This blog post review is focused on the digital tool Exhibit, developed as a part of the SIMILE widget collection by MIT. For the purpose of this post, I am reviewing this software as a history student hoping to create, host and maintain an online academic site.
SIMILE is an acronym for the “Semantic Interoperability of Metadata and Information in unlike Environments,” the toolset itself is extremely similar to its progenitor, MIT’s Dspace, it being an open source and customizable academic software. Thus we have the SIMILE widgets, an open source variation of the greater SIMILE release presented for public use. Exhibit is but one of six tools available but one most suited (in this blogger’s opinion) to a digital historian. According to the Digital Research Tools wiki, Exhibit is an example of a “Data visualising, collection building and citation management tool.” (2012) Exhibit’s primary advantage for a historian is its multi-faceted structure – historians can visualise and present data sets in a non-linear manner enabling, as the name suggests, an exhibit of their choice.
Exhibit comes in two forms, Scripted mode and Staged mode, with the former running in browser and the later requiring the user to host the server themselves. With the onsite example of the American Presidents project Exhibit demonstrates how one central and overarching theme, in this case presidents of the United States, can be represented from a variety of perspectives; birth year, death place, political party, religion etc. The multifaceted approach is immensely beneficial to historical projects with huge data sets linked together by a common theme. Perhaps the Letters of 1916 project could benefit from such a tool to aid with presenting where individual letters were written, by whom and by theme. In this regard Exhibit seems similar to a set of HTML tags; that clicking one displays data that seems to fall within the purview of the aforementioned overarching tag.
With all that being said, how does Exhibit work? As an open source tool, Exhibit is remarkably user friendly; its free, it scales in association with the project (more on this later) and is available in a multitude of languages. Accompanying these advantages is an active community who are more than willing to help new members adapt and utilise the software to their own advantage via in code commenting (made with JSDocToolKit) and an active developer wiki. Having such helpful features readily available is a tremendous boon for the tool. By their very nature open source digital tools are often complex and rather difficult to come to grips with, possessing a helpful and supportive infrastructure akin to SIMILE makes the this learning curve less steep as a result. The tool itself requires little to no coding experience at the Scripted stage although basic HTML is recommended. The tool itself utilises basic HTML configurations which grants the user quite a flexible base to work from, as stated by the site Exhibit can “publish and visualize data collections ranging from small personal collections in Scripted mode, up to large data sets in the server – based Staged mode” (2012).
However, now we come to the tool’s rather noticeable disadvantage: to run Staged mode requires quite the amount of technical knowledge. The active community described above is necessary to the success of Exhibit as a tool due to its tendency to grow rapidly when running in Staged Mode. The amount of tertiary readings and further third party software required increases in scale with the data sets used by the user. Thus, in Staged mode Exhibit runs the risk of becoming increasingly bloated and flabby to use in step with the difficulty of a particular field of research. Understandably, one cannot expect to run hundreds of thousands of data sets in browser, but having the user required to download and install JVM, Maven and SIMILE butterfly increases the amount of time required to supplement one’s knowledge before they can even work with the tool. Another disadvantage lies with the query mentioned above; is Exhibit simply a flashy set of tags and categories? That comes down to the scale of the project and the vision that the historian has in store for it. Comparing the Presidents project with the Cereal character’s project reinforces this point rather well; the president’s project utilises a Google maps plug-in to visualise these tags whereas the cereals projects simply displays a jpg next to text on a rather stark white background. Both are running in browser though with the code accessible through hyperlink, thus both are examples of Scripted mode. While these examples highlight the tremendous potential offered by Scripted mode the lack of Staged mode examples leaves much to be desired.
According to Exhibit’s official blurb “The previous version of Exhibit, [is] currently used on thousands of websites around the web…” Yet the webpage does not possess any links out to any of these sites. As such, saturation of the tool is rather difficult to ascertain. A quick Google search reveals instances of other SIMILE tools such as Timeline being used, but none for Exhibit. Does this suggest that Exhibit is not as widely circulated as the other SIMILE tools? In actually there simply isn’t enough data present to accurately suggest one or another eventuality.
To conclude, would it be possible to recommend Exhibit to a historian? The answer is undoubtedly yes, with that being said the yes comes with an important “but” tied to it. A historian is recommended to use Exhibit for their project BUT must be aware that they will need to relinquish quite a sizable amount of time to realise the potential of the tool. For smaller data sets (easily over a thousand) the tool is quite useful at providing a multi-faceted hierarchy to the given subject. As stated by the Tool Wiki – Exhibit is a well-rounded tool capable of Data visualising, collection building and citation management tool. When using Exhibit the historian must have a clear topic or umbrella term in mind. As mentioned above, the tool is akin to a series of tags and without a central topic to tag out from, for without such a term the tool is virtually useless.
Digital Research Tools Wiki. Web (2012) Accessed on 29/11/14
SIMILE. “Semantic Interoperability of Metadata and Information in unLike Environments.” Web (2012) Accessed on 1/12/14
Simile Widgets. “Exhibit 3.0 documentation.” Web (2012) Accessed on 1/12/14