LP#1729620: (follow-up) do not suggest timestamp manipulation
authorGalen Charlton <gmc@equinoxOLI.org>
Fri, 24 Sep 2021 20:14:44 +0000 (16:14 -0400)
committerJane Sandberg <sandbergja@gmail.com>
Mon, 28 Mar 2022 02:57:42 +0000 (19:57 -0700)
commitb6151ac81e9b8e56f70ea5ff449da6e402967131
tree2caf33ab220fd2776699fbc87ab7aa5bd9b9b000
parent1aa1e20099cd827672a9818ca2740e44554e87f5
LP#1729620: (follow-up) do not suggest timestamp manipulation

The suggestion in the upgrade script and the release notes
to consider adding triggers to update biblio.record_entry.timestamp
upon updates of linked call numbers or items is not a good one:

[1] implementing it means that any workflows that require knowing
    when a bib itself has changed become untrustworthy
[2] it would cause tremendous bloating of auditor.biblio_record_entry_history,
    as every single change to an item -- including ones caused by
    circulation activity -- would result in a new entry.
[3] item changes due to circulation activity aren't necessarily
    going to want to trigger record updates by the harvester

Wanting the update timestamp for OAI harvest to reflect changes
to items is a perfectly valid use case, but this is not the way.

Signed-off-by: Galen Charlton <gmc@equinoxOLI.org>
Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Jane Sandberg <sandbergja@gmail.com>
Open-ILS/src/sql/Pg/upgrade/XXXX.schema.oai_views.sql
docs/RELEASE_NOTES_NEXT/OAI2/new_oai_opensrf_service.adoc