Challenges and Opportunities

motherlamentationInternet and Web Development

Dec 7, 2013 (4 years and 7 months ago)


Looking Ahead:

Challenges and Opportunities

in VIVO Development

Jim Blake, Brian Caruso, Jon Corson
Huda Khan, Brian Lowe, and Stella Mitchell

August 24, 2012


Introduction to the VIVO development

Dashboards and external content

Scenarios for interconnectivity & search

Reasoning and the application ontology

Semantic features

Modularity, extensibility and collaborative

Questions and discussion

The VIVO development community

Developers within VIVO project

UF Harvester, virtual installations, logging, DV

Indiana visualization team

Contributions above & beyond the project

All the 2011 mini
grant projects (Duke widgets, ORCID,
Google Refine, DV
Docs, HUBzero, UMLS vocabulary

Semantic web services

Drupal connectivity for display and reporting

OpenSocial integration

ANDS/VIVO effort and VIVO at Melbourne, Griffith, QUT

Sakai integration via Solr and caching explorations

Connecting to external data

Linking to vocabulary services as of 1.4

Generalizing this capability to work with any
vocabulary or thesaurus available as RDF

Support for navigating broader and narrower

Applying to other shared external resources




Geographic locations


New page types in VIVO 1.5

Static HTML pages

Pages incorporating SPARQL queries for
embedded summary reports

E.g., latest publications, agencies currently funding
research anywhere in an institution

A little more work delivers contextualized reports
within individual page templates

E.g., a consolidated list of the research areas of faculty
members represented in any department

Datastar example

Data interconnectivity in one institution

Feeding downstream websites and apps

Central source of faculty directory and profiles

Impact via maps of projects by legislative districts

Faculty expertise & library liaisons in library search

Augmenting one VIVO with another

Reuse of Weill Cornell VIVO profiles in CTSC

Colorado Lab for Astronomy and Space Physics
supplementing university VIVO data with discipline
specific project and research resource data

Connecting campuses

Weill Cornell and Cornell Ithaca campus

Now include stub entries with name & title that resolve
directly from Ithaca to Weill

Want to make bi
directional and more granular

Joint search index may simplify discovery, but direct
linkages should also be represented

Other use cases

Stony Brook and SUNY REACH (6 medical centers)

Multiple University of Colorado campuses plus federal

30+ information schools, led by N. Texas and Indiana

Sharing curricular resources, not just profiles prototype

Reusable technology demonstrated at 2011

Linked data index builder

Drupal module & interface

Can be extended to support lookups of people,
organizations or other data across institutions
represented in the index

Scaling up

CTSA PIs voted in October, 2011to recommend
using RDF and the VIVO ontology

Multiple platforms making or planning to make
VIVO RDF available

VIVO DuraSpace anticipates offering hosting
and expanded participation

Will require evaluation of linked data from each
new source for consistency & completeness

Likely to grow via disciplines, regions, or consortia


Several dimensions

content and UI

VIVO 1.5 has basic support for language tags
keyed off browser language preference

VIVO 1.6 anticipates a UI to manage multi

A framework for internationalization of the UI and
system messages is also needed

This is a prime area for participation in VIVO
development even without specialized skills


Triple store independence

For scalability

For other reasoning features

Enables closer integration with the Harvester and
other ingest tools

Will make it possible to submit edits/updates from
outside VIVO

E.g., two
way communication with Drupal

Named graphs and provenance

Keep data from each source separate

Store metadata about each graph

Give users the ability to ask

where did this come

Heavy use of named graphs will require triple
stores designed for that purpose

Notably not Jena SDB

Implications for reasoning

OWL sameAs and related features

VIVO 1.5 reasoner fills in triples using other URIs

This needs UI work for 1.6

The goal is to show only one individual to users

Which URI to choose?

Keep all the information in linked data

Removing the sameAs statement leaves
statements associated with original URI



s built
in reasoner

Enhancements with 1.5

sameAs, inverse property, remove inferences if asserted

What kinds of reasoning might we want to do?

Sub property inference

Full realization of types for all the axioms of OWL

How much should the reasoner within VIVO do?

Reasoning over distributed data

Application configuration ontology

Vitro is designed to store and display data for
any ontology

Goal to adapt the interface readily to accommodate
different data schemas

Mechanism that is flexible, extensible, and
permits fine
grained control, and allows
inheritance of settings

How far to go?

Specifying behavior for individuals

Application configuration ontology

Application configuration ontology

Use ontology approach to solve a software
engineering challenge

Represent the contexts in which given
configurations apply as OWL classes and axioms

Transformation of configuration instance data

Reasoner sorts them into a hierarchy

Ontology editor

One of Vitro main functions

create and edit the ontology before then adding data
and displaying/retrieving it

Least emphasis on the ontology editor in the past few

New class and property hierarchy displays in 1.5

Improved metadata and interaction

Expand all to see more context

Complex class expression editor

Complex class expression editor

Modularity and extension


Improves organization of the code

Makes it easier to modify the code

Allows alternate implementation of core modules


Built on modularity and well
defined interfaces

Allows easy addition of new functionality

OpenSocial for extension

Collaboration with Eric Meeks at UCSF and others

Add new features

With the power of JavaScript and access to VIVO RDF

Develop extensions for multiple platforms

VIVO, Profiles, perhaps others

defined API and enforced modularity

OSGi for modularity and extension

Encourage modular architecture

Develop explicit interfaces and dependencies

Provide a packaging mechanism for extensions
and alternate implementations

E.g., a different search functionality under
development at NYU

Developer community tools

Reassessing our tool set

SourceForge is discontinuing hosted apps such as

Moving some tools to DuraSpace

Generally moving with the open source
development community

E.g., considering migrating from Subversion to Git

What ancillary services belong with the code and what
are independent?

Advantages of linking code repository with issue tracking

But specialized issue trackers have agile features

Supporting all of these paths

VIVO communities





Gaining the experience of DuraSpace

Dspace and Fedora communities

Newer cloud services

Many ways to contribute

User scenarios and use cases

Requirements analysis and design

Voting on features (

Responding on the lists



Implementation support


Development from the simple to the complex

We are you

Help us understand your needs

Help us promote open data and open software

Definitive, shared URIs for people/orgs/journals/events

Let us know about new tools and data

Show us your work

is a way to find VIVO activity
around the world at any stage of implementation

Contribute to our web presence

Participate via social media

Questions and Discussion

Thank you