The MUA’s underwater archaeological bibliographic database project will benefit from groundwork completed over the past two years while I was employed at the US House of Representatives. During that time I devised and created a document metadata database quite similar to that which I am proposing now. Document information was stored in a mysql relational database and was accessed via a web page that used PHP (Personal Home Page Hypertext Processor) code to handle database queries and create pages to display their results. In addition the MUA has been in existence for approximately 2 years and in that time the quality of the html has improved though it still has far to go. I am attempting to make use of these two experiences in order to create the bibliographic tool.

The first major task will be to secure an initial data set and to test the waters to see if additional collections would be forthcoming. As described in the scope section the plan involved securing data from a few state underwater archaeologists to build a proof of concept. Once that was accomplished other states would be invited to participate. Hopefully they would be attracted by the functioning site. Maryland State Underwater Archaeologist Dr. Susan Langley was approached first due to a pre existing close working relationship and her noted practice of encouraging innovation. Dr. Langley agreed to work with the MUA to provide bibliographic information on state underwater reports and to consider including abstracts if financially feasible. Four other east coast states and one Pacific coast state were approached informally about participating in the project. Each agreed in principal to participate. Only Maryland data would be used during the initial stages of development due to the ease of acquiring the data which numbers well over one hundred records. Once testing shows stable results the other previously contacted states will again be approached in a more formal manner. After successful completion of further testing with the new records a plan will be created for acquiring additional data. This approach seeks to build a reputation based on a working tested site.

One of the greatest challenges facing this project is the planning of the database itself. While previous House experience demonstrates that a working set of tables could be created in mysql in a relatively short period of time this does not mean that that is the best approach. An interview with the Scott Gillespie, the Project Manager and Programmer for the Virginia Center of Digital History suggests far more thought must be given to the manner in which the tables and fields should be created. Gillespie worked on the Geography of Slavery website which utilizes XML rather than a relational database. While he agrees that the limited amount of text along with the type of metadata to be associated with the site report records makes a relational database a good choice there are additional considerations that must be reviewed prior to beginning work.

One of the most important trends in the web development is the move toward standards for the creation of metadata storage.7 Gillespie notes that his current project seeks to comply with the Dublin Core Metadata Initiative (DC) This is “an open forum engaged in the development of interoperable online metadata standards that support a broad range of purposes and business models.”8 By meeting this standard other organizations will be able to share and make use of the elements stored in the MUA database because both parties will be utilizing a commonly agreed upon set of elements, element refinements, and encoding schemes. In the common tongue this means an agreed upon way to describe field names and the types of values they will store. Meeting this standard is perhaps the truest expression of the MUA’s desire to live up to its mission to support the professional community. As this is a new standard not immediately known to me it will take additional time to learn how create tables following this protocol. The DC website host numerous documents that should aid in that effort. As this data set is not extensive the additional work should not be overwhelming and will certainly be worth the effort. An example of the potential benefits that derive from following this standard is the ability to use standard spatial codes that could be utilized at a later date to map the general high level locations of the various site thus providing a visual reference tool for analysis.

Having created the standardized database schema and tables PHP (Personal Home Page Hypertext Processor) code will be used to enable web based queries and to format the results. The web interface available to the public will only process select statements and not allow any record insertions, deletions, or updates. Utilities for those processes will be restricted to a small group of individuals possessing the proper authorization and skill sets. After creating the database and web interfaces a copy of the Maryland data to will be loaded for testing. The database will reside on a University of Rhode Island, Kingston (URI) server; the current home of the MUA website. Upon successful completion of those tests the tool will be incorporated into a test version of the main MUA website. Navigational links connecting the tool to the site will also be tested and appropriate color schemes and logos will be applied to maintain a consistent look and feel. Additional data sets from those states willing to participate would be added at this time followed by new tests until all of the original six states have been included and thoroughly tested. Finally announcements regarding the tool would be released to the general public and the underwater archaeological community.

Various tools and assistance will be required for this project including Dreamweaver and Photoshop. Software development web sites such as the PHP web site and mysql.com are also important sources of information. FTP clients and the server expertise of the URI information technology staff play a critical role in supporting this project and the MUA in general. This support is made possible by the URI History department and in particular Dr. Roderick Mather, Associate Professor of History and Director of the Archaeology and Anthropology option of their M.A. program. Taking all of this together this project could take upwards of one year to reach the stage where additional data sets could be added. This is a result of the numerous people that have to be contacted and upon whom we would be dependent for the data sets. From a financial perspective costs would be incurred with some travel for early data acquisition in Maryland along with phone calls to coordinate other issues as they arise. Start up cost for the site has already been covered by the MUA. This includes equipment costs such as the purchase of a desk top computer, a scanner, printer, and media storage devices. Additional specialty software for images and video was donated by several companies.9

_________________

[7] Many thanks go to Scott Gilespie for his willingness to discuss and share his views and work on building data .....structures that will comply with the DC standard.
[8] The Dublin Core Metadata Initiative (http://dublincore.org/)
[9] See the sponsors page on the MUA website for a complete listing (http://www.uri.edu/mua).