Curatorial

Collections Registry Cataloging Guide

Introduction

Every MENA librarian is responsible for helping students and researchers find digital resources needed to study the MENA region. As part of this effort, they often create subject guides which are lists of collections related to the Middle East and North Africa. These guides are difficult to create and maintain and quickly fall out of date. A collections registry, collaboratively maintained by the Digital Library of the Middle East (DLME) and the Middle East Librarians Association (MELA) will lower the effort required to assist researchers in finding resources, lower the possibility of error by spreading the effort over a larger network of experts, and allow for better user interactions than a sorted list. It would serve as a starting point for research on the region and a first step in establishing a DLME data partnership. Ultimately, MENA librarians would direct their students and researchers to this collection registry instead of their own subject guide, increasing DLME users, and would add entries in this collection registry rather than building/extending their own subject guide.

Metadata Schema

The metadata standard used to describe collections for the DLME collections registry is the Dublin Core™ Collection Description Application Profile. Please consult the application profile for details. Collections can be entered in the DLME collections registry spreadsheet. We welcome contributions. If you are interested in contributing, please contact the DLME Data Manager (Jacob Hill: [email protected]) for request access to the collection registry spreadsheet.

Below are some field specific guidelines and tips that can be used in tandem with the above metadata application profile.

• Identifier: this field is the DLME collection id field agg_data_provider_collection_id. It can be found by selecting the Data provider facet and then the selecting the specific data provider of interest. Once selected, you can select the Collection ID facet and choose the collection id for the collection you are cataloguing.
• IsAccessedVia: This is the url to the collection in DLME (when available). To get the url, follow the same process used for identifier and select the collection id, then copy and paste the url.

• dataProviderUrl: the url to the collection on the data provider site.
• Subject: values separated by semi-colons.
• Size: the number of items in the collection.
• Language: human-readable values separated by semi-colons e.g. "Arabic; Persian".
• SpatialCoverage: values separated by semi-colons.
• TemporalCoverage: a date range expressed as e.g. 1800-1899.
• DateCollectionAccumulated: a date range expressed as e.g. 1800-1899.
• DateItemsCreated: a date range expressed as e.g. 1800-1899.
• ObjectType: use values from the DLME website Type facet.

Browse groups and categories

Collection review

DLME Environments

There are three DLME environments:

• production: Our production environment that users interact with.
• review: Used for developing new application features and iterating on data mapping.
• stage: This is not for staging features. This is a data staging environment used for staging data for partner review.

Syncing DLME environments

Once the data manager has worked through any mapping issues, data should be loaded in the staging environment for partner review. After incorporating partner feedback, or after a month has passed with no feedback from the data provider, the data should be loaded into the production environment.

Data in the review environment should be cleared regularly. The data in this environment should represent the latest state of the data coming out of the ETL pipeline. In most cases, it should not be more than three months old.

Curatorial settings should be applied in the development environment, tested, then applied in the development environment. Some curatorial work will require different settings in the development environment (e.g. enabling display on some facets that should not be displayed in production). For that reason, settings cannot be exported from development and imported to the production environment. Full curatorial documentation is available here.

Curatorial settings should be applied in the development environment, tested, then applied in the user acceptance testing environment. Once the user acceptance testing environment is configured correctly, the curator settings can be exported and imported into the production environment. Some curatorial work will require different settings in the development environment (e.g. enabling display on some facets that should not be displayed in production). For that reason, settings cannot be exported from development and imported to the user acceptance testing and production environments. The purpose of each DLME environment is documented here.

Much of the curatorial work will be done using bulk tags. For example, a curator might filter records using a combination of facets and keyword searches, then apply bulk tags to all of those records. This will create a stable set of records that can be used to build a browse category without concern that adding new records will inadvertently change the browse category. The work flow for adding bulk tags should be applied as follows:

1. [dev] Filter records as desired and apply bulk tags in the dev environment.
2. [dev] Manually inspect the records with the applied bulk tag and untag any records that should not be included (e.g. if the search results returned a non-relevant record.
3. [dev] Export all records from the dev environment by selecting 'Download csv'.
4. [prod] Delete all bulk tags in prod Caution, never delete bulk tags from dev! Make sure you are in prod. This step is important since importing tags from dev will not remove a tag from any resource in prod.
5. [prod] Import into prod the csv file you exported from dev in step 3.