Infrastructure Implementation Timeline

A detailed workplan has been developed comprising subtasks and milestones, estimated time for each subtask, and resources required. The time frame of the project is short, but the work plan is designed to avoid dependencies so work can proceed in parallel. LINCS prioritizes conversion to harvest low-hanging fruit first, leaving more challenging datasets for later. Within year one, we will set up the Storage with an initial set of at least 40 million triples accessible through initial components of the consumption interface; from there we will roll out more tools and interfaces while converting the data at the same time.

Image illustrating the division of labour over a three-year timeline for the development of LINCS' infrastructure.

Year 1

  • Develop a policy/legal framework and consortium agreements
  • Devise a technical framework
  • Set up initial LOD (SPARQL) endpoint in first quarter
  • Begin text migration, entity and relationship mining
  • Establish ontology and entity store
  • Begin extending modular tools and documenting
  • Preliminary triple-store, API and interfaces
  • User experience design

Year 2

  • Triples generation and inference based on NER
  • High-throughput NER on initial text collections using CC processors
  • Conversion of other data
  • Incorporation of IIF
  • Prototypes for user interface piloted
  • Production-ready SPARQL endpoint/API complete
  • Firm up sustainability plans
  • Begin FRDR integration

Year 3

  • Conversion continues
  • Interface development and testing
  • Dockerize builds
  • Intensive program of training trainers in infrastructure use
  • Full documentation
  • Public launch of LINCS platform