Our panel members are Patrick Dubroy, Luke Church, Tijs Van Der Storm and Alessandro Warth.

This discussion will draw on both examples from the workshop, and prior work in the field. Topics:

Liveness provides continuous feedback on the relationship between program and output, with the potential to fundamentally change the way we interact with programs. What are some of the UI implications? Here are some questions we might like to think about:

• When does it makes sense to manipulate the output and have changes be reflected back to the program?

• How useful is liveness for discrete (non-continuous) domains? Are there any prospects for “datatype-generic” visualisation?

• Immediate feedback may be undesirable for performance reasons, ergonomic reasons, or both. Are there mechanisms (e.g. granularity control) that speak to this concern?

• Liveness makes the language implementation an inseparable part of the “experience”. Do we risk being overly dependent on a particular UI?

• Can liveness scale to large programs and datasets?

• Visualising intermediate values can be as important as visualising output. Such values require an execution context to understand. What can be/has been done here?

• Are there formal principles that might inform UI design, or is it all fundamentally subjective?