Geoff Willard Blog #1: Arabic Music Exhibition at the Yale University Library

I happened upon this Arabic music exhibition by chance after searching for 'music' on the Smithsonian's Library and Archival Exhibitions on the Web. The lack of sound samples nearly caused me to click back and continue my search, until I saw the image on the right. Perhaps this type of "exhibition" would impress 10 years ago; calling this a digital exhibition at all is a stretch given its poor design, layout, and execution. Even as a report on an exhibition the presentation is mind-boggling. The off-center record label bothered me enough that I felt compelled to critique the rest of the collection.

Collection Principles

As a literal translation of a temporary analogue exhibition, I cannot find any justification for the materials chosen. Each digital object is titled, with the author and date noted if known, but notice that all of this information is contained in .jpgs. While the able bodied user may find this annoying - everywhere you click brings you to an enlarged version of the text or image, highlighting text is impossible, search does not return as many search terms as it should - someone who uses a screen reader would be out of luck along with keyboard/voice only users (impossible to tab through the page). So much for coding for usability. No mention is made of copyright, although the bulk of their material is likely public domain. The general lack of context is jarring, beginning with the single page on Arabic music that only goes up to the Middle Ages, and ending with the 20th century singer Umm Kulthum.

Object Characteristics

While the page is littered with .jpgs, I can see that the raw data files were .tiffs. Using the JPEG standard for the web is perfectly acceptable for bandwidth and accessibility reasons, although I've read that GIFs or PNGs are better for text - i.e. the bulk of the collection. Each image is either titled 'picture##.jpg' or 'explanation##.jpg', replacing the ## with sequential numbers. Some administrative metadata is present for the pictures (time stamp, size & color characteristics, software), although information about copyright or long-term preservation is missing. 

Metadata

What would be called descriptive metadata cannot be parsed because it exists as a separate .jpg from the item it's describing, which is a shame because the titles, editors/authors, and dates would be useful as search terms. Metadata on archivability, persistence, condition, and authenticity is non-existent, as is any type of controlled vocabulary. Unique identifiers appear to have been an afterthought. Beyond re-titling the .jpgs, no other data appears to have been added. 

Intended Audience

Survey of Digitization students looking for a bad example of a digital exhibition? Fans of Arabic music might be marginally interested, particularly in the few images of scores, but without actual sound I can't see anyone sticking around or coming back. Those who missed the run of the physical exhibition or saw it but missed some of the details could find something interesting here.