After a year and a half of planning and development, Springer Science+Business Media (www.springer.com), an international scholarly publisher based in Germany but operating in 20 countries, has launched SpringerImages (www.springerimages.com). The massive collection of 1.6 million scientific, technological, and medical images includes photos, tables and figures, charts, graphs, histograms, and other illustrations. Although covering all scientific subject areas, some 61% of the collection focuses on medical and life sciences. Drawing on its own vast collection of content, Springer provides multilayered, in-depth indexing. Subscribers can use the material liberally as long as they do not use it for direct commercial purposes. In an interesting development, SpringerImages includes a small but growing collection of open access images, which are available to anyone, no registration required.
In the STM arena, Springer Science+Business Media publishes about 2,000 journals and more than 6,500 new books a year, as well as boasting the largest STM ebook collection in the world. It also owns Images.MD, Current Medicine Group (CMG; acquired in 2005), and BioMed Central, a leading open access operation acquired in 2008. All these operations provide a flow of images into the new service. The nonmedical/nonhealth sciences content includes chemistry, computer science, economics and management science, education, engineering, environment, geography, geosciences, humanities and arts, material science, mathematics, physics, psychology, and social sciences. Counts of the number of images in different sci-tech categories appear frequently on the website.
Compared to the size of the entire collection, the subset of open access images is minuscule-only about 30,000 items at this point. However, Brian Bishop, director of product development, indicated that the open access images should increase significantly soon, due to an arrangement with a laboratory at Yale University run by Michael Krauthammer. "Michael has already gathered a big collection of images from the open web," said Bishop. "We launched with Images MD, but in the next few months, we'll be adding over 100,000 images. They're [Yale] continuing to grow their database." The open access images are truly open. Bishop said that there are no charges. "They are drill downable for Scirus, Google Scholar, anyone."
In a practical sense, the open access image subset allows users to experiment freely with SpringerImages and its features, forming a kind of marketing function. Any and every user can search the entire file, but only subscribers can view and download the full images. It's easy to see the potential marketing tactic here. Dangle the subscriber-only images before users who came to get a freebie and they may end up scurrying back to their librarians and pressing them into licensing the full service. However, Stuart Koziol, product manager for SpringerImages and eproduct manager for SpringerProtocols America, pointed out that users can narrow their inquiries to searching only open access or free content.
The indexing built to support retrieval draws on captions, keywords, sentences referring to the images even when the referrals may occur far away from the image in the article (e.g., "See Figure 2 ..."), and index terms assigned to the overall article. However, according to Bishop, the service does not use a thesaurus. Success depends upon the rigor and imagination of the user in creating an effective search strategy. For example, if users seek information on a specific drug, they might have to use both the chemical name and the brand name. Search features include an advanced option that can reach sources, publication dates, image providers, image color, author, volume and issue numbers, image type, and DOI (digital object identifier).
Though SpringerImages distinguishes itself by providing the full images extracted from articles and books and its creators promoting the service as offering users quick access to the most valuable parts of scholarly documents, the service does not cut the mooring lines tying images to their source documents. Image details appear with full bibliographic citations, those sentence extracts used for indexing, and-if licensed through SpringerLink-direct access to the journal articles or ebooks. In the latter case, full-text access is only one click away. Users can export images in PDF or PowerPoint format.
Users are classified into three categories: unregistered users, registered users, and subscribers. The unregistered users can search the entire file but can only retrieve open access images. Registered users who have set up an account can add user-generated keywords to individual images, each of which has its own permanent link identification. Subscribers can download the images and use them "for almost all non-commercial purposes." In other words, prohibitions on usage focus on profiting from the image directly. Since Springer certainly intends to license SpringerImages to corporations and other profit-making concerns, it does not worry about profit making in general. If a subscriber does want to use an image for a commercial purpose, e.g., in coursepacks sales or company catalogs, the service provides a feature for requesting permission.
Springer has been very attentive to copyright issues in general. According to Koziol, the company has created algorithmic techniques for finding images not created by authors. Seeking phrases such as "copyrighted by" or "courtesy of" or other publisher names and attributions, Springer has then filtered out the images, blocking them from entering the new service. To searchers, this means that when you find images from an article in SpringerImages, you cannot necessarily find all the images that appeared in the original article. Open access images carry a warning label on their image-details pages: "This image is copyrighted by The Author(s)." If users have any specific concerns about the copyright of particular images, they can state those concerns by clicking on a "Report copyright" link at the bottom of each image.
SpringerImages comes in two packages for licensing subscriptions: the entire collection or just the Medical and Life Sciences portion. Different pricing packages are available for Standard, Academic and Corporate Single-Site, Multi-Site and Consortia. According to Koziol, "Pricing is dependent on the size of the institutions and the research intensity. It can vary from a low of a few thousand to a few tens of thousands for an extremely large institution. Consortial deals can alter prices dramatically. The actual price varies depending on a number of factors." He did indicate that an individual license was available for $595 for the whole collection or $395 for just medical and life sciences.
SpringerImages is not the only service supplying access to sci-tech imagery. For several years, ProQuest CSA has provided CSA Illustrata (www.proquest.com/en-US/promos/csaillustrata.shtml). As a database aggregator, it can reach out to multiple publishers, such as a deal with Elsevier. It even established a connection with Springer last year to add deep indexing for 11 Springer journals. As to how durable or expandable that arrangement will remain, neither Koziol nor Bishop had any comments. Although an aggregator may cover a broader range of sources, a publisher-based service has more flexibility in what it can offer for its own content at least. For institutions looking at licensing images, it might be worth investigating both. The size of the collections is similar, but CSA Illustrata concentrates on natural sciences and technology.
SpringerImages is on an expansion path. Springer expects to add as many as 500,000 images this year. Bishop indicated that the company was investigating some Web 2.0 approaches, such as user-generated images, e.g., those generated at laboratories but not in connection with specific books or journals. "We also have some video in supplementary material," said Bishop. "But videos are very scarce and tricky." One interesting note-Both Koziol and Bishop had a scheduled meeting moments after my last interview ended with Anurag Acharya, the man behind Google Scholar. Under discussion is finding ways to connect the two services. I wonder how that exchange will turn out. Stay tuned!