EBSCO in Cahoots With Harvard Business Press

A controversy is starting to pick up in the business librarian community -- primarily in the U.K. it would seem -- regarding the licensing demands of Harvard Business Press (HBP) for the inclusion of Harvard Business Review articles in EBSCOhost. HBP content in EBSCOhost carries a publisher-specific rider that says use is limited to "private individual use" and explicitly bars the practice of putting "deep links" of articles from EBSCOhost (so called "persistent links") into learning management systems. In my words, HBP is attempting to limit access to its content in EBSCOhost to those who find it through the serendipity of searching. And now HBP is going after schools that are using persistent linking, and this raises all sorts of troubling questions.

The only visible sign of the publisher-specific rider (that I can find) is text appended to the end of each article from Harvard Business Review in EBSCOhost (copied from a post by Paul Pival):

Harvard Business Review Notice of Use Restrictions, May 2009 Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any other means of incorporating the content into course resources. Business licensees may not host this content on learning management systems or use persistent linking or other means to incorporate the content into learning management systems. Harvard Business Publishing will be pleased to grant permission to make this content available through such means. For rates and permission, contact permissions@harvardbusiness.org.

It was Pival's next statement, though, in that same post where he relays a conversation with a colleague at a different institution that raised my eyebrows (emphasis added):

He also mentioned that HBSP [Harvard Business School Publishing] had leaned on his school and when they decided not to pay, EBSCO turned off the ability for them to create PURLs for that publisher.

Huh? How does HBP know that deep links are being used in that way?

EBSCO Information Services Website Privacy Policy

Pival asks "So how does Harvard BSP know whether a given link is being used for 'private individual use' or for within electronic reserves, electronic course packs, a syllabi, or within a learning management system?" The answer is probably the HTTP "referer" ((Yes, they know they misspelled the word "referrer")) header. With every page, your web browser sends the address of the page you came from to the remote web server. You can see this with the BrowserSPY service. If you follow that link, you'll see that the page you came from was this DLTJ page (http://dltj.org/article/ebsco-hbp/ in the HTTP_REFERER row). Whether you know it or not (or have blocked the HTTP referer header before it gets to the remote server), you leave these traces of where you came from with every web request you make. So what I'm surmising is that the EBSCOhost servers record and process the HTTP referer information for deep links, and can see patterns when a number of people come to EBSCOhost from the same web page -- that web page is probably a reading list in a course management system, an electronic reserves page, or something similar. ((This is probably the kind of business intelligence that EBSCO already collected about how its service is used. I doubt that the recording and processing of HTTP referer headers is something specific to HBP content.))

So now that we know how it is probably happening, we can ask "Is there anything in EBSCO's terms-of-use that permits them to share usage information with content suppliers?" The answer would seem to be "probably yes". The place to look is the EBSCO privacy policy. Here is an extract from the policy dated December 26, 2006 (the current version as of the time of writing). The HTTP Referer header seems to fall in the category of "Non-Personal Identifying Information:

B. Collection of Non-Personal Identifying Information
We collect and use non-personal identifying information, including IP addresses and web server log files to track trends, administer the website, track user movement, and gather demographic information. We use this non-personal identifying information in the aggregate. We do not combine these types of non-personal identifying information with personal identifying information [a term defined earlier in the privacy policy; e.g. user's name, address]. We may also share aggregated demographic information with our business partners, sponsors, advertisers, and companies that control, are controlled by, or are under common control with EBSCO Information Services.

The HTTP referer information comes from the web server log files (it is a byproduct of running a web server), and HBP is probably considered a "business partner", so HBP can make a request of EBSCO like "Give me all of the HTTP referer addresses that link to HBP articles, and the number of times each referer addresses is used." It would be pretty simple then for HBP to determine which institutions were deep linking directly to Harvard Business Review articles.

EBSCO Publishing License Agreement (Terms of Use)

So the next logical question might be "Can Harvard Business Publishing create these added restrictions?" The answer to that question is most definitely "yes". As librarians, we may not like the fact that publishers can put added restrictions on content in our aggregation databases, but EBSCO's Terms of Use certainly allow for it (emphasis added):

C. Licensee and Authorized Users agree to abide by the Copyright Act of 1976 as well as any contractual restrictions, copyright restrictions, or other restrictions provided by publishers and specified in the Databases. [...] Publishers may impose their own conditions of use applicable only to their content. Such conditions of use shall be displayed on the computer screen displays associated with such content. The Licensee shall take all reasonable precautions to limit the usage of the Databases(s) to those specifically authorized by this Agreement.

And the repercussions of violating the Terms of Use? They are spelled out in the "Termination" section:

A. In the event of a breach of any of its obligations under this Agreement, Licensee shall have the right to remedy the breach within thirty (30) days upon receipt of written notice from EBSCO. Within the period of such notice Licensee shall make every reasonable effort and document said effort to remedy such a breach and shall institute any reasonable procedures to prevent future occurrences of such breaches. If the Licensee fails to remedy such a breach within the period of thirty (30) days, EBSCO may (at its option) terminate this Agreement upon written notice to the Licensee.

B. If EBSCO becomes aware of a material breach of Licensee's obligations under this Agreement or a breach by Licensee or Authorized Users of the rights of EBSCO or its licensors or an infringement on the rights of EBSCO or its licensors, then EBSCO will notify the Licensee immediately in writing and shall have the right to temporarily suspend the Licensee's access to the Product(s). Licensee shall be given the opportunity to remedy the breach or infringement within thirty (30) days following receipt of written notice from EBSCO. Once the breach or infringement has been remedied or the offending activity halted, EBSCO shall reinstate access to the Databases. If the Licensee does not satisfactorily remedy the offending activity within thirty (30) days, EBSCO may terminate this Agreement upon written notice to the Licensee.

I haven't seen mention of EBSCO going so far as to terminate access to all or part of EBSCOhost, but there are indications that EBSCO is disabling the deep links to HBP content. There are reports of HBP asking libraries to pay an additional fee to EBSCO for the ability to have deep linking for HBP content. By one account, a UK university might have to pay an additional £15,000 (about $25,000 at current conversion rates) to "create persistent links for use in teaching."

Do We Have to Take It?

So that is the technical and the legal perspectives on this controversy. The final logical question is "Do we have to put up with it"? Andy Priestner, Judge Business School's Head Librarian (at Cambridge University), asks several good questions about this twist of electronic article publishing and distribution. In an article a few months ago, Jon Rochkind talks about the eroding of rights associated with the 'first sale doctrine' in U.S. Copyright.

I'm of mixed minds about the issue. HBP can license its content however it likes, and it probably does make a lot of money from reselling articles ((In its "Academic Permission Form", HBP says the cost of permission to reproduce and distribute articles is $3 per copy. In addition, high-quality hard copies can be purchased from HBP for $3.42 each in quantities of 10 or more.)) in course packs and the like -- revenue that is lost by the sorts of deep linking that can happen into a journal aggregtor's service (like EBSCOhost). On the other hand, the added restrictions by the publisher are not clearly spelled out on the page where the permanent link is displayed. (See the figure to the right; click on it to open up to a larger screen capture of an EBSCOhost record display.) Unless an instructor knows of the special conditions through some other channel (or reads the mind-numbing text at the end of the article PDF -- mind-numbing text that never seems to change from article to article, except in these special circumstances), they won't know that they are doing something that violates the publisher-specific terms of use. And more to the point, the library -- the one paying the bills -- won't know that terms have been broken until the publisher comes knocking on the door asking for more money. And if the library balks, who looks like the bad guy?

And what is EBSCO's role in this? Isn't a library's contract with EBSCO, not Harvard Business Publishing? Is EBSCO earning more revenue from this HBP license requirement to enable deep linking to article content? If so, isn't that just an incentive for EBSCO to do the same with other high-profile publishers?

I don't have answers to these last questions, but I do get a very uncomfortable feeling.

Update

June 28, 2009: I missed this post from Meredith Farkas when I was looking for reactions to Paul Pival's post. She says her university has had their deep links into EBSCOhost for Harvard Business Review turned off, although it wasn't for links in the course management system.

August 28, 2009: An opinion piece in Library Journal from Chris Flegg, Bodleian Business Librarian of the Saïd Business School at the University of Oxford (UK) goes into more depth, and includes Oxford's response -- not to pay and have deep linking turned off.

The text was modified to update a link from http://www.caul.edu.au/datasets/hbr2008course-use.pdf to http://replay.waybackmachine.org/20091003235023/http://www.caul.edu.au/datasets/hbr2008course-use.pdf on February 11th, 2011.