JSTOR and the Harvard University Library announce the availability of the beta release of JHOVE, an open source, extensible Java-based framework for the format-specific identification, validation, and characterization of digital objects <http://hul.harvard.edu/jhove/>.
Identification answers the question, "I have a digital object, what format is it?"; validation, "I have an object purportedly of format F, is it?"; and characterization, "I have an object of format F, what are its salient properties?"  Object characteristics (technical metadata) can be displayed in a simple plain text format or as XML, using public schemas where appropriate, i.e., the MIX schema for raster still image metadata.
JHOVE uses an extensible plug-in architecture, with modules available for the following formats and profiles:
GIF     - 87a, 89a
JPEG    - ISO 10918, JFIF, Exif, SPIFF, JTIP, JPEG-LS
PDF     - 1.0-1.4, Tagged, Linearized, PDF/X, PDF/A
TIFF    - 4.0-6.0, TIFF/IT, TIFF/EP, Exif, GeoTIFF, TIFF-FX, Class B, F, G,
P, R, Y
UTF-8
XML
JHOVE is distributed as a set of jar files, for embedding JHOVE functionality into existing infrastructure, with additional driver classes for pre-built stand-alone applications, with command line and Swing-based
GUI interfaces.  The JHOVE distribution is 100% Java and is self-contained, requiring only a J2SE 1.4 JRE.  It has been tested under Solaris and Windows 2000/XP operating systems.
JHOVE is available under the terms of the GNU GPL license <http://hul.harvard.edu/jhove/distribution.html>.
The development of JHOVE was funded in part by the Andrew W. Mellon Foundation.
<http://hul.harvard.edu/jhove/>