“It [hemoglobin] is a good substance from the standpoint of a chemist, because of its availability. All you need to do is to catch somebody, introduce a hypodermic needle and draw out a sample of blood. A standard victim of this practice, weighing perhaps 120 pounds (it’s easier to catch them small!) contains in the red corpuscles in his blood one and two-tenths pounds of hemoglobin.”
– Linus Pauling, 1966.
Some reasonably big news to share today. As announced here, we have launched a revised and expanded version of our 2005 release “It’s in the Blood! A Documentary History of Linus Pauling, Hemoglobin and Sickle Cell Anemia.”
Similar to the revised version of our “Nature of the Chemical Bond” documentary history website, which was launched this past February, the “second edition” of “It’s in the Blood!” contains a ton more content: the final tally runs to 53 new letters, 458 pages of added manuscripts and papers, 18 new pictures and 11 new audio and video clips. The metadata for all of the site’s content is drastically improved as well — a fact that is most immediately evident on the various Key Participants pages, which have been transformed from rather spartan affairs to content-rich resources like this page devoted to Harvey Itano.
Aside from the self-evident benefits of adding more content to our pages, revising the older documentary histories has also prompted our digitization work more in the direction of a uniform METS-based platform. We’ll talk a lot more about them at a later time, but for now it’s sufficient to define METS records as all-in-one containers for digital objects.
We use METS (Metadata Encoding and Transmission Standard) and MODS (Metadata Object Description Schema – both are flavors of XML) not only to describe a scanned item in a qualitative sense, but also to define how the item displays on a page.
For example, the METS record that “powers” the hemoglobin molecule above includes an internal i.d., the date and creator of the record, the image caption and it’s copyright data, the creator of the image (Roger Hayward) and any individuals or organizations associated with it. (Linus Pauling, in this case, since the drawing was published in Pauling and Hayward’s The Architecture of Molecules.) The record also stores the date of the item’s creation (1964), and the genre type of the original document. (We use the Library of Congress’ Basic Genre Terms for Cultural Heritage Materials as our genre authority. The Hayward item is defined by BGTCHM as an “illustration.”)
The METS record also defines certain display characteristics that are then interpreted by the XSL stylesheets that build our HTML pages. Again using our hemoglobin molecule as an example, the METS record which defines the object’s output declares that it can be displayed at one of four different sizes. The 150-pixel width display is used for all images inserted as Narrative sidebar images (Hemoglobin is on this page), as well as all images aggregated onto a given All Documents and Media index page. (Hemoglobin is about 3/4 of the way down the Pictures and Illustrations index.) A 400-pixel width version will be used in a revised version of our “Linus Pauling Day-by-Day” project, which we hope to launch later this year. The 600-pixel width “reference images” display like this, and the 900-pixel width big kahunas look like this.
METS records take a while to create, but the payoff is well worth the effort. The flexibility that METS provides both within and across projects is of huge importance to us — when building really big websites and/or multiple websites with subject matter that tends to overlap, (the documentary histories, the Day-by-Day calendar and the Pauling Student Learning Curriculum, e.g.) it is way more efficient to be able to describe an object once but use it again and again.
Right now, “Linus Pauling and the Race for DNA” is the last of our documentary history sites still requiring METS attentions. Once it’s revised, we’ll be able to start thinking concretely about providing different types of portal views and search tools for our growing METS cache (well over 3,000 records currently), an eventuality that promises a whole new range of possibilities for our entire digitization workflow. But that’s a different topic for a different day…