Rice University Press went all digital in 2006 and will be closing completely in September. As noted in this Chronicle article, this is particularly sad news because Rice’s experiment with going all digital held so much promise for greater sustainability.

The Chronicle article mentions an important point that I hadn’t realized: that Rice had a catalog of only approximately 20 titles. This makes sense given the labor required to edit and publish an academic text – whether digital and/or in print. However with so few volumes, the long tail effect doesn’t have much of a chance of being sustainable.

I hope that the announcement of its closing is premature, and that Rice UP is given more time. If not, the experiment hasn’t been given enough resources and time to succeed. Rice UP needs a larger catalog and a more diverse portfolio (academic books, books digitized at Rice, collaborative projects with the library, collaborative projects with the bookstore for university history books, etc)  before the actual test of sustainability can even begin, much less complete. I hope that we’ll see this experiment continue with the necessary resources for it to be a valid test.

From the SNAC website:

Leveraging the new standard Encoded Archival Context-Corporate Bodies, Persons, and Families (EAC-CPF), the SNAC Project will use digital technology to “unlock” descriptions of people from finding aids and link them together in exciting new ways. We will:

• Create efficient open-source tools that allow archivists to separate the process of describing people from that of records.
• Create a prototype integrated historical resource and access system that will link descriptions of people to one another and to descriptions of resources in archives, libraries and museums; online biographical and historical databases; and other diverse resources.

Mark Sullivan, the UFDC/DLC/dLOC programmer, recently shared this information. It’s exciting to see that SobekCM (our digital asset management system, digital library system, and digital production tool set) is such a streamlined solution with so much functionality. There are seven projects which make up the SobekCM solution. In those projects, there are:

113,643 lines of code ( not comments or empty lines )
23,452 lines of comments
420 files
60 folders
544 classes ( 55 abstract classes, 1 windows form, 5 ASPX pages )
14 interfaces

The main two projects are:

1) SobekCM_Bib_Package which has all the code to represent digital objects, read metadata, write metadata, etc.. This is used throughout all the DLC/UFDC/dLOC applications.

36,554 lines of code
5,796 lines of comments
121 files
19 folders
165 classes ( 3 abstract classes, 1 windows form )

2) SobekCM_Library which does all the rendering, navigation, authentication, etc for the SobekCM library. This relies heavily upon the above library for reading and displaying of digital resources and is utilized by both the builder and the customization manager.

68,803 lines of code
13,825 lines of comments
251 files
30 folders
328 classes ( 52 abstract classes )
11 interfaces

This does not include the 22 separate javascript files of which eight are written by me and include 3951 new lines of code and 702 lines of comments.

3) While the main SobekCM web project is not strictly a library, it is the third project in the SobekCM solution. It is the first project which a user interacts with when entering the library. This project is actually very small, containing only about 1300 .NET lines. It does house the five web forms used in the application, although these forms are quite small and are just basically skeletons into which the SobekCM_Library renders HTML or controls.

4) SobekCM_URL_Rewriter is a tiny library which is essentially just a HttpModule for rewriting and translating the URL to allow for cleaner URLs.

5) SobekCM_Tools is a small library ( about 4000 code lines ) which contains additional classes for logging, interacting with the tracking database, and interacting with the Florida Dark Archive (DAITSS). This is kept seperate from the general library since this is not strictly involved in rendering the HTML but is used by some modules and is used with the SobekCM_Library and SobekCM_Builder libraries for building collection text indexes and loading new items through the Builder.

6) FileUploadLibrary ( written by Darren Johnstone ) is about 3000 .NET code lines and 5500 lines of javascript used for uploading data via HTTP with a real-time upload progress bar. Quite useful and cool library which was adopted with very few changes and worked quite simply. Highly recommended… ( http://darrenjohnstone.net/ )

7) SobekCM_Builder. In addition to these libraries/projects used by the digital library, this library is employed (along with the SobekCM_Bib_Package, SobekCM_Library, and SobekCM_Tools) for the builder software which runs constantly in the background on another server, loading new items which are deposited into network folders or FTP folder. It also updates and builds all static pages, OAI feeds, RSS feeds, and builds the text indexes. Additionally, it reads and loads all of the FDA ingest reports from DAITSS.

6793 lines of code
858 lines of comments
26 files
3 folders
37 classes

Press Release:

For immediate release: August 4, 2010

For more information, contact:
Karla Strieb
Association of Research Libraries
202-296-2296
karla@arl.org

ARL Promotes Member Use of Large-Scale Digitization Principles

Washington DC— The Association of Research Libraries (ARL) Board of Directors unanimously voted on July 26, 2010, to endorse a set of nine principles to guide vendor/publisher relations in large-scale digitization projects of special collections materials, recommended by its Transforming Special Collections in the Digital Age Working Group. The Board’s vote strongly encourages ARL member libraries to refrain from signing future agreements with publishers or vendors, either individually or through consortia, that do not adhere to the principles.

The ARL Board recognizes that research libraries are increasingly finding that large-scale digitization of special collections materials involves partnerships with commercial vendors and publishers, and that those partnerships should be governed by principles that protect special collections materials and promote the broadest possible access to digital versions of them.

Special collections often include valuable and unique materials, but also incur special responsibilities for their stewards. Digital access to special collections materials has become important in revealing hidden materials and promoting humanities research, and ARL member libraries often require appropriate collaborations and partnerships to implement large-scale digitization activities.

The nine principles address issues including implications of the distinctive character of special collections, the need for libraries to retain their own copies of the products of digitization projects, the importance of promoting broad access to digitized collections, and concerns regarding the collection of data about users of digitized collections.

“I am thrilled that the ARL Board has endorsed these principles, which encourage research libraries and archives to provide digital access to special collections while safeguarding institutional interests and promoting broad public access,” said Anne R. Kenney, Chair of the Working Group. “At Cornell, we plan to draw on the framework to improve many terms in our negotiations with vendors, including to shorten embargo periods on all our digitized collections to five years or less.”

To view these principles, please visit: http://www.arl.org/bm~doc/principles_large_scale_digitization.pdf