Project goals: Binder’s long-term vision¶
Artefactual believes that Binder can help cultural institutions meet the challenges of long-term preservation by providing access to metadata (technical, administrative, descriptive, and preservation) that is critical in informing digital preservation policy and practice.
We hope to see the Binder project grow, and are interested in partnering with other cultural heritage institutions looking to solve similar challenges. Below are some of the priority goals we see for the project as it develops.
Please review the Current project status first - the below goals are proposals for Binder’s long-term development. Our immediate goals are described on the previous page.
Generalize to support a broad set of use cases¶
Binder was originally developed to help MoMA meet its goals implementing standards-informed, best-practice digital preservation policies and practices. As such, in its present form the application is not ready for broad use in other institutional settings without further development.
Our first development goal is to generalize the application, so it can easily be used to support a broad set of use cases and institutional practices out of the box. In any future development project, we hope to build this kind of generalization into the work carried out.
Remove AtoM as a dependency for Binder¶
In the initial development of Binder, the app was built by leveraging existing AtoM functionality and adding new features via AngularJS (1.2) front-end pages, which Binder communicates with via an HTTP API. See the Technical overview for more information.
The use of AtoM as a backend dependency was a development strategy necessary to meet the desired goals with the resources available. As the project grows, we hope to see Binder gain its own stack (database, web server, search index, etc) tailored specifically to the goals of the project. This will not only improve the long-term maintainability of the project, it will also improve Binder’s scalability and, by implementing a common data model already in use by other open-source tools, help to support better interoperability and integration with existing resources used by the cultural heritage community.
Integrate with more existing open-source tools¶
Binder’s strength is in its ability to pull data from several disparate applications used in preservation and conservation, and bring that information together into a user-friendly interface. As the project develops, we would like to see this approach maintained and expanded, so that Binder will be able to communicate with a wide source of existing open source preservation tools, including Islandora, Hydra, BitCurator, and more. Integration with CollectionSpace and other open source alternatives to TMS are also high on our priority list. In general, Artefactual prefers to collaborate with and participate in many of the exciting community development efforts already underway, rather than compete by duplicating the functionality of existing projects in novel ways.
Implement standards-based preservation of Binder data¶
At present, many of the complex relationships that can be created and managed in Binder are only preserved in the MySQL database that Binder inherits from AtoM. We would like to see those relationships - between AIPs, artworks, components, and supporting technologies - preserved in a standards-based way. This could mean greater integration with Archivematica, to allow for AIP re-ingest or versioning, and/or the creation of Archival Information Collections (AICs) designed to preserve relationships between preserved objects and packages in an OAIS-compliant manner.
Further descriptive metatada import/export functionality should also be considered to support the management of descriptive and technical metadata, the documentation of process history, provenance, and more. Existing linked data ontologies should be explored for their utlity in both preservation and interoperability.
Artefactual has always aimed to support standards-based description and preservation via its AtoM and Archivematica projects - we look forward to continuing and broadening this work via the Binder project.
Develop new tools, reports, and visualizations¶
Binder’s strength is in the access to collections data and metadata that it provides, and as the project grows, we hope to enhance the utility of the application by expanding these capabilities to support more use cases and functions. Below are a few areas where we are excited to consider how best to enhance Binder.
Presently, Binder includes a number of reports that can be generated by repository managers, including reports on fixity, ingest, download, significant properties, and more. Reports can be viewed in HTML tables via the user interface, or downloaded in CSV format.
Adding further configuration options to existing reports, as well as expanding the number and kind of reports available, are both prime goals for Binder as the application grows.
Binder’s dashboard currently includes a number of widgets providing analytics on collection size, growth, character, and activity. Making these widgets configurable, adding further options, and allowing users to better interact with the data displayed (including the ability to download and reuse it) would vastly increase the utility of the Binder dashboard across a broad set of use cases.
One of Binder’s most striking features is the context browser, available on Artwork records. Built using the D3 library, it provides users with a graphical user interface to create and manage complex relationships between a work, its components, related AIPs, and the supporting technologies required for ongoing preservation and access. By providing this information visually, repository managers have a powerful and intuitive way to explore and maintain collections data.
With the context browser, we have only begun to scratch the surface of what is possible. We see great potential for optimizing and enhancing the existing functionality, to provide repository managers with better and more intuitive tools. Adding the ability to view the current tree in different ways could also open up support for other kinds of cultural heritage data, including trees with large hierarchical structures, such as those found in archival description.
Further views could also be added to the context browser - here are but a few possibilities:
- AIP view: shift the context browser’s focus from artworks to AIPs, and use the tree to display and manage contained directories, files, objects, as well as related supporting technology and component nodes. Submit changes back to Archivematica for AIP versioning and/or re-ingest.
- Relationships view: view and manage relationships between one work in the collection and others.
- Taxonomy manager: visualize and manage linked data ontologies and taxonomical hierarchies; manage relationships between a term and other node types in the repository.
The goals and ideas outlined above are not exhaustive - they represent some of the possibilities that have arisen during the development of Binder, and areas which we see as important to consider for the long-term utility of the project.
Have you got other ideas? Why not start a discussion in our User Forum? If you’re a developer testing out Binder locally and would like to contribute code, we love pull requests! Check out our Github repository here:
If you are interested in discussing Binder development with Artefactual, please feel free to contact us at firstname.lastname@example.org