About the Project


The InsideWood Working Group (IWG) includes wood anatomists and NC State University Libraries professionals. The IWG has been fortunate to have had support and advice from technical specialists and numerous content contributors.

InsideWood Personnel:
Elisabeth Wheeler (Professor Emerita, Department of Forest Biomaterials, NC State University).

Richelle Bangi, David Barker, Chris Bazzle, Marcq Brooks, Alan Coulson, Sara Decherd, Amanda Garrett, Patrice Gladden, Lauren Grove, Jennifer Harper, Ryan Rodgers, and Sarah Schwarzer (students) scanned black and white negatives and 35 mm color transparencies, and recorded metadata.

Major Partners:
Pieter Baas
National Herbarium of the Netherlands
The Netherlands

Hans Beeckman
Royal Museum of Central Africa
Tervuren, Belgium

Peter Gasson
Jodrell Laboratory
Royal Botanic Gardens, Kew United Kingdom

Jugo Ilic
Wood & Fibre Quality

Shuichi Noshiro
Forestry & Forest Products Research Institute
Tsukuba, Japan

Other individuals and institutions contributing images and data to the InsideWood Database are listed in the Acknowledgments section.

Project Details

The original system consisting of ColdFusion, Oracle, PL/SQL, PERL, and XSLT needed to be enhanced for the new fossil wood data and the searching capabilities that were desired. It was decided the best approach would be to rearchitect the system using Java 1.5, Hibernate, Apache Wicket, Ajax, Apache Lucene, Apache Axis2, Apache Maven 2.0, Subversion, and GlassFish and to modify the Data Model.

The fossil wood data, which had been accumulated and maintained over several years via Excel Spread Sheets, was parsed using Apache POI and persisted to the new Oracle Database Model. The existing modern wood data was migrated from the old database schema to the new one using Oracle SQL scripts. Also, during this data migration, we took the opportunity to cleanmuch of the data and add new data constraints and indexes that had been previously overlooked.

The enhanced functionality includes: listing mismatches on feature searches along with the number of mismatches, results sorting capability (alphabetically or by the number of mismatches), improved user input validation, feature item description popups on both Fossil and Modern Wood Menu pages, a browseable taxonomy tree, displaying taxa with images only (no descriptions) as well as taxa with descriptions via keyword search or family/genus browsing, session based description selection and export. Many of these are functions that were requested by users of the 1.0 version of InsideWood. Additionally, there is a new data collection / editing interface with more functions, allowing editing of all parts of the database (e.g., creating synonyms as well as adding authorities, references) and including internal image validation and processing, and various new reports on database content. A new scalable CRUD application framework was designed and implemented for the data collection site.

Version 1.0 of InsideWood was developed using Oracle 9i in a UNIX-clustered environment. The complex data module was comprised of three major components: taxonomic information, image information, and descriptive information. Data were converted from text files and loaded into Oracle using PERL, XML, and ColdFusion. A controlled access interface allows editing of the content, including addition of new descriptions. Image data were loaded using ColdFusion, which relates the images to taxonomic and description records. The web-accessible interface to the InsideWood database has been developed using ColdFusion, Oracle indexes, and pre-formatted display data.

More details about the project development are shown in the documents below:

Workflow Diagram for building the database.

Progress report / overview, February 2005 [2 MB, PDF of slide show]

Afro-European IAWA Meeting, November 2004, Montpellier, France. Abstract. [2.8 MB PDF of poster]

Botanical Society of America, Botany 2004, August 2004, Snowbird, Utah. Poster Abstract.