We are proud to announce the release of Hume version 2.18. This release introduces new features for Hume Core, Annotate and Orchestra. Hume 2.18 builds on features such as Graph Editing and Hume Filters which were introduced in Hume version 2.17.
GraphAware Hume is an intelligence analysis solution. It represents your world as a network of interconnected entities, builds a single view of your siloed intelligence holdings, and brings powerful, user-friendly, machine-assisted link analysis capabilities to end users.
Hume 2.18 introduces:
- Fine Grained Access Controls for Hume Graph Editing.
- New Access Points for Advanced Expand.
- New Hume Filters Functionality including item deselection.
- Pausing Components in an Orchestra workflow.
- Manually Commit Consumed Kafka Messages in Orchestra.
Access Control for Graph Editing
In this release we continue to expand on the newly introduced Graph Editing feature with access control functionality. Access control consists of two features which can be used together or separately to give you more control over graph editing capabilities. In 2.18 we introduce:
- Restrict Editing capabilities which allow you to limit Graph Editing only to data which was created using the Graph Editing feature. Restrict Editing allows you to disable Graph Editing for data from a source system which you may not want to be editable.
- Fine Grained Access Control allows you to define the user roles that can view and edit the data that you have created in Hume. Fine Grained Access Control lets you define the level of granularity that you want for your Graph Editing access and enables controlled collaboration for users that share common access roles.
New Access Points for Advanced Expand
Advanced Expand is a powerful feature that allows users to write queries in an easy, schema-driven, visual way, without the need to learn the query language.
Advanced Expand is now easier to access from multiple points in your visualisation and no longer requires a starting node in order to use. With 2.18, Advanced Expand can be accessed from:
- the blank canvas without a selected node or node class.
- any selected node or relationship in the canvas
- the search bar which can open classes and relationship types, or the selected search result as the starting query point.
- the schema overview by selecting a node class, relationship, or the blank canvas.
Advanced expand is now available when you need it, to help you query the knowledge graph in an easy-to-use visual manner.
In addition, we have made performance improvements and Advanced Expand is now faster when working with large databases.
New Hume Filters Mass Operations
Hume 2.18 introduces new functionalities that make Hume Filters easier to use. The mass operations introduced are:
- Item Deselection: You can now deselect previously selected filter categories in the Hume Filters panel.
- Clear Filters: Reset your current filters selection.
- Remove Filtered Items: Items that have been hidden can be removed from the canvas.
Other Hume Core & Annotate Improvements
- Configure Metadata Input Types: Now you can configure metadata attribute input types in your Graph Editing forms allowing Hume to automatically set the following metadata attributes: created by/at, updated by/at, published by/at.
- Review Changes Before Publishing: Graph Editing now shows a detailed list of pending changes before publishing. The Review Changes overview window alerts you to any conflicts and allows you to edit or discard changes directly from the window.
Pause Components in an Orchestra Workflow
Components in an Orchestra workflow can now be paused and resumed within the workflow if predefined conditions are met. Pause and resume is useful for several use cases, including:
- Pausing (and waiting) to set a configuration in the context before consuming Kafka messages.
- Reading files containing a configuration and setting it in the context before reading files to be ingested in a second branch.
- Pausing the consumption of Kafka messages as soon as a failure is detected.
Manually Commit Consumed Kafka Messages
Consumed Kafka messages can now be manually committed at a desired point in the workflow. This new feature is relevant in scenarios where a message consumed from Kafka needs to be transactionally processed by another system (e.g. a write transaction to Neo4j). Being able to ensure that the transaction was successful before marking the message as committed ensures data integrity in your workflows.
Other Hume Orchestra Improvements
- Workflow History Search: History versions can now be searched for so that they are easier to find.
- New Webhook Producer Component: The new Webhook Producer component allows you to configure communication between workflows via Webhook.