I would echo much of what Chuck said in his email. We don’t curate to the degree you’re suggesting, in part because we don’t have time, but primarily because of the OCR possibilities. But, we’re also not dependent on the administration to pay for storage, and that was/is a game changer in keeping our program alive. Just to be clear, we have little administrative support for our digital newspapers, so we’ve had to find creative ways to save our collections and take in new ones.

We use the Internet Archive for storage and access. We’ve got a Blacklight portal to get users to the materials in IA. (This is undergoing a facelift at present) The University also has a deal with Google for unlimited storage, so we’re using that for dupe storage. Between Google and IA (especially with a Canadian mirror) we’re now offered some geographic distribution. From both locations we can push and pull data without incurring costs. The Libraries is working on a repository with on-campus storage. Once we’ve been given the costs for that we will probably add the newspapers to that repository for obvious reasons.

Our only real costs remain what a public library, for example, might want digitized. They have to find a way to afford a vendor, or we have to work with our special collections center to digitize a few reels every so often. Sometimes we can do that, sometimes not.


Kopana Terry
Oral History Archivist:
Curator of Newspapers:<>
Special Collections Research Center
Louie B. Nunn Center for Oral History
105 Margaret I. King Library
University of Kentucky
Lexington, KY 40506-0039
tele: 859.218.1423
fax: 859.257.6311
email: [log in to unmask]<mailto:[log in to unmask]>

On 2/7/17, 1:17 PM, "Data, API, website, and code of the Chronicling America website on behalf of Kirkham, Christine" <[log in to unmask]<mailto:[log in to unmask]> on behalf of [log in to unmask]<mailto:[log in to unmask]>> wrote:

Hello, all. As we grow our local CHRONAM instance,, we are struggling with the question of content.

We provide the work of digitizing, uploading and hosting newspapers using a cost-recovery model. Clients come to us with a title and date range in mind, and we provide them with a quote to digitize, upload, and host the content. Then, when they have raised the necessary funds, they return to us and we launch the project.

Some local libraries raise funds to digitize a particular date range of a small-town weekly. Often they wish to start with the newspaper’s first decade, in the period 1890-1920. On examining the microfilm, however, we may find that the paper was rife with readyprint, reprinted content from other state or regional papers, reprinted national and international news, and local and national ads. The local, Montana-specific content of these newspapers is almost always less than 50% and sometimes lower than 10%. For this reason, we have to question the wisdom of including these titles in our repository.

We certainly try to have frank discussions with these clients and steer them toward alternative titles that would consistently populate search results with Montana-created and Montana-centric content. However, we find that clients are resistant to this feedback and/or beholden to donors who simply want to see the early decades of the town’s first newspaper online.

As you know, the cost to store this data long term and the cost to host it on the web (we are currently deployed on Amazon Web Services) is not insignificant. We’re wondering how you are approaching this, that is, how do you balance the requests of paying clients against what you deem to be the actual research value (or lack thereof) of a proposed run? Have you declined to accept self-funded projects for this reason? Do you have a curation policy you’d be willing to share? We’d be grateful for any and all feedback.


Christine (Tina) Kirkham
Digital Services Group
Montana Historical Society
[log in to unmask]<mailto:[log in to unmask]>