Purcell and Elmslie, Architects

Firm active: 1907-1921

Minneapolis, Minnesota :: Chicago, Illinois
Philadelphia, Pennsylvania :: Portland, Oregon


Navigation :: Home
"William Gray Purcell Job Files" digital images database (launches new browser window)
IMAGES System, University of Minnesota Libraries

The IMAGES project at the University of Minnesota Libraries is subtitled "Metadata Aggregator for Campus Digital Collections."  Using an XML coding system, this project provides digital access to facsimile documents from many significant archival collections held at the University of Minnesota.  A portion of the Purcell Papers is mounted for Internet access, consisting of images found in the shelf boxes of the commission files in the Purcell and Elmslie Archives record group.  Drawings (small sketches or reproductions of larger ones stored in flat files), occasional reductions of blueprints or working drawings, and photographs can be searched geographically, by date, string.  Links to the IMAGES citation pages are provided on my web site to offer another and perhaps more convenient approach to this resource.  This creates a nice "virtual" exhibit with materials not served by IMAGES (e.g. text files) or which are not part of the Purcell Papers, such as the recent photographs of P&E buildings by Scot Zimmerman.  Also, I have added many images of unique Purcell and Elmslie documents, including drawings, correspondence, photographs and artifacts, found out in the field, other collections, or given to me over the years, that dovetail with the IMAGES facsimiles. 

Regrettably, an extraordinary amount of useful digital information contained in the commission files has been omitted, apparently because it is not primarily graphical in nature.  Ironically, the scanning of key correspondence, cost calculations, or significant notecards makes these documents into images which then convey information that is much more valuable than, say, some of the very bad photographs taken in the 1950s which appear in the current presentation.  Sometimes a thousand words can be worth more than a picture.  In addition, the vast majority of the Purcell and Elmslie drawings are stored in flat files whose contents are generally not included (to date) in the IMAGES database.  Hopefully, the digitizing project will eventually add all of these materials. 

There are also some document identification errors in the IMAGES database, which have been corrected here.  For example, a leaded glass panel for the John Leuthold [aka Ward Beebe] house in St. Paul returns under a "Edna S. Purcell residence" query; there are a number of other like mistakes.  The geographical sorts have problems, as a look at the listbox on the IMAGES page reveals (e.g. there is no "Rhinelander, IL" or "Rose Valley, MN" work, being Rhinelander, Wisconsin, and Rose Valley, Pennsylvania, respectively).  You should look at all permutations to get a complete return.  There is also a recursion fault in the code that omits the last image from sets of 15+1; the IMAGES pages are designed to show fifteen thumbnails at a time.  A sixteenth image may be indicated, but the web page does not allow for retrieval.  A workaround for this is to search on the name of the commission, or similar keyword, but that does not always help when the query return set is still 15+1.

In some cases, particularly with colored pencil sketches on tracing paper, the color of the IMAGES digitizing is greatly distorted.  Because I have photographed many of these drawings for research using a high-end digital Nikon (thank you, Michael McGee!), sometimes I can provide different views of the same document which have more realistic color.  An example of this can be found in the A.B.C. Dodd residence sketches. 

Finally, my suspicion is that the IMAGES database is harvested by an "assembly line" crew, without the presence of someone who is sensitive to the content and context of the materials.  Clearly, seasoning the raw data collection with some intellectual selectivity would improve the usefulness of the resource, not to mention a better interface; the limitations of XML are self-imposed.  However, I praise strongly the existing database for the merit of being present and accessible.  On the whole, the IMAGES database is an extremely useful resource that provides access to many documents not otherwise electronically available.  Salutations, IMAGES!