Skip to content

4.2.0

Compare
Choose a tag to compare
@netcanvas-bot netcanvas-bot released this 18 Oct 21:45
· 1571 commits to master since this release
01e5e78

Welcome to Network Canvas 4.2.0

This release includes the following fixes and improvements:

  • Important: Updates the app to support the features of protocol schema version 2. This change should not impact existing protocol files, which will continue to work normally. We encourage you to update your protocols (using Architect) wherever possible, and explore the new features supported by this version. New features include the user of ordinal/categorical variables in filter rules and skip logic, as well as the ability to filter the nodes that are shown on a given stage, which adds powerful possibilities for shaping the flow of your interview.
  • Fixed a bug in the roster interfaces that prevented nodes from showing on the sociogram.
  • Fixed a bug that prevented the start screen from rendering correctly on devices running iOS 10
  • Modified the behavior of node labelling in response to community feedback. Node labels are now drawn from either codebook variables or node attributes called "name" regardless of case.

Known Issues

  • Skip logic rules using ego do not work at present. We are looking into this issue for the next release.
  • Loading CSV formatted data performs poorly, particularly on mobile devices. This operation is CPU intensive, and causes devices to stall, sometimes for multiple seconds.
    • Strategies for mediation: Use alternative data formats, and/or minimise the size of external data files. Use more powerful devices. Use the Name Generator Roster (Large) interface, which performs slightly better.
    • Long term resolution: Parse external data when session starts (potentially to temporary storage), and stream it into memory on stage load.
  • The Name Generator Roster (Search) interface does not work well on devices with software keyboards.
    • Strategies for mediation: Use alternative platforms if this feature is important for your research.
    • Long term resolution: Adjust the styling to better work within the constraints of devices with limited vertical space.
  • Large protocols may cause the app to crash during import on iOS.
    • Strategies for mediation: Do not include large assets in protocols. Avoid using iOS if large assets are essential.
    • Long term resolution: Attempt to implement unzipping within a stream.

This app is part of a free, open-source suite of tools for the collection of social network data developed through the Complex Data Collective, a registered not-for-profit, and funded by the National Institutes of Health (R01 DA042711). Network Canvas is a collaboration between Northwestern University and the University of Oxford managed out of Northwestern’s Institute for Sexual and Gender Minority Health and Wellbeing.

For documentation, further information about the project, and download links for the other apps in the suite, visit https://networkcanvas.com.

Please also support this project by providing feedback! To contribute feedback, email our project team at [email protected].