For many reasons, I found the copyright protection weak and the patent protection inappropriate for the legal protection of software, and I made a detailed proposal for a specific (in legal terminology: sui generis, or "on its own right") protection. My book was published on the subject and the Library of Congress acquisitioned it, but that's the end of the story—my proposal was not implemented by the legislators.

It was quite an asset—especially at that time—to be familiar with both fields, and I was at a crossroad where I could choose between a career as an intellectual property lawyer or as an educator and practitioner in computer and information science. I chose the latter, so the following is not a legal opinion but merely an editorial from someone who is a) not ignorant about the legal and the computer sides of the issues; b) an author of 300-plus articles, conference papers, columns, and reviews; c) familiar with the plight of the librarians and the consumers; and d) opinionated about digital copyright and "copywrong."
 

Loyalty Above Royalty
My opinion in some matters is certainly flavored by the fact that Information Today, Inc. (ITI) has paid me very decently for my articles for over 10 years under a simple gentlemen's agreement—without requiring the transfer of my copyright. There have been other perks, such as free admission to any conference organized by ITI. Although there is still much buzz since the Supreme Court's decision that freelancers own copyrights for digital reproductions of their works if they did not specifically sign them away in writing, I am not discussing that issue here. Above legal considerations are the moral ones, such as loyalty in this case. That's why I don't dispute this publisher's right to license my copyrighted works to online information services, which in turn create powerfully searchable full-text and/or page image databases and set reasonable prices of $2 to $4 for printing my full-text articles. I tackle that aspect with regard to a former British publisher of another column of mine in an upcoming Savvy Searching column in a 2002 issue of Online Information Review, 26(4).

This column is not the appropriate place for a legal treatise, so I will instead focus on the everyday anomalies in copyright matters that concern librarians and other information professionals who provide fee-based services to patrons and clients using document delivery services. I also cover some of the current copyright infringement cases in my annotated storybook of screenshots at http://www2.hawaii.edu/~jacso/extra.
 

The Wide Spectrum of Copyright Infringements
There is an array of possible copyright infringements, ranging from the blatant information highway robbery of snatching an entire Web site (the Academic Info v. Blacknet case) to selling derivative works (the ISI v. Prestige Factor case) to—hopefully and presumably—non-willful copyright infringements (many cases).

While authors are the "little people" in Leona Helmsley's parlance, the bigwigs have their own fights among themselves.

Copyright infringement disputes often erupt between well-established information companies. Large publishing houses have nothing to worry about, as they have armies of lawyers and can sue the life out of the infringing parties. Last December, FDC Reports, Inc. settled a lawsuit against NewsEdge Corp. and IMS Health, Inc. for unauthorized delivery of copyrighted materials owned by FDC, Inc., a Reed Elsevier company.

Last November, Elsevier and John Wiley filed a lawsuit against Kessler-Hancock Information Services, Inc., a traditional document delivery company, for "unauthorized photocopying from the publishers' journals for resale to the service's customers." The complaint alleges that "Kessler-Hancock solicits and fills orders for copies of copyrighted materials distributed by the publishers without the publishers' authorizations, and even collects purported copyright fees in connection with the copies without returning them to the copyright holders" (http://www.sspnet.org/public/news/details.cfm?id=83). The jury is still out, but librarians have known for quite a long time that the copyright fees that they pay may not always get to the legitimate copyright holders.
 

Copyright Fee Anomalies
This leads me to the anomalies that I keep discovering when I'm doing research about document delivery companies and their primary sources. Although my research features a number of such entities and databases, including The British Library (a favorite supplier for many document delivery companies), TheScientificWorld, UnCover Plus, and Infotrieve, I will use only the latter to illustrate my points because of space limitations. You can find examples of the confusing handling of copyright fees in the sciBASE database of TheScientificWorld service in the September 2001 issue of ONLINE, and in a follow-up ("Who's Inaccurate?") at http://www2.hawaii.edu/~jacso/extra.

Librarians have been struggling with copyright problems, including copyright fees. Stephanie C. Ardito and Paula Eiblum provided excellent, albeit troubling, insights about some of the copyright oddities in the document delivery world in their excellent series in the 1998 issues of ONLINE. My current survey just reconfirmed that their past experiences apply today in an even broader spectrum, triggered by the omnipresent and very convenient digital delivery of documents on the Web. CIL had a useful feature article about copyright issues for librarians, and an informative interview with Dave Davis, a project manager at the Copyright Clearance Center (and a library student then), in its June 1998 issue.

Still, there is much confusion, fear, uncertainty, and doubt about copyright fees, which are of deep concern for librarians and other information specialists. These people are the mediators and the interface between the customers and the document delivery services, and they have to answer questions and explain procedures and prices (which they themselves may not understand), and all the while they are trying to get the best price for their customers and to comply with copyright rules.
 

A Flat Copyright Fee?
At first glance you may let out a sigh of relief when you realize that Infotrieve has a laudable copyright policy (http://www4.infotrieve.com/docdelivery.asp), and that it uses a flat copyright fee for articles within the same journal. This implies that once you know the copyright fee, it will be the same for every article from the same journal (and often from all or most of the journals of the same publisher, as is the case with ITI) without respect to the length of the document (which can be a nightmare to determine, as we shall see later).

Imagine that you are searching for a CIL document, "Know When to Hold 'em, Know When to Fold 'em," by Péter Jacsó. It may comfort you to see the same $15 copyright fee in the result of a search showing two versions of a bibliographic record for the same document from CIL—courtesy of The British Library, which provides many of the bibliographic citations for the Infotrieve database.

Your relief will be cut short, however, because you will probably start thinking what a high copyright fee this is for Jacsó's usually two-page column. But wait, one record says it is from pages 38–43 (six pages), but the other claims it is on pages 38–39 and 42 (three pages). Should you fold 'em?

No. After all, you pay the same royalty for CIL articles, regardless of the number of pages, right? But then you get a sinking feeling in your stomach. You recall that just a short time ago you were charged a $5 copyright fee for a three-page CIL column by this author, and a printout shows that your recollection is correct. That price was "by the book," the same as ITI set and kept unchanged for many years: a base fee of $3.50, plus a 50-cent-per-page fee.
 

Will the Real Copyright Fee Please Step Forward!
The new $15 copyright fee for CIL articles is discomforting. It sure makes the article (service) fee look better (lower), but otherwise it does not make sense. You can't fathom how Infotrieve ever arrived at that number.

Did it get it from The British Library catalog? No, BL says the copyright fee is £5.27, which is about $7.50. This is half again as much as the fee set by the publisher, but still only about half of what Infotrieve charges for such short, two-page articles that are typical for columnists in many ITI publications.

Maybe Infotrieve got the copyright information from the Copyright Clearance Center. No, the CCC provides the same information ($3.50 plus 50 cents per page) as the publisher.

This copyright fee, which does not jibe with any of these published fees, may be difficult to justify to your patron. Is there perhaps a profit added to the copyright fee? That's unlikely. Now, should you fold 'em? Maybe give another try.
 

Base Fee Plus Page Fee
Look up how much the copyright fee is for articles from the Information Today newspaper through Infotrieve. Good, it is only $4.30 for the first document. But the price looks a bit odd. Why? Because for articles in Information Today, Infotrieve does not use a flat fee, but a base fee of $3.50 plus a page fee of 50 cents, and then adds the 30-cent Copyright Clearance Center processing fee. This is odd, but moneywise better than the $15 flat copyright fee, and you can justify it to your client.

Fair enough, you think—until you realize that the bibliographic records from The British Library, which are the backbone of the Infotrieve database, very often claim longer page spans than the articles cover, and now (as well as in the case of all of the journals for which you also pay a per-page fee) this common error hurts. A two-page column becomes a six-page column. You learn this when the document arrives, but do you remember that you paid more for this document than you should have? 

Sure, this 200-percent increase in pages raises the copyright fee for this article only by $2, but $2 here, $2 there can grow into a pretty slush fund somewhere in the chain when thousands of articles are ordered each day. Maybe you should change jobs and work for the company where those extra fees land.

Are such mistakes in page numbers pervasive? It's hard to say what percentage of the more than 32 million records in the Infotrieve database display excess page numbers, but it caught my eye that records for four of my five 1998 CIL columns had wrong page numbers. In the case of other ITI journals for which Infotrieve charges a base fee plus a per-page fee, that is an 80-percent error rate that directly hits you or your client in the pocket. Interestingly, I couldn't find any article where the BL page numbers would have been less than the article really is. In all the cases the extra pages clearly were unrelated to the column, such as full-spread ads and news pages.
 

The Check's in the Mail
It would take more research time to test a representative sample from such a large database, but the implications are remarkable—and you can hardly help wondering what happens when Infotrieve tries hard to transfer the copyright fee to the publisher directly, to the Copyright Clearance Center, or to the author (if he is the copyright holder). The third alternative is easy to dismiss. This author has never received royalties from Infotrieve or any other document delivery service (except for a handsome lump sum from UnCover in a class action suit, settled out of court).

The publisher may be more lucky, even though it may be a tad more difficult to deliver and cash the royalty checks for CIL articles if they are addressed to Meckler Corp. in Westport, Connecticut (as the Infotrieve records suggest). Meckler was the publisher of CIL quite a long time ago, and may not even be able to determine whether the checks belong to someone else. Why? Read further to find out.

Even if the check reaches the publisher, this does not necessarily mean a happy ending, as the publisher may need to forward the royalty or part of it to the author (if he is the copyright holder). Well, this may not happen, as most if not all of the royalty payments are forwarded from the document delivery services to the publisher in a lump sum, for copies of many different articles. Under this current practice, the publisher can't even guess whose articles were copied and distributed.

As a copyright-compliant information specialist dealt such a deck of cards, is it time to "run"? Probably you should explore alternative document delivery options.

Going to the publisher directly may be more work, but you pay far less, and you help the publisher. It may then have funds for a decent deal with most of its authors to compensate them with a lump sum for reproduction and distribution rights, as ITI did with a progressive and fair offer for articles published after 2001. Now you know the other reason why I said "loyalty above royalty" with regard to ITI. It's good for the goose and good for the gander—as long as third parties charge appropriate copyright fees and remit royalties to the copyright holder. I am pursuing that aspect of the copyright fee conundrum, as it opens a new angle: the infringed copyright of the author who specifically retained the reproduction and distribution rights, but who still does not get royalties from document delivery transactions.
 

Postscript: Infotrieve
I ordered and received in 3 days the two-page document I mentioned in the column. The invoice repeated the same bibliographic data, indicating the page numbers as pages 34–39, even though it is on pages 34–35. I sent an e-mail to Infotrieve outlining the problem. They responded promptly, and after clarifying some issues offered to reimburse me $2 for the four extra pages (which I declined, knowing that this would be a lot of extra work for Infotrieve). However, we got stalled in the basic question. The rep claimed that I had indicated that I wanted pages 34–39, and that's what they had charged me for. I pointed out that users have no choice (even if they suspect that the information may be wrong) in specifying page numbers, but must click on the "Place in Cart" button in the bibliographic record presented by Infotrieve. Ordering is a totally automatic process. The users assume that the page numbers and thus the copyright fees provided by Infotrieve are correct, especially when they search the Infotrieve database. One mistake of four extra pages is not significant, but articles with excess page numbers in the ad-hoc test I made were excessive. Considering the volume of transactions that take place each day, and the proportion of journals where a base fee and per-page fee are used to determine copyright, this may be a large-scale problem, may undermine the customers' trust, and may cause them to be regularly overcharged. While our discussion stalled at this point, I trust that Infotrieve and its customers can work out some solutions for compensating those who have been overcharged, and that Infotrieve will make changes to prevent future overcharges.