search

bodleian / ora_data_model

Documentation and crosswalks relating to the ORA data model
1 stars 1 forks source link

SE repository status 'Private' (OAM 'Deposited (Not live)') #172

Open mrdsaunders opened 4 years ago

mrdsaunders commented 4 years ago

if there is a record_publication_date -> repository -status Public else if there is a record_review_status -> repository-status Private (In review) else -> repository-status Private

In OAM: repository -status Public -> Live repository-status Private (In review) -> In review repository-status Private -> Deposited (Not live)

It would seem more accurate to crosswalk all records without a record_publication_date as 'In review' since they have been deposited but are not live. Does anyone know what the rationale was for AB to write the crosswalk the way he did originally?

tomwrobel commented 4 years ago

@mrdsaunders I think the rationale was that Oxris might want to know (or care) about the distinction between 'deposited (but otherwise untouched)' and 'reviewed (and therefore under embargo if not live)'

We can possibly re-create this logic for objects that have a reviewer action associated with them somewhere in their histories but aren't in the access db

mrdsaunders commented 4 years ago

Thanks Tom that's helpful. We can discuss at ORA-SE.

mrdsaunders commented 4 years ago

Depending on outcome, solution could be ORA side, crosswalk or both.

eugeniobarrio commented 4 years ago

Are those repository-status Private -> Deposited (Not live) ever going to be reviewed?

tomwrobel commented 4 years ago

@eugeniobarrio they're probably already reviewed, just that no review status was recorded in ORA3 that corresponds to record_review_status in ORA4. We can probably imply that for those. Otherwise, yes.

mrdsaunders commented 4 years ago

We could just as easily change the crosswalk to map has ORA record but no status to In Review instead of Deposited (Not live). Are there any problems anyone can foresee with that?

We can also have a discussion about whether we actually want a status of 'Deposited (Not touched yet)' and how whether we could get S Ltd to map that to a status that doesn't trigger reminders to deposit.

mrdsaunders commented 4 years ago

In the short term, can we do as you suggest Tom and imply a status for those currently with no status (in order that they show as 'In review' in SE)?

In the longer term, we could use this status to distinguish between Deposited (review not started) and Deposited (in review). This could in turn potentially be used to distinguish between embargo end of 31/12/9999 where it is genuinely indefinite vs temporarily entered pending review.

What are the current values for record_review_status?

tomwrobel commented 4 years ago

The short term request is being tracked here: https://trello.com/c/uhkJz0bq

@mrdsaunders the allowed values for Review status are contained within:

https://gitlab.bodleian.ox.ac.uk/ORA4/reviewinterface/blob/master/config/authorities/review_status.yml

If one is not set, which value would you like?

@jjpartridge @tobypitts I assume the value "Incomplete/In progress" would indicate that an item was under review but that we had no other information?

@tobypitts can I have some sample object UUIDs for testing?

tobypitts commented 4 years ago

@tomwrobel Yes, we've generally been using the "Incomplete/In progress" value when we have begun a review but have not finished it or arrived at the point where we are able to assign another status

@tomwrobel I will gather some sample uuids shortly. Do you need a mix of Review status values?

mrdsaunders commented 4 years ago

Hi Tom. We don't mind what the value is as we don't interpret any specific values, just the existence of one. Use whatever value is appropriate to the the actual condition of the record.

tomwrobel commented 4 years ago

@jjpartridge are we happy with Incomplete/In progress?

jjpartridge commented 4 years ago

@tomwrobel @mrdsaunders can you confirm for me the scenario that we applying this value to?

has no record publication date has no review status recorded

Is that it?

Didn't we create "Deposited (not reviewed)" for this purpose

How many objects do we have that do not have a record publication date and/or a review status? Surely it is only the new deposits that have not yet been reviewed...

tomwrobel commented 4 years ago

I think the issue is that lots of records did not have their record_review_status set, but they have been reviewed. New deposits in ORA4 are not a problem, because the deposited (not reviewed) status should be set for these records

jjpartridge commented 4 years ago

Those records will have record publication dates though - correct? And therefore set as 'Public'

mrdsaunders commented 4 years ago

@jjpartridge Corect, it is applied to : has no record publication date has no review status recorded

mrdsaunders commented 4 years ago

@jjpartridge If you filter OAM by status 'Deposited (not live)' you will see objects that are not recent deposits but which have neither record publication date nor record_review_status.

jjpartridge commented 4 years ago

@mrdsaunders Are there many?

mrdsaunders commented 4 years ago

@jjpartridge there are 325 in the OAM

tomwrobel commented 4 years ago

325 starts to look like a migration edge case. Examples?

jjpartridge commented 4 years ago

Publications in OA policy.csv.xlsx

jjpartridge commented 4 years ago

Here is an example: https://ora.ox.ac.uk/objects/uuid:9cd3f03f-6dd5-41b4-9d27-e5c3d79e4956 - looks happy in ORA4-REVIEW and ORA4-SYMP

Perhaps failed on pairing or harvest?

tomwrobel commented 4 years ago

It's not a failure. We have a duplicate here:

https://oxris.ox.ac.uk/viewobject.html?cid=1&id=815179

has two ORA deposited objects:

One (the one you linked to) is public and live, the other is not.

Are the other items similarly affected?

mrdsaunders commented 4 years ago

@thomas-wrobel yes that is the common scenario. Here is one with four ORA records, one live the others not: https://oxris.ox.ac.uk/viewobject.html?cid=1&id=1077177