At 08:00 PM 10/26/2004 -0600, you wrote:
>I know that file by file indexing is more efficient, because it uses less
>resources than document by document indexing in the traditional paper
>records environment. I was also told that retrieval is more efficient, for
>finding a paper document in a file, if file by file indexing is used, and
>that is my experience in 20 years of practice. However, my references
>don't have much detail about why this might be so. Does anyone out there
>have any research papers or references addressing the value of retrievals
>with file by file indexing? I have a hard "sell" ahead of me!


I think much of this is based on the assumption that you know the document
types that exist in the "files" already by having a robust file plan in
place.  For example, you know that an HR file consists of document types A,
B, C, D, E, F, G, H, etc... but that doesn't mean that EVERY file will have
ALL of those document types, just that IF those document types ARE in a
file, they would be in an HR file.

If the requirement is to index at the document level, you will obviously
have a much more detailed index, but it will also require a significant
additional amount of time to support and naturally, you will STILL require
a file entry in the index associated with each document entry to identify
which file the document is located in.

The fallback position you could take would be that on a "day forward
basis", as existing files are accessed (or new files are generated) that
they are indexed at the document level, and that you keep track of the
amount to time it requires to generate this data and the requirements in
storage space for the additional data, then offer to have this compared
against the benefits of having the indexing done at a document level.

Larry

List archives at http://lists.ufl.edu/archives/recmgmt-l.html
Contact [log in to unmask] for assistance