Allow Multiple Data Sources in Custom Metadata Block

[Saixel]

Background

In the platform, each dataset can have multiple data sources contributing to its final content. Currently, the custom metadata block "Metadata about Data Sources" allows the entry of only a single primary data source. For example, Jen Stowell’s NDVI dataset has five different sources used to derive the final uploaded dataset, but only the primary data source (Landsat satellite) can be entered into the custom metadata block.

Here is the metadata block: https://github.com/IQSS/dataverse.harvard.edu/blob/master/metadatablocks/customCAFEDataSources.tsv

Here is a dataset that uses the block: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/ART54Z

Recent example of making fields multiple:

• Make Series multiple #9255 dataverse#9256
• Make productionPlace multiple, facetable, and enabled for Advanced Search #9253 dataverse#9254

Issue

The current limitation of the custom metadata block to accept only one primary data source means that it cannot capture all relevant information about the various data sources contributing to a dataset. This reduces the accuracy and completeness of the metadata, which can negatively impact the usage and understanding of the dataset by users.

Feature Request

The ability to add multiple data sources in the custom metadata block "Metadata about Data Sources," with clear distinctions between the different sources (e.g., separated with line breaks, etc.).

Justification

Having the ability to add multiple data sources will increase the accuracy and completeness of the metadata, allowing users to better understand the different contributions to the dataset. This is especially important for complex datasets that derive their data from multiple sources.

Implementation Considerations

• Evaluate how multiple data source entries can be captured and stored in the metadata block.
• Ensure that the user interface allows for easy and clear entry of multiple data sources.
• Implement backend changes to effectively store and retrieve these multiple entries.

Additional Context

Jen Stowell’s NDVI dataset is an example where this functionality is important. Currently, only the Landsat satellite can be entered as the primary source, while there are four other sources that should also be captured in the metadata.