Digital copies of Charlottesville’s daily newspaper dating to the 19th century are now online and searchable, thanks to a joint project between the University of Virginia Library and the Jefferson-Madison Regional Library.
The Daily Progress – Digital Edition is a powerful resource for researchers interested in local history, an example of a fruitful University-community partnership and a sort of pilot program for future large-scale digitization projects in the U.Va. Library, said Bradley Daigle, the library’s director of digital curation services.
The collection includes copies of the newspaper from its founding in 1892 through 1923, the span during which the paper is free of copyright restriction. U.Va. got involved when Daigle learned that the regional library hoped to digitize its microfilm copies of the newspaper.
“The Daily Progress collections here at U.Va. and at the Jefferson-Madison Regional Library are both heavily used, so the idea of making them available online was of interest to both of us,” he said.
Working with the regional library and the paper, U.Va. provided consultation and helped create the methodology by which about 60,000 images were digitized by a vendor and then put online. JMRL funded the project while U.VaLibrary supplied the microfilm copies and staff time.
The regional library wanted the images online by the end of the Charlottesville’s 250th anniversary year in 2012 – a deadline it met – and U.Va. integrated the digital images of the newspaper into its online collections. Users can browse the newspaper collection in its entirety or through the library’s online search tool, Virgo.
“This was an excellent collaborative town-gown opportunity,” Daigle said. “We’re both approaching librarianship from slightly different angles – we’re a research library, they are a public community library – but we found common ground in making this project happen.”
Though the collection is online, the project isn’t complete. Library staff hope to refine it by adding missing copies and plan to rely on community support to help make the collection more searchable and useful.
Users who come across an article on a particular person or topic can contact library staff to suggest that additional metadata – the information search engines use to identify relevant results – be added to the entry, so it will be more likely to appear if a future user searchers for that term.
So, for example, a user who came across an early-20th-century article referencing infamous former Charlottesville Mayor J. Samuel McCue could recommend that his name be added to the metadata for that particular copy of the paper, making it easier to find for future users interested in McCue.
User feedback has already been a help, said Barbara Selby, research and information services manager in the library who worked on the project.
“I received an email over the holidays from someone who had realized that an issue we’d identified as being from June 9, 1892, contained an article about the death of King Edward VII’s aunt, who actually didn’t die until 1923,” Selby said.
Selby discovered that a printer’s error in 1923 had caused that particular copy of the paper to be misdated on the masthead, so the date of the paper’s founding – 1892 – appeared as the date of the edition.
The library also hopes to use a technology called optical character recognition, or OCR, to digitize the text of each article contained in the collection. This would make each article searchable, and greatly ease the research process. But the poor quality of some of the newspaper images, and OCR’s limitations reproducing proper nouns and names, means that community support could be vital in refining the results, Daigle said.
“It’s really a living project, which is the fun part about it,” he said.
For the library, the project could help pave the way for future large-scale digitization projects. The library’s digitization services staff respond to requests from faculty and other researchers to create digital copies of manuscripts or other items housed in the library – such as manuscripts in the Albert and Shirley Small Special Collections Library – but the Progress project may hold lessons on the best ways to digitize and index entire collections, Daigle said.
“We’re using the Daily Progress as an example of how to take hierarchical data and integrate it with our other collections,” he said. “The library has over 5 million books, but we also have over 17 million manuscripts. Our manuscript collections are filled with primary source documentation, and a lot of them are the jewels in our crown, such as the records of Thomas Jefferson’s correspondence.”
Though many of these manuscripts have been digitized and published, many more have not, he said. Many libraries are wary of massive digitization projects because of justifiable fears of data overload, but the Progress project could provide valuable insights on the best ways to manage large digital collections.
“Is there a way we can take our primary source digitized materials and present it to our users in an integrated way? We’re not quite sure how it would look, but these are open questions that are being considered throughout the library,” Daigle said.
Such projects are contingent on the laws that govern intellectual property, such as copyright restrictions and fair use. In the case of the Daily Progress archive, the library hopes to add additional copies of the newspaper beyond 1923 to the archive, if it can secure the rights to do so.