Navigating Cyberspace: Exploring & Rendering Linked Data in the Terminal

Navigating Cyberspace

Exploring & Rendering Linked Data in the Terminal

About

Anatoly Scherbakov
Software developer (primarily Python)
Linked Data enthusiast
Invited expert @ W3C JSON for Linked Data Working Group
- working on YAML-LD
- helping with JSON-LD

RDF

Consists of quads

Cyberspace

© William Gibson
A set of all quads available
for an individual
Quad is:
- Subject
- Predicate
- Object
- IRI of the source where the above where obtained

pip install iolanta


        iolanta https://purl.org/np/RAYskLSM5x29icArnWvo9nVrIVEN2mfPoDq3TQSgm-9kk

Let us copy and paste this command into the terminal. This is a URL of a nanopublication. Unfortunately, not all parts of this nanopublication have been converted to a human readable form. We can see this is Tobias, and we will have to guess that he retracts another nanopublication. Let's click on it.

Now this is a bit more interesting. This nanopublication asserts a few triples about red things, which are a class, have a label and a definition. All written by Tobias again, and then retracted using the publication we have already seen.

Let's click on the label. From the header of the screen, one might guess that this is a regular label property from RDF Schema vocabulary, but this page is trying to make it more explicit. We can see that RDFS defines this term and we can jump to the whole vocabulary.

Which conveniently shows us its terms, grouped by rather arbitrary categories. This is only one of possible views for the RDFS ontology, it is designed to present it, well, as ontology - that is why Iolanta chooses it as a default. However, there are other views, this is the Properties view which describes the ontology itself.

Note that one of its properties is Term Group, and that's exactly what defines grouping of terms that we have been looking at. This term is defined by VANN ontology by Ian Davis back in 2010.

I would like to show one more example. Let's go back, here, and see what we know about Tobias. His information is being sourced from ORCID portal, and this page shows a lot of pieces that we fail to properly render, they are just blank node identifiers. Like this one, which could theoretically be nicely rendered as Scopus ID. This still needs a lot of work.

On this screen, I will return to the presentation; if you have questions or thoughts, we can always come back during the discussion.

Use Cases

Browse the Cyberspace
Generate tables, lists, diagrams, and more from Cyberspace
Inject Cyberspace-sourced information into human readable content
Ideas are welcome!

What is Iolanta good for? Why is it built? - In order to make machine readable data more readable for a human. Now opinions might differ as to how human friendly a terminal is, but as a programmer, I use terminal a lot, and this environment was a natural habitat for me to experiment within.

In addition, terminal gives you quite a lot of interoperability. At this point, Iolanta is drawing a text based user interface, but it will be even easier to generate JSON, YAML, HTML, TeX or any other textual format - based on Linked Data available in the Cyberspace.

Therefore, in addition to merely browsing the Cyberspace, retrieval of information from there for use in other programs, or as part of documents, is also an important use case.

I will be thrilled to hear your ideas of how it might be useful in your work.

Links

Site: iolanta.tech
GitHub: iolanta-tech/iolanta
Feedback, bug reports, issues, PRs are very much welcome!