Question about documentMode about polyfill-service

[triblondon]

Good point. Seems switching in and out of compatibility mode changes the UA, so that would be supported, but document mode doesn't, so isn't. Would be interested to hear about possible implications of that and any workarounds you can suggest.

Hi, by the way.

[tmaslen]

Hello :-)

DocumentMode is detectable in the DOM, we do this...

var browserType;
if ('querySelector' in document && 'localStorage' in window && 'addEventListener' in window) {
    browserType = 'html5';
}
else if (
    (document.documentMode && (document.documentMode < 9)) ||
    (/MSIE (7)/.test(navigator.userAgent))
) {
    browserType = 'legacyie';
}

We load in a slightly different JS app if its legacy IE.

It's an adaptation of our original CTM but for the interactive specials content I mostly work on now.

/t

[triblondon]

OK, so if you cared about that, you could sniff those things and then set the UA query param. Otherwise all we have to go on is the user agent header sent by the browser.

[tmaslen]

Yes that would work. DocumentMode is definitely an edge case and causes IEs to behave strangely. I've stopped using conditional comments because of it as they stop reporting accurately too.

[triblondon]

Would you consider using this service?

[tmaslen]

I can't speak for the whole of BBC News and I'd have to look at it more but it does sound like we're trying to address similar issues as the FT.com right now. We'll need to support at least IE8 with all of our modern JS. Speak with John Cleveley, he'll have a better idea of what is going on for the responsive news' support of IE8.

It's definitely something I'm considering for my bespoke interactives.

[triblondon]

Wilco. Closing this then - if we have problems in future because we need to deliver different polyfills to the same IE UA depending on what its documentMode is, we'll have to have another head-scratch.