We had previously relied on the act of publishing (e.g. Repo::publication()->publish()) to indicate that metadata (new or changed) is available for downstream service to consume and used this as a jumping off point. By making it possible to edit this metadata post publication, I propose that we should treat each save event of the metadata as a MetadataChanged event and make use of the MetadataChanged and BatchMetadataChanged events.

I would propose a naive approach that simply always dispatches the MetadataChanged event when the form is saved/updated post publication, and let each individual implementation be in charge of deciding what metadata is important to process.

This approach has the advantage of using already existing events and jobs and allows any updates to happen asynchronously to the web request. It is also extensible for any future services or third-party plugins that need to know when metadata has been updated.

Examples of downstream services that would make use of this (either currently implemented or future implementations):

• Mark DOIs for a submission as stale. The DOI logic already handles resubmitting stale DOIs. This should already implemented for major the registration agencies we support.
• Update ORCID records for authenticated users in their ORCID profile. I don't think this has been implemented yet, but it could be added independently of this work.

@asmecher Ready for review
OJS_main:
ojs: pkp/pkp-lib#10263 Relax editing metadata on published/posted materials #4410
pkp/pkp-lib: pkp/pkp-lib#10263 edit metadata post publication #10335
pkp/ui-library: pkp/pkp-lib#10263 Relax editing metadata on published/posted materials #398

Thanks, @Hafsa-Naeem! @Vitaliy-1, would you be able to review this one?

@asmecher @Hafsa-Naeem On ui-library side of things this make sense to apply to the new side modal workflow page. Trying to get that side modal workflow page as default on main branch ASAP.. Its probably 2-4 weeks away.

This is a good change, because journals get stuck and create new versions easily without any good reason (correcting a typo etc.)

Just asking that when we now edit the metadata, is this visible in the editorial history? ie. do we stamp who did what? I think this is important especially after something has been published.

There would be an event in the event log, but if the editor chooses not to create a new version, the change will not be tracked.

Note that we'll be extending versioning further with #4860 -- description there.