The long term preservation of media assets is a primary objective of most audio visual archives. When an archive has a collection on discrete media, in either a digital or analogue format, this preservation work can involve a continual process of migration of content from old media to new.
Media formats become obsolete. This obsolescence puts pressure upon the archive to migrate content, often long before the media itself has worn out, so that the assets remain accessible. The obsolescence of media also makes it increasingly difficult and expensive to support the machinery needed to access the content, and with the need to migrate, this becomes more crucial over time. The issue of obsolescence in both discrete and mass storage is addressed in the tutorial on Media obsolesence and migration.
Once media content is migrated of discrete individual specialised media storage formats and onto a mass storage platform this issue is mitigated to an enormous extent, due to three main factors:
In moving to mass storage you break the link between physical format and data format. This means that the process of transferring data between physical formats is far less complex, and no longer dependant upon technologies approaching obsolescence.
The available physical storage media are no longer limited only to specialised broadcast and professional audio visual formats. The archive can now utilise the technologies available in the larger and more competitive IT storage market, reducing costs and offering greater flexibility.
Digital transcription between successive generations of media can be, theoretically at least, loss less. Unlike traditional analogue preservation processes, or even digital processes that involve decoding, decompression, recompression and re-encoding, the transfer of digital data files between media need not involve any loss of information. Quality of the content can be maintained indefinitely.
In short, preservation in a digitised mass storage archive is simpler, cheaper and more reliable, or can be if done well.