PREVIOUS   NEXT   CONTENTS   HOME 

10. The critical role of a digital archive

While an individual institutional archive, such as that provided by the University of Arkansas' Fedora solution, is an important element in the preservation and dissemination of digital heritage we argue that a formally recognised national (or international) digital repository is also of critical importance. It is a truism that the more copies of a digital object exist (or any form of an object for that matter) the greater the likelihood of its preservation. Additionally, such nationally recognised/supported repositories have a number of access advantages as well (McManamon et al. 2010). In the case of the Hampson and Amarna materials, agreements were initiated to place digital objects from each in such a repository. For the Hampson material the repository was tDAR, developed as part of the Digital Antiquity Initiative supported by the Andrew Mellon Scholarly Foundation, for Amarna the repository was the Archaeology Data Service (ADS) at the University of York. Both repositories have adopted the metadata standards and depository formats created by the recent Guides to Good Practice 2 project discussed earlier.

Though both repositories have similar file and metadata standards and have developed robust interoperability tools, deposit of digital objects in these two different repositories is quite different. For the ADS depository data-files are provided along with metadata in a standard form but ADS staff perform the actual entry of the data and metadata. The ADS staff review the various components and interact with the depositor to correct any identified inadequacies. In the case of tDAR metadata the depositor performs entry with various automated checks-and-balances to identify problems.

The point-cloud data files placed in the archives are in an ASCII format. This file format was chosen for archival purposes for a couple of different reasons. The ASCII format stores data in a raw text-based form that is easily viewed and exported from a number of different software packages. The American Society for Photogrammetry and Remote Sensing (ASPRS) Lidar Exchange Format (ASPRS 2005), which was originally established for airborne LIDAR, was also strongly considered for the archival format. LAS preserves important information from the original point cloud (such as point cloud organisation) where the ASCII format does not; however, it has not yet been adapted to support terrestrial datasets completely, and is not a common format used between scanning practitioners. It is also a binary format that can be more difficult to decipher without the appropriate software tools. While the ASCII data format tends to generate large file sizes, it is considered to be the easiest, simplest, and safest format to archive point-cloud datasets. In mid 2011 ASTM International (a standards setting organization) announced a data transfer specification for terrestrial laser scanning data, ASTM E2807-11. It is too early to be certain but it appears likely that this data format may be widely adopted. The specification can be found at http://www.astm.org/Standards/E2807.htm.

The comments facility has now been turned off.

Archived Comments

Selection of a digital archival format where a technology is evolving (as is the case for terrestrial and short range scanning systems) is a challenging task. The basic ASCII point cloud format has the clear advantage that it will be readable in almost all situations but it is large and the least "structured." Hopefully the field will develop either a de facto or de jure specification - perhaps ASTM E2807-11 will be it. But as of July 15, 2011 only one vendor announced support, Trimble Realworks. A search of the Innovemetric (Polyworks), Rapidform and GeoMagic web sites found no mention. So we'll see. Fred Limp

 PREVIOUS   NEXT   CONTENTS   HOME 

© Internet Archaeology/Author(s)
University of York legal statements | Terms and Conditions | File last updated: Tue Jun 28 2011