Implementing a risk assessment service

This approach to format risk assessment raises several questions.

How is the Risk Score Calculated and by Whom?

If using PRONOM (from v7) the risk score is calculated by The National Archives (UK). Other services may in future provide format risk scores. Clearly the scoring agency will be an important factor in choosing the service. Format assessments are very detailed, with dozens of risk factors, and the final score is calculated from a series of ratings that are split across several categories. For more information on the categories and scores see this Preserv wiki page.

Can I generate a score based on selected risk categories, or in a different way from the TNA?

The functionality available from PRONOM is such that the user or agent can obtain risk scores broken down not only for each category but also for the individual parts of each category. These scores might then be weighted differently for each category to generate a score that may be more applicable for the preservation needs, and resources, of your repository. It should be noted that no risk scores have been provided by any service yet, so there is no recorded experience of weighting scores.

How can information in the PRONOM registry be corrected or updated?

There are two possible ways. First, contact and inform TNA. This process could take some time. The second possibility is to deploy a PRONOM proxy that would act as your local registry, storing information you have submitted or changed while all other information would remain up to date as maintained by TNA.

At time of writing the software to deploy an updatable PRONOM proxy is not available. Preserv2 developed the PRONOM stub code (downloadable from the preserv svn), which allows development of code referencing features not yet publicly available in PRONOM. Preserv2 views the stub server as a starting point for a possible PRONOM proxy. There has been talk about making PRONOM available to others who want to construct their own registry.