nearly linear increase in the annual percentage of bad links, from about 10% in our most year of publication (2001) up to about 40% in our first year (1997). There were a total of 1236 unique external links in the 28 issues of RLG DigiNews published prior to this issue. How did we determine the validity of so many links? For the most part, we did what most people would do when faced with such a large task: we took advantage of automation and employed a software link checker. Writing a link checker must either be a very popular assignment in computer science classes or else part of a rite of passage into the world of open source computing, because there are dozens of link checkers available. Many are freeware or shareware or free online services. A good list of available products and services (with a few of its own bad links) is available at http://www.elsop.com/wrc/comp_ls.htm. As one might expect with such an abundance of products on the market, there are some substantial differences in features and performance. We tested only a few products, but found some considerably more flexible and powerful than others. The differences have an impact not only on ease of use and reporting capability, but on the validity of the results. The operational basis for link checkers is fairly straightforward. Much like a browser, a link checker connects to Web sites using http (hypertext transfer protocol) and gathers data from the server. However, instead of rendering the data into a viewable Web page, it extracts status information that is part of every http exchange, but not always displayed to the end user. Most everyone has seen a message reading "404 Not Found" when trying to access a URL that's no longer available. The 404 is one of dozens of http status codes defined in the protocol. A "successful" http transaction usually produces a status code such as 200 (meaning "OK") or perhaps 301 or 302 (for a redirect to a new location). Status codes for properly completed http transactions are not normally displayed to users. How accurate a picture of a Web site's link status does a simple compilation of http status codes provide? That depends a lot on one's definition of a "good"' or "bad" link. For our purposes, we defined a good link as one that leads to (roughly) the content intended by the original reference, either directly or through an automated redirection or refresh. We even accepted situations where one additional manual click would be required to find the original content, as long as the path to that material was fairly evident on the first page brought up. Unfortunately, even given this fairly liberal definition of "good," most link checkers relying on simple status codes will significantly overestimate the number of good links on a site. There are many situations in which a site that appears good to a link checker may in fact fall short of our definition of good. Status code does not equal content. Just because a URL is still good doesn't mean the content hasn't changed dramatically. The domain name registration may have lapsed and been purchased by another entity. There may be no connection whatsoever with the original content or content provider. Putting out mixed messages. While displaying html content that clearly says "bad," the status code sent out via http says "good." This happens most frequently with sites that have substituted a custom error message for missing pages and failed to associate the proper status code with it. It's obvious on manual examination that these pages are bad, but without special effort, a link checker won't detect it. Depending on how they obtain their status codes, different link checkers may vary in their reporting on such sites. Moved or lost in transit? There are various ways for sites to indicate that content has moved to a new location. One method (called a meta-refresh) is usually reported as good as long as the page with the "we've moved" message loaded correctly. There are also different kinds of redirects. Redirects produce their own status codes, so at least one is alerted that something may be amiss, but a link checker may assume that as long as the redirect succeeds, the page is good. Frequently this is not the case. Also, since assignment of redirect codes requires manual intervention by site operators, it is prone to misapplication. The figures in the graph above were not taken at face value from a link checker. We did some manual checking of the results in order to verify their accuracy. We manually checked all reported codes for permanent or temporary redirections. We found that 25% of the permanent redirections and nearly 30% of the temporary redirections did not pass our definition of good. We also checked certain other reported server problems. We adjusted the results in the graph to reflect these findings. We did not manually check all the links reported as good by the link checker, but we did test a subset, just to get an idea of the error rate. Of the 68 links reported good in the three issues of RLG DigiNews for 1997, 14 (a little over 20%) did not pass muster. This discrepancy is not accounted for in the graph above, so, according to our definitions, the percentages of bad links in RLG DigiNews it shows should be higher. Despite these limitations, however, link checkers can form an essential part of a Web maintenance program, whether for your own site, or an external site that contains resources important to you. But it is important to understand that link check reports cannot necessarily be taken at face value. More expensive link checkers provide configuration and customization options that can help produce more accurate status assessments. Given the abundance of available link checkers, our advice is to take advantage of freeware as well as trial versions of shareware and commercial products to find the product that meets your needs.

—Peter Botticelli, Robin Dale, Carla DeMello, Barbara Berger Eden, Richard Entlich, Anne R. Kenney, and Nancy McGovern

                      print this article

Figure 1: A scribe in his study. Valerius Ma imus, "Faits et dits mémorables." Bruges, 1479

Gutenberg's invention altered the job. Archives branched off into a parallel institution for hand-written materials. Librarians largely abandoned the hands-on skill of the copyist. They lost control of the means of production. Arguably the world's first capitalists, publishers appeared on the scene to decide what would appear in print and, incidentally, they redefined librarianship. Publishers made reading portable and developed the crucial trappings for the modern book—e.g., the title page, tables of contents, indexes. The new industry also helped spawn the concept of authorship and the concomitant need to protect the right to profit through copyright (1).

Library involvement with reproduction techniques returned to a degree at the turn of the twentieth century. A new profession had just appeared in response to the rise of mass culture, which followed the introduction of wood-pulp paper and advances in the printing press. The first generation of college-trained librarians benefited from printing advances for their professional tools, especially the Library of Congress's (LC) introduction of inexpensive catalog card sets. The typewriter with supplementary mimeographs and carbon copies also entered the consciousness of information specialists. These technologies encouraged the development of new descriptive tools, especially the invention of archival finding aids at LC's Manuscript Division (2).

Figure 2: Taken from the Atlantic Monthly Advertiser in The North American Review. / Volume 82, Issue 494 from Cornell's Making of America Collection

Reprography blossomed anew in the 1960s. The advent of photocopiers brought unexpected change. "Xeroxing" became a new verb in short order. We could do away with messy mimeographs and carbons. The ability to make facsimiles helped repositories to deter theft and mutilation, as well as assisting with preservation. Photocopiers even impacted the law. They joined the cassette recorder as the technical impetus for the 1976 Copyright Act: Title 17 US Code. The largely unfettered "right of sale" was revised to protect authors, but still give libraries and archives reasonable abilities to make copies. Section 107 established the principle of "Fair Use." Special dispensation was granted in Section 121 to reproduce for the blind or others with disabilities. Most importantly, Section 108: Reproductions by Libraries and Archives gave the institutions the authority to make preservation copies and allow users to make their own copies on their premises. In exchange, the law called for clear procedures, postings, and rights management policies (3).

Technology and repository interests moved forward. The first generation of specialized "wet" papers and awkward processing matured to plain paper copiers. Machines added production capacity and ever more bells and whistles. Repositories developed the copy center—a new business service and revenue stream. The new equipment brought many institutions face-to-face for the first time with the intricacies of repair contracts and licensing arrangements. In the 1980s, an onslaught of facsimile machines pushed toward the provision of electronic means of distribution. Although somewhat constricted by the danger of viruses and copyright implications, many libraries also responded to the delivery potentials of downloading to computer disks and in the 1990s to e-mailing through the Internet.

Coming Impacts
The most recent advance in reprography is the move from analog photographic process to digital scanning. This is a robust and differentiated marketplace with competition among Bell & Howell, Canon, Ricoh, and 3-M. The Reuther Library is considering three types of equipment. One is a volume machine replete with a hopper feeder, collation bins, and automatic stapler. Secondly, our reference room will trade in its microfilm reader/printer for a significantly enhanced product. Third, the conservation lab will get a "Cadillac." The new machine has an oversize scanning area and can cradle delicate materials. Its microprocessor will correct for curvatures in the page and remove the image of a finger holding down the page. The technology offers other equally intriguing prospects and implications:

• Although the first copy is slower than with analog machines, warm-up time is shorter.
• The "scan once, print many" feature reduces the physical scanning for large jobs and allows for efficient storage for recurring jobs.
• Storage is in the computer's temporary RAM (random access memory) as standard TIFF files, which offer crisp and exact output (utilizing non-lossy compression).
• Identifying marks and time stamps can be automatically embedded in the image. You may also be able to embed electronic watermarks or signatures.
• Files are available for faxing or e-mailing.
• Machines can be tied into a local area network for remote control and collective action; e.g., several copiers could be tied together for large jobs.
• The printer is separate and can be networked into a LAN as a remote station for a variety of jobs.
• Use can be monitored and for the first time we can actually capture or identify the images being scanned.
• Images can be exported for fine-tuning by imaging software, converted to text by optical character recognition (OCR) systems, saved as PDF files to preserve the original layout, or to JPEGs for posting on the Web.

• Scanning files are temporary, a one-shot application to reproduce an in-kind copy on paper.
• Patron copy stations are limited to printed output.
• In-house duplication services, especially for items in copyright, are similarly print only.

• How much do you put on the Web?
• Is the aim complete public access or to tease people to research in-house?
• Are there financial prospects to be examined and are you afraid to "give away the store?"
• Are there privacy, legal, or political factors that mitigate against mounting certain materials?

                      print this article

The Final Cedars Workshop: a report from Manchester, UK

On 25-26 February 2002, 70 invited delegates met at UMIST in Manchester for the final Cedars Workshop. Digital preservation has been a high strategic priority for the Consortium of University Research Libraries (CURL), as evidenced by the Cedars (CURL Exemplars in Digital Archiving) project, which was a collaboration between three CURL institutions, the universities of Cambridge, Leeds and Oxford. The project was due to end in March 2002 after four years. The outcomes of the project were therefore of particular importance to CURL libraries and most of them were represented at the workshop. Publishers have also been regarded as key stakeholders in digital preservation and constituted another group of invitees, as well as playing an active role in the programme for the workshop. Other key decision-makers were also invited. The main purposes of the workshop were:

• to disseminate information about the Cedars project to the main beneficiaries of the project,
• to put that work into a wider context, recognising some of the key developments in digital preservation which have developed in parallel with the Cedars project,
• and last but not least, to look forward to what should happen after the end of the project.

Footnotes
(1) Consultative Committee for Space Data Systems, Reference model for an Open Archival Information System (OAIS), CCSDS 650.0-R-2. Red Book, Issue 2, July 2001. (back)
(2) CAMiLEON project (back)
(3) Robin Dale and Neil Beagrie, Digital preservation conference: report from York, UK, RLG DigiNews, 4 (6), 15 December 2000. (back)



Highlighted Web Site

                      print this FAQ

Assuming most publications that produced content in 2001 are still functioning, more than half of the publications launched in 1997 appear to be still viable. Nevertheless, DLN has plenty of company. Of the titles we reviewed, a dozen either didn't survive in any form, or appear to have stopped publishing by 1999.

Determining at what point an online publication has ceased to be a going concern can be tricky. In most cases, all one can do is try to judge from the length of the hiatus and other signs of neglect. As with DLN, it is rare for a Web publication to openly acknowledge that it is no longer operating. There are a few exceptions. The Annihilation Fountain, which was described as "a journal of culture on the edge" published 12 issues and then wrote its own edgy obituary.

Others are less willing to throw in the towel. The Journal of Credibility Assessment and Witness Psychology, a "peer-reviewed archival scientific journal focused on the scientific study of credibility assessment and witness psychology" has published three issues, the most recent in 1999. A note on its Web site says "JCAAWP is still in business. We have been suffering from a lack of publishable articles. Your submissions would be welcomed." The site was last updated in December 2000.

Even more unusual than a publication that acknowledges its death is one that has written a last will and testament. Consider Videre: A Journal of Computer Vision Research, "a refereed, archival internet journal of computer vision" which posted this note on its site:

One reason publicly acknowledging death might be important is to provide a signal that help is needed to preserve the publication. As indicated earlier, we were unable to find any archive for six of the 65 titles. In a few cases the Internet Archive captured fragments, but typically not whole publication runs.

Does it matter if the output of these short-lived electronic publications disappears? Should Web publications be given more attention than print publications of similar stature?

Such judgments are for collection development specialists and cultural historians to make. We will note, however, that a search of OCLC's WorldCat shows that five colleges and universities thought enough of DLN to catalog it. We don't know whether they have learned that DLN stopped publishing four years ago or would be concerned if its archive disappears from its server at Rutgers.

Clearly, more analysis of electronic publishing patterns and the factors that lead to loss of content is needed. As part of Project Prism, Cornell is studying these and related issues in a Digital Information Longevity Study.

—PKB, RGE

Calendar of Events

Multimedia Archive Preservation: Practical Workshop
May 22-24, 2002
London, UK
This workshop will incorporate the combined experience of ten major European broadcast archives in developing a solution to digital preservation.

Off the Wall and Online: Providing Web Access to Cultural Collections
May 30-31, 2002
Lexington,MA
The Northeast Document Conservation Center (NEDCC) is sponsoring this conference, that will explore the powerful potential of digital technology to transform education, outreach, and marketing activities in museums and other cultural institutions of all sizes. The faculty will include museum professionals, technical experts, educators, and administrators, who will discuss new digital access possibilities as well as the complexities of producing digital products.

The Fifth International Symposium on Electronic Theses and Dissertations (ETD)
May 30- June 1, 2002
Provo, Utah
The theme this year is: Pioneering on the Electronic Frontier: E-theses and Intellectual
Solidarity. Topics include: ETD costs: direct cost versus hidden costs, digital libraries and ETD access issues, and multimedia content in ETD.

Digitisation Summer School
June 30th - July 5th 2002
Glasgow, Scotland
Participants in the course will examine the advantages of developing digital collections of heritage materials, and will investigate issues involved in creating and managing access to such collections. The lectures will be supplemented by seminars and practical exercises.

Fourth Annual UCLA/Getty Museums, Libraries and Archives: Summer Institute for Knowledge Sharing
July 29 - August 1, 2002
Los Angeles, CA
The institute is an intensive four days of instruction and dialogue for professionals involved in creating, sharing and preserving electronic information in museums, libraries, and archives.

Creating Electronic Texts and Images-A Practical "Hands-on" Exploration of the Research, Preservation and Pedagogical Uses of Electronic Texts and Images in the Humanities
August 18 - 23, 2002
New Brunswick, Canada
This course will center around the creation of a set of electronic texts and digital images. Topics include: XML tagging and conversion, using the Text Encoding Initiative Guidelines, e-books, the basics of archival imaging, and the form and implications of XML.

Symposium 2003- Preservation of Electronic Records: New Knowledge and Decision-Making September 15 - 18, 2003
Ottawa, Canada
Submissions are now being invited for this symposium. The purpose is to build awareness of digital preservation by bringing expert and leading edge opinions to a larger audience including small and medium-sized archives, libraries, and museums. The focus will be on making decisions and finding practical solutions that can be implemented immediately, especially for the materials that are at risk of being lost within the next 10 to 20 years. For further information contact: cci-icc_publications@pch.gc.ca.



Announcements

The Illinois OAI Protocol Metadata Harvesting Project Repository
This project is testing the viability of using the Open Archives Initiative (OAI) protocol for harvesting metadata, and exposing it with a search interface to enhance resource discoverability for materials that represent cultural heritage. The repository includes metadata records donated by over 26 institutions. Please send comments to Joanne Kaczmarek.

Evaluation Initiative for XML Document Retrieval
The DELOS Network of Excellence for Digital Libraries invites participation in an evaluation initiative for XML document retrieval. The widespread use of XML in digital libraries, catalogues, scientific data repositories, and across the Web has prompted the development of appropriate searching and browsing methods for XML documents. This initiative provides an opportunity for participants to evaluate their retrieval methods using uniform scoring procedures and a forum for participating organizations to compare their results. The invitation is open to all research groups with an interest in XML retrieval.

California Digital Library Opens Online Repository for Working Papers
The California Digital Library has launched a Web site and associated digital services to store and distribute academic research results and working papers. The eScholarship Repository includes a set of author and reader services for the rapid dissemination of scholarship authored or sponsored by faculty from the University of California. Its initial focus will be on working papers from the humanities and social sciences.

OAIster Project at the Digital Library Production Service of the University of Michigan Libraries
The OAIster project is one of the Andrew W. Mellon Foundation Metadata Harvesting Initiative grants. The goal is to create a wide-ranging repository of free, useful, previously difficult-to-access digital resources irrespective of subject area or format, and that will be easily searchable by anyone.

Digital Document Quarterly (DDQ)
Worth a look is the Digital Document Quarterly. This online publication will address issues such as document qualities that can be conferred and managed automatically, including integrity, confidentiality, and authenticity.

The Tibetan and Himalayan Digital Library
Making use of new digital Web based technologies to collaboratively solicit, generate, publish and integrate all aspects of scholarship on the geographical, cultural and linguistic regions associated with Tibet and the Himalayas, this site will be of interest to our readers.

Publishing Information

RLG DigiNews (ISSN 1093-5371) is a newsletter conceived by the members of the Research Libraries Group's PRESERV community. Funded in part by the Council on Library and Information Resources (CLIR) 1998-2000, it is available internationally via the RLG PRESERV Web site. It will be published six times in 2002. Materials contained in RLG DigiNews are subject to copyright and other proprietary rights. Permission is hereby given for the material in RLG DigiNews to be used for research purposes or private study. RLG asks that you observe the following conditions: Please cite the individual author and RLG DigiNews (please cite URL of the article) when using the material; please contact Jennifer Hartzell, RLG Corporate Communications, when citing RLG DigiNews.

Any use other than for research or private study of these materials requires prior written authorization from RLG, Inc. and/or the author of the article.

RLG DigiNews is produced for the Research Libraries Group, Inc. (RLG) by the staff of the Department of Preservation and Conservation, Cornell University Library. Co-Editors, Anne R. Kenney and Nancy Y. McGovern; Production Editor, Barbara Berger Eden; Associate Editor, Robin Dale (RLG); Technical Researchers, Richard Entlich and Peter Botticelli; Technical Coordinator, Carla DeMello.

All links in this issue were confirmed accurate as of April 12, 2002.

Please send your comments and questions to preservation@cornell.edu.