...
Content in LIBSAFE Go is organized hierarchically. The highest level of organization in LIBSAFE Go is a node. When you are added as a user, you will either be added to an existing node or have a node created for your library. Nodes can have subnodes, which correspond to collections, record series, or divisions within a library. Each library should determine how it makes sense to organize subnodes within their node.
...
Workflows
A “workflow” in LIBSAFE Go is a set of sequential steps used to track completion of tasks and allow users to view containers based on what needs to be done with them next. Workflow templates are applied at the container level, and the container can be manually advanced through the workflow steps by clicking the “Send to Next Step” button at the top of the container view. Steps that are not applicable to a given container can be skipped. At this time, there are no automated actions or triggers associated with workflows.
All containers generated through the submission areas or created in the ACC node should use the “Accessioning Workflow” template to ensure consistent handling. Permanent containers in collections nodes should use the “Processing Workflow” template.
Archival Accessioning Workflow
Any content uploaded the Accessioning Node either directly or via submission areas should go through this workflow. It includes the minimum required steps to gain basic control of the content. If staff upload content directly into a collection container (for example, after processing files received on physical media) they should ensure equivalent steps are taken prior to entering the processing workflow.
Rename container with accession number
a. Enter the accession number for the content in place of the automatically generated alphanumeric title.
Review submission
Check to make sure all expected content is present and matches any submitted metadata.
Weed files that should not be preserved.
Create md5 manifest
Click “Explore Content,” select all folders/files within the viewer, and click “create-md5-manifest” in the “Functions” menu on the right side.
Enter a file name for the manifest using the accession number or batch identifier (ex. “2020-08-08_manifest”).
After the process finishes running, refresh the container to make sure the manifest has appeared.
Save submission form and agreement
This step is for content uploaded by external contributors through a submission area.
At minimum, save screenshots or PDFs of the form fields, submission agreement, and container metadata within the container.
Follow local practice for saving copies in control files.
For Rose: Accessioner should also save a generic copy of the submission form and agreement, merged with a screenshot of the submission container metadata, in SharePoint in place of a transfer form/deed and link to this file from the accession record.
Create/update accession record
This step is performed in ArchivesSpace but is used in LSG to flag content that is not yet officially documented in an accession record.
Follow Emory and repository-specific guidelines for accession records.
For Rose: Before content can move on from this workflow stage, it should be recorded in the Acquisitions & Accessioning Airtable (by submitting an EUA transfer form or collections received form), with the location noted as the LSG Accessioning Node, so it can be assigned for accessioning and an accession record created.
Once the Accessioning Archivist creates the record in ArchivesSpace and assigns out the accession (typically to the University Archivist or Head of Digital Archives), the accessioner can update the record and complete the accessioning workflow in LSG.
Any important information received as part of the submission container metadata should be added to the ArchivesSpace accession record, as this information won’t transfer with the files to their permanent containers.
Include in the record a list of any original carrier/batch identifiers associated with the content.
Create permanent collection container (if applicable)
Check the appropriate node or subnode to see if a container for the collection already exists.
Create folder in collection container for each disk, accession, or other grouping you intend to save.
See Container Organization above for guidelines on grouping and naming.
Copy files to collection container
Select all content and use the Functions pane to copy it to the permanent collection container in the appropriate node.
Do not do anything else in either the source or destination container until the job is complete.
Delete submission/accessioning container
Do this only after the move is complete, you’ve verified the md5 manifest, and the content in the permanent container has gone through the processing workflow.
Archival Processing Workflow
Any time new content is added to a permanent collection container, the archivist should restart the Archival Processing Workflow and go through each applicable step.
Assign/update container metadata
Review content analysis
Dedupe/weed/arrange files (if applicable)
Note that for large or complex collections, it might be best to complete this step outside LSG on a processing workstation and ingest the content already arranged. This is because 1) the move and copy functions in LSG can be slow, 2) it is easy to cause discrepancies or lose information by changing container content while previous processes are running, 3) LSG lacks tools that might be useful in processing (e.g., for normalization or PII detection).
Update resource record
This step takes place in ArchivesSpace but must be completed as part of the process to ensure all content stored permanently in LSG is described in a resource record (aggregate representation is fine).
Follow ArchivesSpace field usage guidelines and local practice to decide which notes and other fields should be updated.
Create DOs
Follow digital content description guidelines for ArchivesSpace to create digital object records and record LSG links.
Your collection might have a single DO for all content in LSG or more granular DOs. See the Container Organization section for more guidance.
Any DO record, regardless of level of description, must correspond 1:1 with a folder (top level or subfolder) or item within the collection container.
Assign object metadata
See the Creating Object Metadata and Using Excel Metadata Import sections of this guide.
Tag restricted objects (if applicable)
Create md5 manifest
Content moved from the Accessioning Node will already have a manifest, but creating a second one is recommended to ensure it reflects any deletions or changes made during processing and contains all files, including those that might have already been in the collection container.
Submission-level manifests and all previous validation results should be moved to the PDI folder, but the latest manifest for the entire container should remain at the top level to enable periodic fixity checking.
Validate bags (if applicable)
If any content was bagged prior to submission/ingest, use the validation function to ensure it is complete and correct.
Upload documentation
Gather any manifests, logs, submission agreements, metadata spreadsheets, etc. and place it in the PDI folder.
Tips and Troubleshooting
In submission areas or when uploading directly to containers using the web interface, you can select multiple files when using “Add Files” but cannot select multiple folders when using “Add Folders.” If you want to add multiple folders at once, dragging and dropping is the best option.
There is no file size limit for browser-based uploads, but uploading either very large files or large numbers of small files can lead to errors and skipped files. If you have a large number of files to upload at once, it often works better to tar or zip the files first, then upload a single package.
For staff ingest of large files or large groups of material, use FileZilla to upload via FTP instead of uploading through a web browser.
Moving, copying, or altering files with the toolbar or the right-click menu options visible in the “Explore Content” window can cause problems. The system takes a long time to complete these changes in the background and does not block other interactions from happening at the same time, which can lead to discrepancies or incomplete tasks. It is a best practice to use the functions menu instead whenever possible: