Fourier Making Waves - Publish 1.2 with PhET-iO

[pixelzoom]

A PhET-iO version was not part of the plan for the Fourier 1.0 release. But the reality is that it's pretty much impossible (or very unwise) to do sim development without considering PhET-iO, and doing a fair amount of the instrumentation to test what you think you're doing. Consequently, I've been doing basic instrumentation of this sim as I've been implementing it. Because the sim has no dynamic elements that are instrumented, the instrumentation has been relatively straightforward, and I feel like the instrumentation is ~90% of the way there.

So... Is there any interest in including the PhET-iO version as part of the 1.0 release?

To help inform this decision, I would encourage @arouinfar and @kathy-phet to review the PhET-iO version (Studio, etc.) in master.

[kathy-phet]

If it's that far along, seems worthwhile to do the PhET-iO release. I'm wondering how the Game part is instrumented though? We will take a look in studio.

[pixelzoom]

Instrumenting the game was straightforward. Each level has 2 Fourier series, answerSeries for the answer to the challenge, guessSeries for the users's guess. In Studio:

If you expand answerSeries or guessSeries, amplitudesProperty shows you the 11 amplitudes for the 11 harmonics in the Fourier series. Or you can inspect the individual harmonics. For example, for answerSeries in Studio:

[arouinfar]

@pixelzoom I haven't done an in-depth review of the tree, but the sim is generally looking great in Studio. I would be in favor of including PhET-iO in the 1.0 release.

The basic functionality of the game feels pretty good to me. I could see a client wanting more control over the game -- customizing challenges, changing the threshold for the RewardNode/Dialog. However, I think we could punt on that until it becomes an actual client request.

[pixelzoom]

8/12/21 design meeting @arouinfar @kathy-phet @ariel-phet @pixelzoom

Let's try to publish the PhET-iO version.

Basic instrumentation is done. So this is ready for @arouinfar and @kathy-phet to continue the PhET-iO design process. I'll rename this issue to "PhET-iO Design" and assign to them.

[pixelzoom]

FYI, #4 (comment) identifies some potential requirements, where @arouinfar said:

As far as I know, there isn't any general PhET-iO infrastructure in vegas, as we've never instrumented a game (build-an-atom excludes the game). Thinking generally about any game, a client might want to:

• Hide specific levels
• Change the condition for triggering the RewardNode (e.g. lower the threshold to 5 stars)
• Specify their own challenge, or at least have some way of ensuring all students get the same sequence of challenges

[pixelzoom]

8/16/2021 @arouinfar @kathy-phet @pixelzoom had a design meeting on Zoom. We created GitHub issues for PhET-iO feature requests, and design work that needs to be done.

[pixelzoom]

8/1/2021 Slack discussion @arouinfar @kathy-phet @pixelzoom

We decided NOT to include PhET-iO in the 1.0 version. As we got into it, the amount of work grew quickly. It was pretty clear that we'd need to push out milestones, possibly into November for 1.0 release. @kathy-phet pointed out that while the milestone for Fourier are flexible, there are other higher priorities.

So I'll wrap up the PhET-iO issues that I'm working on, put good notes in the issues, and defer those issues.

@arouinfar will also pause PhET-iO design, and put her notes in a GitHub issue.

[arouinfar]

Design review notes can be found in the PhET-iO design doc and comments are summarized here:

• PhET-iO design of Keypad, see Instrument Keypad.js for PhET-iO scenery-phet#540
• These should all be read-only
  • dialog.visibleProperty (applies to all dialogs in the sim)
  • dialog.closeButton.enabledProperty (applies to all dialogs in the sim)
  • dialog.closeButton.visibleProperty (applies to all dialogs in the sim)
  • discreteScreen.model.oopsSawtoothWithCosinesEmitter
  • discreteScreen.view.charts.amplitudes.amplitudeKeypadDialog.visibleProperty
  • discreteScreen.view.charts.amplitudes.amplitudeKeypadDialog.closeButton.enabledProperty
  • discreteScreen.view.charts.amplitudes.amplitudeKeypadDialog.closeButton.visibleProperty
  • discreteScreen.view.charts.sum.sumChartNode.autoScaleCheckbox.autoScaleText.visibleProperty
  • discreteScreen.view.charts.sum.sumChartNode.infiniteHarmonicsCheckbox.infiniteHarmonicsText.visibleProperty
  • discreteScreen.view.controlPanel.infoDialog.closeButton.enabledProperty
  • discreteScreen.view.controlPanel.infoDialog.closeButton.visibleProperty
  • waveGameScreen.model.level*.correctEmitter
  • waveGameScreen.model.level*.incorrectEmitter
  • waveGameScreen.view.level*Node.charts.amplitudes.amplitudesChartNode.amplitude*Slider.visibleProperty
• Remove Studio Control
  • discreteScreen.model.fourierSeries.harmonics.harmonic*.amplitudeProperty
  • waveGameScreen.model.level*.answerSeries.harmonics.harmonic*.amplitudeProperty
  • waveGameScreen.model.level*.numberOfAmplitudeControlsProperty
  • wavePacketScreen.model.wavePacket.centerProperty
• When pressListener.enabledProperty or dragListener.enabledProperty is false, the cursor should not be a hand on hover. Affected elements:
  • amplitude*NumberDisplay.pressListener.enabledProperty
  • discreteScreen.view.measurementTools.periodCalipersNode.dragListener.enabledProperty
  • discreteScreen.view.measurementTools.periodClockNode.dragListener.enabledProperty
  • discreteScreen.view.measurementTools.wavelengthCalipersNode
  • wavePacketScreen.view.measurementTools.componentSpacingToolNode.dragListener.enabledProperty
  • wavePacketScreen.view.measurementTools.lengthToolNode.dragListener.enabledProperty
• Setting discreteScreen.view.controlPanel.fourierSeriesSubpanel.soundLayoutBox.outputLevelSlider.visibleProperty to false should also hide the min/max labels.
• Instrument the separators in the panels on the Discrete and Wave Packet screens. Refer to phScale.microScreen.view.beakerControlPanel.separator as an example.
• The individual radio buttons within seriesRadioButtonGroup should be instrumented.
• waveGameScreen.view.level*Node.feedback.smileyFaceNode should have a visibleProperty (frownyFaceNode has one).
• On the Wave Packet screen, the "Component Spacing" title can be occluded by the info button when the string becomes long, but this doesn't happen to the "Fourier Series" title on the Discrete Screen. See wavePacketScreen.view.controlPanel.componentSpacingSubpanel.componentSpacingText.textProperty
• These instances of titleNode don't appear to do anything, as the text is handled in a sibling. Let's uninstrument these:
  • wavePacketScreen.view.controlPanel.componentSpacingSubpanel.componentSpacingControl.titleNode
  • wavePacketScreen.view.controlPanel.wavePacketCenterSubpanel.centerControl.titleNode
  • wavePacketScreen.view.controlPanel.wavePacketWidthSubpanel.conjugateStandardDeviationControl.titleNode
  • wavePacketScreen.view.controlPanel.wavePacketWidthSubpanel.standardDeviationControl.titleNode

Additional Coments & Open Questions

• Several buttons do not have an enabledProperty, including the eraser buttons, Show Answer, Check Answer, and New Waveform. Was this an oversight, or is there something different about these buttons?
• There are lots of ranges/values in the sim that can get broken in PhET-iO, for example:
  • Set amplitude of hidden harmonic to a non-zero value. It will show up in the Harmonics chart and contribute to the Sum, but doesn’t appear in the Expanded Form.
  • Set equationFormProperty to something incompatible with domainProperty (function of frequency, but in space domain) and the equation reads “undefined”.
  • TimeControlNode is only enabled when domainProperty is space & time, but a client could enable it for a different domain.

[pixelzoom]

Status for 1/4/2022 quarterly-planning meeting:

Basic instrumentation has been done. It's unclear whether design is done, or whether there's implementation can resume. @arouinfar's notes are in #123 (comment).

[arouinfar]

@pixelzoom I would consider the design review to be complete. Change requests are summarized in #123 (comment) with game-specific requests in #144.