摘要：One of the most important aspects of the long-term digital-image preservation strategy is maintaining data fixity, i.e., assuring the integrity and authenticity of original data. This article aims to highlight the limitations of the approaches used to maintain the fixity of digital images in the digital preservation process and to offer perceptual hashing as a way to alleviate some of the limitations of current approaches, as well as discuss some non-technical implications of the described problems. This paper is exploratory, and while it includes a simple experiment description, it only outlines the problem and testing environment for a possible solution that could be elaborated on in further research. The most commonly used fixity maintaining techniques are immutability of data and file checksums/cryptographic hashes. On the other hand, planning for long-term preservation necessitates the need to migrate data into new future formats to maintain availability and sustainability, and the concept of the file itself should not be assumed to remain forever, which calls for other tools to ascertain the fixity of digital images. The problem goes beyond one that is exclusively technical: bitstream content is not ready for human perception, and the digital preservation strategy should include all the necessary technical steps to assure the availability of stored images to human eyes. This shifts the perspective on what should be considered the digital image in digital preservation. It is not the file, but a perceptible object, or, more specifically—instructions to create one. Therefore, it calls for additional tools to maintain fixity, such as perceptual hashing, transformation logging, and others.
