The JupyterLab Metadata Explorer is a step forward in providing rich context to data stewards, analysts, and scientists into the world of meaningful, semantically enriched linked scientific documents and data.
Jupyter Notebooks and JupyterLab have succeeded as general-purpose tools to wrangle, manipulate, and analyze documents and data with programming languages.
The JupyterLab audiences combine various computational documents (eg. narrative, code, applications, notebooks) and media types (e.g. figures, data) to contextualize their expertise in different scientific method.
Rich context metadata connects vocabularies and schema that enrich the context and meaning of information in computational documents.
The JupyterLab Metadata Explorer, a feature of the rich context ecosystem, is a purpose-driven extension that exposes supplementary meaning related to data. It is designed for individuals and organizations capturing knowledge in JupyterLab and Jupyter notebooks.
There is valuable contextual information–metadata–surrounding all of JupyterLab entities (notebooks, datasets, file, etc) which we call “rich context”. This rich context, when visible, enables the collaborative authoring of an emergent narrative around your work within JupyterLab. It empowers you to collaborate with your peers, discover new information, techniques, and results to help you make informed decisions. It gets to the heart of the underlying value of a dataset for stakeholder agencies and organizations and enables researchers to work with the data in a more effective manner. With the introduction of the JupyterLab Metadata Explorer, we enter a new phase of tooling which will surround the practitioner with rich context.
Classical sciences, the information poor predecessor to modern science, can manage their metadata knowledge without information systems.
The modern information rich landscape of data science is accelerated by global asynchronous collaborations.
The velocity of modern science is too fast for individuals to manage the complex meaning in data and computational documents; the rich context ecosystem is needed for managing meaning.
Specifically, we are defining metadata as multi-vocabulary knowledge graphs encoded as RDF, through JSON-LD, served by HTTP. Organizations - with their developers and scientists - collaboratively contribute to collections of interlinked descriptions of entities into knowledge graphs. These entities range from things, publications, datasets, events and people, along with other objects used in coding languages and scientific frameworks. The JupyterLab Metadata Explorer consumes an organization’s metadata catalog as knowledge graphs, to enhance interactive computing practices with supplemental meaning about the fundamental entities of data and their relationships. Ultimately, it connects knowledge generated by multi-faceted organizations of data consumers and creators.
Taken together, the result is a flexible system that serves an organization’s scientists, analysts, and data stewards. Organizations with existing metadata catalogs can serve their catalogs to their JupyterLab users by linking knowledge though the metadata provider; data stewards can also supplement their organization’s knowledge through links to various external metadata providers. Projects in JupyterLab benefit doubly from the auto-generated context supplied by the previous notebooks while maintaining the agency to curate personalized domain knowledge graphs. Rich Context helps organizations, data stewards, and scientists discover information, techniques, and previous analysis extracted from datasets, notebooks, publications etc., to access the underlying value of data in a more effective manner. The end result of linking a Metadata Provider for your organization is less duplicated work and faster innovation.
Metadata Catalogs already exist in various formats, are hosted in various ways by various organizations, and have many uses outside of JupyterLab. The Metadata Explorer connects the Metadata Service and Metadata Providers. It relies on well-tested and documented web standards to expose catalogs through the JupyterLab interface to end-users. The result is that the Metadata Explorer can merge many catalogs together into a unified view! Insert gif of Metadata Explorer here The Metadata Explorer, and the general rich context suite, aids you as your JupyterLab assistant. When the Metadata Explorer is open in JupyterLab it automatically accesses the knowledge graph, to surface links between what an end-user is working on and any associated knowledge within the graph. Keep your hands free as you browse the data that the software has linked for you. Boom, you’re a scientist collaborating with other Jupyter communities.
Go download and install the JupyterLab Metadata Explorer, and the other tools in the Jupyterlab rich context ecosystem. Your metadata awaits!