Dear Natasha:

I think the list might get tired of our discussion as I do not see others jumping in, so I will respond briefly and then we can move to our personal emails. 

Let me premise that I do not believe that the decision of whether something is a record or not is an appraisal decision, so we agree on that point. 

With that said, if we refer to backups as records, as you believe we should, there are 2 scenarios that I see.

1. each complete backup as a whole is a record of the security and disaster recovery activity. As such, it is the by-product (as opposed to the product--which I believe it is) of such activity. This means that the back-up is not a copy but a record itself. Each backup results from a reproduction of the records system as it was at the time of its making, not any time later... Which means that it is not an exemplar of the records system other than for the brief time during which the system has had no change. In other words, a series of backups in succession is a series of records that requires a retention and disposition schedule. 

2. each complete backup contains documents that are copies of the records in the records system at one moment in time. We all know that two identical documents included in a different file, series, or aggregation are different records (a record is composed of a document and its relationships). The copies included in the backup are placed in a context that is different from the original context and which in a short time no longer reflects that of the original records in the records system. To keep one backup only as a container of copies of records (which you say are records themselves) does not make sense as it reflects the relationships of one moment in time, while keeping the sequence of backups for as long as the original exists at least would attest to the transformation of the context.

Neither of the two scenarios makes sense to me. I do not believe that backups are or contain records as their content is never used in the usual and ordinary course of business until such time when it is reintegrated in the system and becomes the records of the creator, if ever. I believe that producing backups is an IT function and that their management should be determined by IT people on the basis of the systems requirements, needs, emergencies, etc. Backup exists to recover records in the context of the system in which they existed rather than to use copies of records in the context of the backup. There is nothing academic about this, although, being sound practice, it is theoretically justifiable (practice informs theory and the theory so informed guides practice). 

The new Canadian Government Standards Board standard on electronic records as documentary evidence explicitly assigns backup responsibility to IT, and the standard has not been written by academics but by RMs, lawyers, IT specialists and archivists (30 professionals in all), followed by a public ballot. It is now being translated into French and it should be published by the end of the year or in January 2017. 

Luciana

Dr. Luciana Duranti 
Professor, Archival Studies
School of Library, Archival, and  Information Studies
The University of British Columbia | The Irving K. Barber Learning Centre
470-1961 East Mall, Vancouver, BC V6T 1Z1
Phone 604 822 2587 | Fax 604 822 6006
[log in to unmask] | www.slais.ubc.ca |
Director I Centre for the International Study of Contemporary Records and Archives
www.ciscra.org
This message is confidential to the parties I intend it to serve.

   

-----Original Message-----

Your implicit assumption that all exemplars of a record in a possession of an organization should be kept for the same time period is IMHO incorrect. 
I accept your point about back-ups as containers; I agree that themselves they are not records of line-of-business activity. However these containers do contain spare exemplars of various records (that’s one of the points of making them). Digitally signed records in back-ups are special because, by law, they are more or less self-sufficient: if the signature can be validated and limitations set in digital certificate are observed, the record will be effective for its purpose – so it won’t be just evidence. 

List archives at http://lists.ufl.edu/archives/recmgmt-l.html
Contact [log in to unmask] for assistance
To unsubscribe from this list, click the below link. If not already present, place UNSUBSCRIBE RECMGMT-L or UNSUB RECMGMT-L in the body of the message.
mailto:[log in to unmask]