Data retention

In order to prevent re-distribution of a "copy" of data which later changes, or becomes non-public due to processes such as expungement, the best practice is to never retain or re-use Data.  Retention is actually expressly prohibited for certain specified data elements.  Avoiding the use of "stale data" is a lesson learned from credit report problems caused when a financial institution makes an error in their records, and that error winds up copied into the files of multiple credit-reporting agencies.

Where Data is retained, the CaseMD5 attribute from the search results should be retained in the Compiled Index Record in order to make it easy to monitor the court source data for changes.  the CaseMD5 is effectively a "fingerprint" for a case at a given point in time.  Before re-using retained data, it is prudent to execute another search based on the retained CaseMD5 and compare it with the new CaseMD5.  if the value is different, the case has usually changed, and a GETCASE should be run. 

Unless otherwise noted in a Web Service Definition,  only a handful of key/ID attributes explicitly identified in the Case- search results version are eligible for retention in  the Compiled Index Records.  These are typically retained for use in comparing one set of search results with another.  explicitly indicates such elements.

Some Web Service Definitions may override these general retention prohibitions and allow retention of more Data.  Generally, this only applies to web services (such as the Collections) which, rather than providing a complete snapshot every time, are sophisticated enough to provide updates when a case has changed since a specified date.