data sharing

[Una Smith]

Hugh Wilson wrote:

>I don't "reject the utility" of 'audit trails' but only suggest that
>- to me - it seems strange to worry about this when, at this point
>in time, there is little to audit.  What sort of networked info is
>being used - at this point in time - for biodiversity policy
>decisions?  ITIS?  TNC? maybe CNN?

As I understand the term, "audit trail" has nothing to do with policy
decisions, and there is a VAST amount of data to audit.  Another term
for "audit trail" is "record history";  an audit trail allows future
curators to painlessly, effortlessly reconstruct any and all changes
made to a record, if necessary, and recover any prior instance of the
record.

I'm sure we've all seen specimen labels (or catalog entries, whatever)
where person #1 creates the record, person #2 corrects an error in the
record, and person #3 crosses out person #2's correction.  An audit
trail allows everyone to see precisely what elements of a record were
added, when, by whom, and may explain why.  Actually, the only people
who see this audit trail are the curators, not all users, and certainly
not the general public.  Any "good" database management system will
have a built-in mechanism for maintaining a complete audit trail of all
data.  The trail usually does not contain information on usage of the
data, only changes made to it, a date and time stamp, a user ID, and
perhaps annotations about the reason for the changes.

        Una Smith               Department of Ecology and Evolutionary Biology
                                Yale University
        una.smith at yale.edu      New Haven, CT  06520-8106