When talking about Open Access to technologists, developers etc. the focus quickly shifts to repositories as well as the issue of automated metadata exchange. Essentially repositories are huge document stores that allow for the systematic description of documents with metadata. Additionally contemporary repository systems like DSpace, Fedora, Opus etc. usually provide a standardized OAI-PMH interface allowing data aggregators to harvest data from the repository and to include this data into their own catalogues. That’s an important feature which comes at considerable cost with the above mentioned systems. Even though they are Open Source they are designed for huge collections of Open Access publications and are thus rather complex. Installing and maintaining them easily becomes a full time job.
For small and medium publishers in the area of Open Access DSpace and the like are just too big, too powerful, and too complicated. Thus, we searched for a different solution. And we found one.
Looking for an alternative one quickly ends up with Open Monograph Press (OMP), a software developed by the Open Knowledge Project for Open Access Book Publishers. Similar to Open Journal System OMP serves as a one stop solution for submitting, reviewing, editing, producing and publishing books. In the end OMP even offers an OAI-PMH interface for metadata exchange of published books. Even though this software is quite powerful and well suited for many contexts of use it has its limitations especially if a publisher dislikes the public face of OMP: it’s web front end. That’s exactly the problem we ran into in our experimental publishing project “meson press by Hybrid Publishing Lab”. Looking for a more contemporary, flexible and aesthetically appealing solution for presenting our books online we quickly abandoned OMP and switched to WordPress. For us the new website of meson press, which has been released in the past month is the proof that we made the right decision.
Choosing WordPress over OMP, however, causes new problems, because WordPress just isn’t designed for standardized metadata management. It’s a blog software not a repository and an easy to use and up to date OAI-PMH Interface to WordPress does not exist. Since we still wanted to allow metadata harvesters to access our Open Access publications using the OAI-PMH standard we had to find an alternative. An obvious solution would have been to turn back to OMP and to use the software just as backend for our editorial workflow, production and metadata dissemination. We tried (honestly!), but we failed since OMP again turned out to be too restrictive for our purposes and needs.
Instead of using a software designed for Open Access book publishing and metadata dissemination we ended up using Omeka, that is, an Open Source software designed to assemble archival collections and to create online exhibitions based on these collections. Omeka is a lightweight content management system that is almost as easy to install as WordPress. Developed by the Roy Rosenzweig Center for History and New Media Omeka is a well maintained software that offers an easy to use interface for adding items to the database using Dublin Core metadata that can be harvested via an OAI-PMH interface.
Despite this advantage the Dublin Core metadata set that Omeka relies upon isn’t well suited for describing books. It consists of rather general descriptors such as “Title”, “Creator”, “Description” that can be used for ascribing metadata to a wide range of heterogeneous resources. As a result users need to decide on how to make use of the metadata set and comply to these rules while adding books to the database, e.g. adding author names to the Dubin Core field “Creator” or the year a book was released to the field “Date”. Even worse Dublin Core does not allow for distinguishing between titles and subtitles. Therefore the user would need to add titles and subtitles to one field separating them with a standardized separator like a colon. Since Omeka allows users to quite easily add new functionalities to the system by creating a plugin we decided to do exactly this: We created a simple but quite helpful plugin called “BookCore” that adds a specialized metadata set for books to Omeka consisting of metadata fields like “Title”, “Subtitle”, “Author/Editor”, “Year Published”, “Series”, “Blurb”, “ISBN Print”, “ISBN PDF” etc. Once an item is stored these metadata are mapped on the Dublin Core metadata set. The plugin thus allows users to easily add books to the database and ensures that Dublin Core metadata harvested from Omeka are consistent. The plugin is released under an Open Source license and can be downloaded from Github.
Install this plugin to turn Omeka into a lightweight repository for Open Access book publications suited for small and medium publishers!