such a long term project, beginning as it is in the early development
of digital imaging protocol, there will undoubtedly be a rolling obsolescence
of the data formats and codecs. Maintaining a consistent appearance
throughout the movie may be desirable, although the changing feature
of the image quality due to upgraded systems may become visible as
an interesting corollary effect and theme.
the rate of development of digital imaging, capture chips should
be as large as possible, collecting as many pixels per image as
is currently possible. Even with resolutions on the order of 10
mpixels/image, the data rates and storage needs can be easily accomodated
with current networking and flash ram technology.
process of updating seems inevitable to maintain the readability
of the images from the rapidly evolving world of browsers and internet
protocols. This is a significant problem for many different sets
of digital data, and one that has as yet been poorly addressed.
One possible solution is the programmed upgrade of archives. Such
an automated conversion system may already be in existence, if not,
the analysis and construction of such a scheme may be useful to
many other managers of long term data sets.
more obvious solution would be to archive the images in a less volatile
format. Such an archive clearly adds a layer of expense and complexity,
but will assure long term accessibility of the images. One such
method of archiving would be to convert the digital data into a
color separated monochrome film positive print via an electron beam
recording technique. This technology is fully developed and being
used on a large scale to archive color film negatives and digital