ReDBox 1.10 Release Planning

Introduction

The next release of ReDBox (v 1.10) is being planned in consultation with the ReDBox user community and will incorporate as many of the feature requests as possible. These will be prioritised by the responses from the upcoming 1.10 Feature Release survey. This page will provide information into each of the feature requests.


Requested Features

Create a plugin for ReDBox and Mint to publish metadata to Proai

Implement an optional plugin that can send metadata to a Proai instance so that it can provide the OAI-PMH interface to Research Data Australia and National Libraries Australia. This could be beneficial as Proai is likely to be more stable than the current in-house developed OAI-PMH interface and support the specification more completely. Institutions that are already hosting a Proai instance may also find having harvest sets consolidated in the same place.

Modify forms so that a record doesn't need to be created before any values are filled.

Currently in ReDBox, when the user requests to create a new record (e.g. a dataset, data management plan or self submission record), an empty record is created with the title \[Untitled\] before the user is redirected to the form to input values. This means that there are often several empty records in the system created by users that requested to create a new record in error. Modifying the forms so that records aren't created until some data is entered will remove this as an issue.

Upgrade to Solr 5 to improve search capabilities 

Apache Solr is used in ReDBox to provide search and discovery functionality, there have been a number of updates to Solr that improve search capabilities including spatial searching, search suggestions and spell checking. Newer versions of Solr run in standalone mode only so it will be an optional feature enabled via configuration.

Better support for research data storage

Support integrations with Object Storage or similar systems to store the research data. Currently the application can store the data on the file system but it isn't suitable for large datasets.

Bring Mint's functionality into ReDBox, removing the need for 2 separate web applications 

By bringing the Mint functionality into ReDBox, it will allow for simpler maintenance from both a technical and business perspective. Features that are available in ReDBox would also be optionally accessible to Mint records such as the ability to edit additional metadata via forms.

Create an administration screen to report on storage and check integrity 

Currently there isn't an easy method to view the status of the metadata store. This feature would create a administration page that will provide storage statistic such as:

  1. The number of records in the metadata store
  2. The number of records in the search index

And provide the ability to scan storage and report on the integrity of the records.

Allow an administrator user to delete all records that are returned in a View 

Currently an administrator can request to reharvest all the records visible in a view (e.g. all published datasets). A feature that would be useful particularly during testing is to be able to delete all the records in that view.

Allow Services to be created and edited via the User Interface 

Many institutions don't have a reliable source of truth to harvest service information into Mint and therefore it'd be useful to be able to create and manage them via the user interface.

Standardise the Harvesting APIs to allow easier ingestion of records  

In the 1.8 release, standardised REST APIs for ReDBox and Mint were added. These are suitable for integrating systems working with individual records in real time but is not so suitable for batch harvesting of records. By implementing standardised harvesting APIs that accept structured CSV and JSON will be easier to integrate in situations where batch processing is required.

Implement a new web portal

With implementation of the REST API in the 1.8 release, it is now possible to create a brand new web portal. This will provide the opportunity to develop a modern front end that is mobile friendly.  The technology stack of this portal will be a standard framework that will be easier for technical staff to customise.

Implement webhook feature to improve integration capabilities with systems

With implementation of the REST API in the 1.8 release, ReDBox has strong capabilities to integrate with upstream systems (i.e. a system that needs to push records into ReDBox). While there has always been the capability to integrate with downstream systems (such as VTLS Vital) there isn't a standardised method and generally required developing a custom plugin. By implementing a webhook plugin, it will be possible to have a standardised and configurable method to notify downstream systems of events in the ReDBox system and also simplifying the process.

Add support for international metadata standards such as DataCite and DDI

Currently ReDBox supports the RIF-CS, Dublin Core and MARC metadata standards for data collections. Adding support for DataCite and DDI schemas will add support for international repositories.

Update the Vital plugin to support instances running versions higher than Fedora 3.5

Fedora changed their APIs in version 3.6 rendering the current Vital plugin incompatible. An update to the libraries used as was done for the storage plugin