Creation of the SDL
The Synesthesia Digital Library (SDL) project grew from our shared interest in synesthesia as a unique and fascinating way of experiencing the world. Defined as the intermingling of one or more senses, synesthesia is only now beginning to be understood by the medical and psychological communities, although its effects have been written about and discussed for centuries. Many synesthesia resources can be found on the Internet, whether through YouTube, popular news media or personal websites. However, we decided there was a need for scholarly information on synesthesia that was both reliable and would appeal to casual users, including people with synesthesia, researchers, psychologists, artists, musicians, educators, students, and therapists. Since many well-known artists have or had synesthesia, we decided to create a digital library with biographical sketches of 14 famous synesthetes, complemented by explanations of some types of synesthesia and a list of additional scholarly resources for further research. To make our collection engaging for a wide audience, we enlisted the talents of Grant’s friend and illustrator, Jon Stich, who illustrated the synesthetes and synesthesia types.
The goal of the SDL is to provide an aesthetically pleasing and reliable source of information on synesthesia for a broad audience, and to spark greater interest in the topic.
The materials used were: 72 dpi JPEG copies of Jon Stich’s illustrations, created from archival TIFF scans; extensive metadata on the synesthetes and synesthesia types; and scholarly resources. While there are hundreds of scholarly articles and books on synesthesia, we selected 40 resources to serve as a representative sample of the existing literature.
The basic Dublin Core elements were found to be the best fit for the SDL collections, since they are both flexible and allow for interoperability with other collections. The SDL is comprised of 3 distinct collections: Famous Synesthetes, Synesthesia Types, and Additional Resources. Three separate CSV uploads were used to import collection content and enabled us to use only the Dublin Core elements that best fit each section. We used shared Google docs spreadsheets to enter the metadata. This helped with consistency and allowed us to work through questions about how to use some of the elements, in particular the 'Subject' field. For the biographical sketches and synesthesia types, the metadata elements are: Title, Subject, Description, Creator, Contributor, Rights, and Identifier. For the scholarly resources, the elements are: Title, Subject, Creator, Source, Publisher, Date, and Identifier. We also used HTML codes in our CSV file for italics and special characters. Because we used the Google docs spreadsheets, exporting to CSV was very easy.
Deciding how to represent each synesthete’s birth and death dates presented a challenge at first. We tried using Omeka’s “Person-type” metadata, and very much liked the additional information this allowed us to add. However, our concerns about interoperability led us back to the original Dublin Core elements. We used the Library of Congress Name Authority Headings for each of the famous synesthetes since these often include birth/death dates and increase interoperability with other collections. The Dublin Core extended plugin for Omeka was applied at first, but this plugin did not add substantial value to our particular collection, and it caused subsequent upload problems in Omeka.
After uploading our CSV files, we were able to make edits to individual item fields. We used this ability to add in hyperlinks to Jon Stich's website, and link famous artists to their type of synesthesia.
The SDL features several plugins: Exhibits, Simple Pages, Simple Vocab, Social Media, CSV import, Item Order, Library of Congress Subject Headings, and OAI-PMH Repository.
The Exhibits plugin helps users navigate through the three collections without having to browse items individually and helps prevent users from getting lost among all of our items. We very much liked that the exhibit option allows users to see navigation tabs for each of our famous synesthetes and synesthesia types for easy back-and-forth browsing. The Simple Pages plugin was used for the About page, as well as the landing page and this Creation page. The Simple Vocab plugin was used to create a controlled vocabulary in the Subject field. The Social Media plugin helps make our collection visible across a variety of social media platforms, and the CSV plugin allowed for large batch uploads of content. The Item Order plugin helped us to alphabetize our resources, famous synesthetes, and types of synesthesia, regardless of the order in which we uploaded them. Before discovering the Item Order plugin, we ordered our csv file "backwards" so that our famous synesthetes would be added in alphabetical order. With the Item Order plugin, this level of detail in our csv file was not necessary. The Library of Congress Subject Headings plugin allowed us to make sure subjects and tags conform to LC Subject Headings, and the OAI-PMH Repository plugin will enable our collection to be harvestable.
Although the SDL ascribes LC Subject Headings to the famous synesthetes and synesthesia types, the LC Subject Heading 'Synesthesia' is a bottom-level term. It was decided a finer level of granularity was needed, so the SDL employs a custom controlled vocabulary to define the types of synesthesia. All of this information is duplicated in appropriate item-level tags to facilitate searching and site navigation.
At the start of the project, we noticed that the default, "Thanks, Roy" theme would not allow our images to be posted at the top of the metadata. Also, the default theme caused some upload problems with a test CSV file: the HTML codes didn't always work and sometimes html-modified text appeared in a noticeably smaller font. Switching to the "Berlin" theme fixed both of these issues.
Before uploading a CSV file, we recommend choosing which plugins will be used. The CSV files we uploaded with Omeka’s default features worked fine. After we installed the Dublin Core Extended plugin, however, CSV uploads became very problematic; some fields did not register at all, while others were missing data or were formatted incorrectly. Subsequent attempts at tweaking the process yielded different, but not better, results.
It was also discovered that the image constraint settings must be applied prior to uploading image files. Because Omeka creates three derivative files of any image uploaded -- small square thumbnail, item thumbnail, and full-size image -- changing the image constraints after uploading an image has no effect. For this reason, using the largest possible image is recommended to ensure the best resolution.
We found that the Simple Pages have "breadcrumbs" in the top left corner of the pages. Grant was able to go into the CSS file and make the text white. This is something future users might want to look into removing entirely, but requires tinkering with the PHP code.
We experienced a couple of instances where changes were made in Omeka, and appeared to save properly, but then did not. Saving work in Notepad or TextEdit and CSV files in Excel or as a Google Spreadsheet prior to uploading content into Omeka can somewhat alleviate the frustration this causes!
Ultimately, it would have been helpful if the DU Omeka installation had allowed more than one user to hold administrative privileges for the digital library account. This would allow for a more evenly distributed workload in the group and might encourage more experimentation.
Ways to Improve the Collection
To improve the SDL, we would like to add more synesthetes and types of synesthesia, as well as additional resources. If copyright permissions could be obtained, it would be nice to include some of the musical and artistic works of our Famous Synesthetes. If copyright permissions could not be obtained, links could be provided instead. Because these improvements will depend upon our access to the DU server, Grant plans on migrating the site to the server that hosts his website to ensure that it remains accessible.
Other future projects include connecting our site with Google analytics to learn more about our users and how they are navigating our collection. Finding ways to make our site searchable via Google--perhaps through SEO--would also greatly increase the collection's visibility.
A very long term goal for the SDL would be to pair it with an interactive model for users, where those users who have synesthesia could create profiles of themselves and post creative work to the library. Using Omeka's Contribution plugin might be one way to accomplish this. It would align well with Web 2.0 features and might draw more awareness to the prevalence of synesthesia among a wider population of users.
To facilitate maximum exposure of the SDL, we plan to register our collections with an OAI-PMH repository.
We believe that building the SDL in Omeka was a very positive learning experience. Experimenting with plugins and file transfers to the DU server, thinking critically about the structure and navigation of our collection, and considering the desires and needs of our potential users helped put the course readings and discussions into perspective. We look forward to sustaining the SDL, migrating to new media as necessary, and continuing to learn what it takes to maintain a digital collection.