MIT’s SIMILE Project: Innovations in Metadata Interaction and Analysis

^[1]Well-formed Data’s post on Exhibit ^[2] led me to explore what was available from MIT ^[3]‘s Semantic Interoperability of Metadata and Information in unLike Environments ^[1] (SIMILE) project. I took a little time to examine some of the SIMILE project tools with an eye to how they could impact interaction with archival records and metadata, as well as how they might support the work of archivists. All the tools appear to be available via an open source BSD license ^[4].

Babel

Babel ^[5] converts files from one format to another. I did a test to see if it would convert one of the Library of Congress EAD Finding Aids ^[6] from XML to some other format – but it gave me an error (‘unqualified attribute ‘repositoryencoding’ not allowed’). I love the idea that I could just point this at an EAD finding aid and get something useful out the other side – but apparently that is a bit on the wishful thinking side – at least for the moment.

Exhibit 2.0

Exhibit 2.0 ^[7] is described on the Exhibit homepage as follows:

Exhibit is a three-tier web application framework written in Javascript, which you can include like you would include Google Maps. If you just want to show a few hundred records of data on maps, timelines, scatter plots, interactive tables, etc., why bother learning SQL, ASP, PHP, CGI, or whatever when you can just use Exhibit? To use Exhibit, you write: a simple data file, and an HTML file in which you specify how the data should be shown. Data + Presentation. That’s all there is to publishing, as it should be.

Sounds fabulous, doesn’t it? I wish I had a week to play with this tool. They have a whole slew of examples ^[8], but I think the two I list below do a fine job of showing what you can create (not to mention being fairly thematic for those of you paying attention to the US Presidential Primaries news coverage):

• 2008 Presidential Election Candidates on the Issues ^[9]
• US Presidents ^[10]

Gadget

Gadget ^[11] is an XML inspector designed to create useful summaries of vast pools of XML data. I didn’t download and play with this one – but it sounds like something that might be very interesting to pump a big pile of EAD XML format finding aids into to see what could be discovered from an aggregate ^[12] point of view.

Longwell & RDFizers

Longwell ^[13] is a faceted browser ^[14] for RDF ^[15] formatted data, while RDFizers ^[16] is actually a directory of tools which convert other data formats into the RDF format. It doesn’t exist now, but if there was an RDFizer that went from EAD to RDF then Longwell would become more interesting to archivists.

That said, they already do have both a MARC/MODS RDFizer ^[17] and an OAI-PMH RDFizer ^[18]. I suspect that many archivists could put their hands on archival data in one of these two formats – which makes experimenting with Longwell more plausible in the near term.

Final Thoughts

There are lots other tools that are part of the SIMILE project (screen scrapers ^[19] and timeplotters ^[20] and more ^[21]), but the ones listed above most ignited my imagination. Surely there are geek archivists even now rolling up their sleeves to figuring out how to leverage free open source tools like these, both to improve access to records and increase understanding of what we have and how well it is (or isn’t) documented.

I hope to find time to play with each of these over the next few months – but I would love to know if anyone else out there has already tried any of these tools. Have suggestions for likely datasets? Have knowledge of existing archive related applications using these tools? Please post your comments below or drop me a line via my contact form ^[22]!

Image Credit: The Simile Project logo displayed above is from MIT’s Simile Project website ^[1].