Overview

Information is often only in regular active use for a comparatively short space of time. In due course its level of use will decline, perhaps because the information has been superseded or simply because the initiative to which it relates has itself ended. But this does not mean that its use, or indeed its usefulness have disappeared altogether. The information may still have residual value and be referred to on occasion for reference purposes, or may no longer have any informational value but still be required due to its possible evidential importance.

In many respects this represents the most problematic and potentially dangerous phase of all in the information lifecycle. This is because when information is in almost continual use this in itself often dictates its own management: we keep it close at hand and know instinctively where to find it, we know who we are working with and therefore who needs access to it and we know it has value because of its relevance to the task in hand. But when we have moved on to the next task and that peak of use begins to decline things become less clear-cut - a problem compounded by the fact that our interest and efforts have now been transferred to tackling the next task in front of us.

This is often the period during which people forget where the information came from, what its place in the process was or indeed why it was created at all. Such uncertainly often leads to the accidental breakdown of whatever carefully constructed management controls were put in place during the creation and active use phases. This can often lead to quite different and contradictory patterns of behaviour depending on the nature of the individual user. Where one member of the project team may come across project files from a long completed project and delete them as obsolete, five others may find the same files and each retain copies 'just in case'.

The following sections are designed to try to help you to steer through this grey period in the lifecycle and impose some structure on what can otherwise become the least well defined and managed part of the model: