▲ 86 points
back
9 comments
Kudos for having their hearts in the right place.
No kudos for presentation. I've tried to look at two works, typewritten MSS by Vachel Lindsay and Sara Teasdale, and the text display is awkward. The initial presentation is too small to read. Looked long for a zoom control. Finally tried "full browser" link which does display the current page with a zoom control and the ability to drag that page around in a small frame to read it.
While it is nice that UT is doing this, it is sad that each uni has to do its own thing. Can there be no one virtual repository where all such collections -- or at least their catalogs -- could be amalgamated? It's like the bad old days of physical libraries: the scholar has first to find out which collection has the thing she needs, and may never learn that it exists. If they don't want to deal with google books, how about using TIA as a front end?
Edit: ...and then I looked at the URL, hrc.contentdm.oclc.org and thought, huh, that doesn't look like UT or EDU... turns out oclc.org is something like I was just asking for, a front end content distribution platform for multiple libraries. OK then. TIL and all that.
It always irritates me when these sorts of sites decide to implement their own often quirky document-viewing web app and hide everything behind it instead of just providing a direct link to a PDF or similar. I mean, if you're going to have this material freely available anyway why not make it easy to access?
IMHO archive.org is a good example to follow; I know their recent redesign is somewhat controversial, but they have not changed the fact that it is easy to get to the content directly, and their URLs are also mostly descriptive.
I've been working in this space for the last 6 years. One of the challenges of these efforts is the funding model -- a lot of the work is funded through grants. If a grant funds a project for 5-10 years, maybe 80-90% goes towards conservation and digitization of the material, leaving a relatively small amount of funding for sharing the material online. As a result the websites (and related tech) tends to be fragmented even within the same organization. Usually the folks building the website are brought in as short-term contractors who tend to use the tech of the day.
In the last couple years there has been a renewed interest among culture heritage organizations to create and formalize technology standards. International Image Interoperability Framework (http://iiif.io/) is one of those efforts and has resulted in a couple of good open projects. Of course there are many more budding standards and open/closed SaaS and COTS products, but the industry is improving. As more folks use these systems, they become more appealing to the folks offering grant money, more grants/visibility results in better offerings and more unification/exchange between organizations.
It's always exciting to see these cultural heritage project pop up on HN.
My university does something similar with its Digital Library and its related projects. The main repository hosts around 140,000 objects which can be accessed alongside metadata and in multiple formats here:
http://digital.library.unt.edu/
The Django app and libraries running the site are built and maintained in-house, and are open source:
https://github.com/unt-libraries (the Django project/app itself is 'coda')
This library is famous, I had often used it online even years ago. When I left the US I donated a lot of material from Asia gathered over a decade or so to their excellent map collection http://lib.utexas.edu/maps/ (much of which is digitized) since shipping anything out of the US is ridiculously expensive.
Technically, the Perry-Castañeda Library and the Harry Ransom Center are only loosely related, if that. (The HRC is the big fortress/vault-looking building at 21st and Guadalupe and the Pickle is the warm and welcoming architectural monstrosity at 21st and Speedway.)
But the HRC, it's forbidding exterior notwithstanding, is pretty nice, too.
Cool! Looks like they're free to use with attribution.
Direct link to the collections: http://hrc.contentdm.oclc.org/#nav_top
Nicely done -- they admit that they don't have a copyright where appropriate, and ask for attribution, without requiring it.
UPEI has digital library system called Islandora[1] that does this sort of thing. I worked for them for about 6 months. It's a set of Drupal modules that interact with the Fedora[2] Repository (not the OS). They're pretty successful, with hundreds of installations all over the world. They have been scanning hundreds, if not thousands of books, newspapers, and maps so that they can be stored digital and accessed freely.
One example I think is pretty interesting is Island Newspapers [3] where they have scanned every issue of a newspaper in PEI, Canada called the Guardian going back to 1890.
[1] http://islandora.ca/islandora-installations