CrossRef recently published new guidelines for assigning DOIs to books — including reference works — and a revised fee structure for publisher deposits.

The new parameters advise publishers to deposit DOIs at the chapter/entry level. The 2010 pricing structure, which has significantly reduced fees for backfile and intra-ebook content deposits, supports their recommendation. Assuming publishers adopt this direction (Springer already has — see their new SpringerLink platform), these moves may have far-reaching, long-term implications for e-book functioning and interoperability.

CrossRef identifies the following aims:

  • Maximize reference linking among books, journals, and conference proceedings
  • Enhance the discovery, visibility, and usage of book content
  • Enhance the user’s experience through improved functionality
  • Enable the creation of a book citation reporting mechanism which would give book content the visibility, credibility, and metrics that journal content has

There have been many champions of entry-level metadata, some of the more prominent in connection with Reference Universe. Advocates have been acutely aware of the factors limiting e-book functionality, stemming from the absence of coherent e-book tagging and linking standards.  Reference e-books have suffered particularly in environments external to publishers’ own platforms, because these rely on deep-level tagging to enable discovery and use of the of content within.

Assuming that publishers quickly embrace the new book DOI recommendations, multi-disciplinary reference may yet regain its “royal status” (see David Tyckoson’s presentation for Booklist on “The Rise and Fall of Reference“) in the digital information environment — or, at least, to get back to the table as a relevant, high-use player.

Publishers, particularly those who publish journals, have been cognizant of the potential for DOIs in e-book linking.  However, with hundreds of thousands of backfile DOIs at the chapter and entry-level to deposit, Cross Ref’s pricing has been a gating factor — until now.

It’s easy to envision that, in an environment in which patrons have access to dashboards that help them create and manipulate personalized information folios — e.g., ebrary’s new DASH, which stands for “data sharing, fast” — more granular linking will provide a significant boost.  No one wants to add an entire encyclopedia to their folio, but individual articles make a lot of sense.

Are there unanswered questions? Yes, particularly pertaining to links for titles hosted in non-primary aggregations. E-books may be hosted in 10 or more different locations and formats. With collections of hundreds of thousands of hosted titles in their repositories, e-book aggregators may lack the incentivse to embed granular DOIs that link out to publisher sites.

However, this is an assertive move by CrossRef to help make e-book content — including reference — more parsable, interoperable, and linked.

Reblog this post [with Zemanta]


7 Thoughts on "E-books Get a Leg Up from CrossRef"

Linking to the interior of large documents is potentially very important, so this is great news. On a futuristic note, at OSTI we are studying the concept of user defined links. In addition to being an aggregator, OSTI publishes lengthy DOE research reports at

We call the idea “user anchoring,” or more informally “doc blog” because we envision it being used in science blogging. We have already added a comment feature that lets users comment on any given report.

The basic idea for doc blog is quite simple. The user navigates to the point in the document they want to share. They execute the anchor command and the system places an anchor at that point, creates a URL for it, and gives the URL to the user. The user can then post or email the URL, to direct others to that part of the document. This is much more efficient that saying “go to paragraph 5 on page 68,” or some such, especially if one wants to talk about several sections at once.

User anchoring cannot presently be done, as far as we know, but we think it might be feasible for documents in XML format. It would be wonderful if it could be done in HTML or PDF format. If anyone else is working on, or knows about, similar work I would love to hear about it. (

I think this is brilliant. Especially useful for longer documents, but even for shorter documents it will make better use of readers’ time. I have often been frustrated to click on a link only to find that it’s up to me to locate the cited passage. Very cool. It could prove to be the best thing since the ‘permalink’ — and even friend of same.

interesting idea, but I’m not sure why you think user-targeted links can’t be done- it can, in javascript/html environments.

Thanks for the info. We have been looking at document formats like PDF, PS, PPT and Word because that is what our present collection of 250k+ report docs is in. I will pass this on to our techies, who understand such things. Has anybody actually done this in Java/html, that you know of?

Having an html version makes sense to me, but I am just a function person. Some of my proposals make the techies laugh.


To answer your question about permalinking to XML, yes, you can, and it’s something we’ve been working on for our legal customers. A good example you can check out now is Norman Walsh’s site (norman (dot) walsh (dot) name). Put your cursor at the end of a paragraph on one of his posts and you’ll see a gray paragraph mark. It’s a permalink to that paragraph.

Comments are closed.