Continuity Model Database

Current Progress

  • Working proof of concept (Windows only)
  • Data from continuity problem files (i.e. cont6 files) is broken up according to their different classes which correspond to forms
  • Data is sent to database
  • Opportunities to mark up data
  • Data can be retrieved from database

Meeting notes

  • Version vs variation
    • Versions
      • Keep track of which "version" is the most up to date for a model
      • New version number might indicate either
        • Fixed "bug" in model
        • Updated model to reflect update in Continuity
      • Metadata change always indicates a change in version, not variation
    • Variations
      • Multiple variations of the same model could all be "correct"
      • Example: Roy has 100 different versions of a model where only the material laws are varied.
    • 1.3 = Version 1, Variation 3
  • Metadata should be stored in cont6 file
    • Could create a concurrency issue that could be easily managed by the database indicating to the user that another model has been previously committed where your model belongs, and ask the user if they want to use a higher variation number.
  • We could use more fine grained diffing to determine if data in a form has actually changed
  • Changes to deposit form
    • "Don't Deposit" ► "Not included"
    • "Use previous version" ► "Don't update"
    • "Deposit new" ► "Deposit Updated version"
    • Distinguish what is being updated (version or variation)
  • Schema discussion
    • 6 levels discussed
      1. Properties
      2. Model (whole problem)
      3. Classes (mesh, biomecahanics, electrophysiology, etc.)
      4. Objects (nodes, elements, material properties)
      5. Sub-objects (node coordinates, node fibers)
    • Search dialog has the best current example of this (Library ► Search; click "All" and view hierarchy)
    • Also data origin should be included (automatically generated vs based on user input)
    • New equations editor form should be important for designing schema

Important tasks remaining

  • Get working on other platforms (Mac)
  • Develop complete schema for Continuity data
    • Mark up data with information using schema for design
      • Automatically infer from data
      • User annotation
  • Metadata (data about the data) didn’t work last time I tested it
  • Searching

Fred's current tasks

  • Write a few pages describing:
    • Version system (versions vs variation)
    • Current design / implementation
    • Underlying database
    • Searching mechanism

Ankur's current tasks

  • Learn about current database
  • Figure out what aspects aren't working
    • Fitting-> data cannot currently be desposited

    • Electrophysiology->epEquations cannot currently be desposited

Adrians current tasks

  • Learn about database schemas
  • Work out schema for Continuity models