Will anyone remember us? Thoughts on information loss caused by progress
Peter David Townsend · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2010
speed, fibre optic communication or cost per CCD pixel often follow a smooth logarithmic improvement per year. This seems desirable, but progress is frequently only achievable by introduction of new software, different types of storage media or new operating conditions. Consequently technologies become outdated. For transient information this is unimportant, but for long term storage and archiving of information, images, photographs etc, there is an inevitable loss of earlier records. This is not a new phenomenon as even information on stone or clay tablets has decayed or been lost, either by physical decay of storage materials or loss of understanding because of changing language and cultural nuances. Examples emphasise how technological progress has speeded up information decay and loss. Since logarithmic "laws" have been proposed to describe the trends for electronic improvements, one may consider if equivalent trends apply to information loss. It appears that one may propose that the product of three factors is roughly constant. These are the time needed to write the new information; the quantity of information stored, and the average survival time of the information before the storage medium has decayed or is obsolete. The reality of such a "law" is that, whereas we may currently have records and photographs from many earlier generations, our rapidly stored electronic data may be lost within a few years, and certainly will have vanished in a readable form for the next generation.