Introduction

Tawny-OWL is a new environment for Ontology development
Provides a very different model from existing tools
First, the driving use case for this
Then, move onto details

Use Case (karyotypes)

Humans (normally) have 46 chromosomes
Lost chromosomes or bits of chromosome are bad

Use Case (chromosomes)

Each of the 23 pairs has bands (visible structures)

Partonomy

What happens if we describe all of this
It gets complicated and big very quickly

Protege

Could do this in protege
Technically, it scales well to an ontology of this size
But the user interface does not

Protege

Generating many similar classes is painful
Hard to know how an axiomatisation will perform at the start
Changing them afterwards even worse

Tawny-OWL

Interactive environment, create new ontologies, classes
Built on Clojure, the JVM, OWL API and some reasoners!
Provides an R like environment, in comparison to Protege’s Excel

Ontologies in an IDE

We can now create ontologies in sources files, with comments
Using any of a range of IDEs (Netbeans, Eclipse, Intellij)
Or Emacs and VIM and other power editors
Here using the pizza ontology (because it’s simpler!)

Polyglot

It supports multi-lingual labels with an easy syntax
Tawny-OWL can is easily extensible

Pizza Templates

Can also build ontology patterns
Expands to many classes
Abstracts over axioms, so open to change
This ability comes directly from the hosting language

Human Chromosomes

The karyotypes ontology is an extreme example
~1000 classes from one pattern

Unit Tests

The consequences of reasoning can be hard to judge
Can add "probe" classes to test for inconsistency
With Tawny-OWL, can use a full test harness, with reasoning

Continuous Integration

Once we have tests, we want continuous integration
Because Tawny-OWL is hosted, we can use native facilities
This is Travis-CI — OWL API and HermiT can be tested in the same framework

Driving Protege

Tawny-OWL is not a rich visualisation environment
Protege is. We can use them together

Literate Development

We often forget the reasons why we made decisions
Documentation and ontology are equal
Literate Ontologies enable this

Literate Development

Also, can support argumentation
In this case, the ontology formalizes what the document describes

Tool Integration

Ontologies can become part of our tool chain
This is an extreme example!
One source file both makes music
And describes it’s own metadata

Conclusions

Tawny recasts Ontology Development as a form of software development
We use many, many, very high quality software development tools for free
We can extend for special purposes
We can integrate with other environments

Acknowledgements

Karyotype Ontology, Jennifer Warrender, Newcastle University
Integrated Music, Michael Bell, Newcastle University
Ignazio Palmisiano, OWL API
Matthew Horridge, Timothy Redmond, Jennifer Vendetti, Protege
Birte Glimm, Boris Motik, Ian Horrocks, HermiT

Links