RSS

Fullscreen mode in MinnoJS

Sometimes it is usefull to have your task run in fullscreen mode. One way to do this, is to ask your participants to press F11 (or ^Cmd+F on Safari). A much more fluent way to do this is by having full screen activated by the player automatically. There are two problems that we need to sovle in order for this to happen. First we need to figure out how to request fullscreen mode using raw Javascript. Then, we need to find the way to integrate it into our tasks.

Getting to fullscreen should have been easy: simply call document.documentElement.requestFullscreen() and presto wer’re done. But the fullscreen API varies somewhat across browsers, with many requiring vendor prefixes and/or not implementing the latest specification. Therefore we cannot use it directly and have to find a more comprehensive solution. A quick google search brought me to this Stackoverflow answer, which offers a simple script to normalize all vendors and even support legacy browsers (read: Internet explorer prior to IE11). With some modifications, it looks like this:

var el = document.documentElement;
var rfs = el.requestFullscreen || el.webkitRequestFullScreen || el.mozRequestFullScreen || el.msRequestFullscreen;
if (rfs) rfs.call(el);
else if(window.ActiveXObject){
    // for Internet Explorer
    var wscript = new window.ActiveXObject('WScript.Shell');
    if (wscript!=null) wscript.SendKeys('{F11}');
}

Note however, that this API only works during user interaction (i.e., after the user interacts with our page), to prevent maliciously switching to full-screen in ways that might confuse the user. Therefore we need to call it from within a user event such as a mouse click or keypress.

How do we integrate this code in a MinnoJS study? Typically, our study would start with a consent form, and that might be the best place for triggering the full-screen mode. We could add a message with a fullscreen button and have users click it. But, it might be nicer if the fullscreen mode would start as soon as we continue from the consent to the next page of our study. Let’s do that.

We could add a message with a fullscreen button and have users click it, but wouldn’t it be nice if proceeding from the consent page would automatically activate full screen? Let’s do exactly that. We begin by creating a standard message task. We then add a load hook to run our custom Javascript after the message has been loaded and its elements are available. Within the hook, we grab the proceed button ([pi-message-done]), and attach a click listener to it. When the proceed button is clicked the listener will be triggered, and the code (described above will run).

{
    type:'message',
    template: [
        'This is where the consent text goes',
        'Be sure to note that commencing the study will switch the browser to fullscreen mode'
    ].join(''),
    load: function(){
        document.querySelector('[pi-message-done]').addEventListener('click', function(){
            var el = document.documentElement;
            var rfs = el.requestFullscreen || el.webkitRequestFullScreen || el.mozRequestFullScreen || el.msRequestFullscreen;
            if (rfs) rfs.call(el);
            else if(window.ActiveXObject){
                // for Internet Explorer
                var wscript = new window.ActiveXObject('WScript.Shell');
                if (wscript!=null) wscript.SendKeys('{F11}');
            }
        });
    }
}

Some older versions of Safari do not support keyboard input when in fullscreen mode. If you are writing a study using keyboard input, and expect users to use older devices you should explicitly note this problem within the instructions.