phetsims / john-travoltage

"John Travoltage" is an educational simulation in HTML5, by PhET Interactive Simulations.
http://phet.colorado.edu/en/simulation/john-travoltage
GNU General Public License v3.0
4 stars 8 forks source link

Alternative Input for Leg Swing and Arm Swing - double check #425

Closed terracoda closed 3 years ago

terracoda commented 3 years ago

Notes from Slack

Using my keyboard

Using swipes on my iphone 8 Note that swipes left and right move focus between items when VO is enabled.

I know there is something special about the direction of the Arm swing. I think we intentionally went with making the Right and Left Arrow keys move the hand in the most intuitive visual direction (right clockwise; left counter-clockwise). This means that Up Arrow moves clockwise and Down Arrow moves counterclockwise which is less intuitive.

On iOS with VO enabled, UP and Down Swipes move sliders. Right and Left swipes move focus.

So, on iOS the Up Swipe moves the hand down (clockwise) and the Down Swipe moves the hand up (counter-clockwise).

@emily-phet, and @BLFiedler are you both fine that Up Swipes moving the arm down and Down Swipes moving the arm up?

emily-phet commented 3 years ago

I think however we arrange this, there will always be at least one user scenario that has an unintuitive outcome (spatially). If we could flip the mappings for up arrow and down arrow, and flip the mappings for the up swipe and down swipe, would that be preferable across the board?

So the mappings would then be: Up Arrow - Counter Clockwise Up Swipe - Counter Clockwise L Arrow - Counter Clockwise

Down Arrow - Clockwise Down Swipe - Clockwise R Arrow - Clockwise

I believe at some point, long ago, the position values were read out, so the incrementing/decrementing position values was part of what was read out. I don't think that's the case anymore, so it's just the spatial mapping read out.

emily-phet commented 3 years ago

Also note - I'm sure if asked @BLFiedler might have an opinion, but I don't think he's ever taken on the role of designer for this sim. I think it's ultimately up to you @terracoda.

jessegreenberg commented 3 years ago

@terracoda and I met to discuss this issue because it could have an impact on #391. In the end we decided to do nothing for #391, but that I would take a pass at this change to see if it would be tricky. Turned out to be pretty simple and I did it in the above commit. Can be reverted, but I made the commit so we can play with the change.

Here is a version with the arm motion reversed: https://phet-dev.colorado.edu/html/john-travoltage/1.6.0-dev.43/phet/john-travoltage_en_phet.html

terracoda commented 3 years ago

@jessegreenberg and I discussed this issue briefly when we were looking at #391.

Recently, I realized that the issue with the arm and the leg could stem from the fact that the Arm Swing, from its starting position, seems like a vertical slider whereas the Leg Swing has a more natural fit as a horizontal slider.

Neither, however, are flat-line interactions, the Arm Swing follows a full circle, and the Leg Swing follows a half-circle.

At design meeting today, we can discuss the pros and cons of adding explicit orientation to either or both of these sliders, or adding some more explicit information in the keyboard help dialog to make it more clear what actions make the appendages move forward and backward or clockwise and counter-clockwise.

Currently, the Arm Swing has the opposite orientation when compared to the Leg Swing:

terracoda commented 3 years ago

Thanks @jessegreenberg for https://github.com/phetsims/john-travoltage/issues/425#issuecomment-879228551

I'll try it out.

terracoda commented 3 years ago

I personally like it. I don't hear any additional slider orientation language, so I am assuming none was added.

This is what I like:

terracoda commented 3 years ago

In Voicing design meeting, July 13, we decided to permanently change the direction of the Arm Swing slider so that it goes in the same direction as the Leg Swing slider. This change affects the sim, not a single feature (i.e., Voicing versus PDOM).

This means that

We felt this change required an update to the Keyboard Shortcuts dialog.

Which is discussed and laid out in the Interactive Description Design Document, and in a new related issue.

terracoda commented 3 years ago

@jessegreenberg, I think it is safe to close this issue. Please re-open if you need to.