Why we don’t use popcorn.js even though we love it

November 2012

Playing with popcorn.js by Mozilla lately, I was musing at length whether and how it could be a com­pon­ent of our video strategy at Zeit Online. To come straight to the point – I don’t think it makes sense for us, at least not now. I’ll explain in a moment.

But first let me say that popcorn.js is lovely. If you browse the demo section, you’ll stumble over very inter­est­ing pro­jects. Popcorn.js, even at its early stage, is a fas­cin­at­ing play­ground and I’m eager to see it grow.

So why do I think it doesn’t work for us? Two reasons.

1) Not everything that’s fun to play with fulfils an actual demand 

Looking at the variety of popcorn.js demos, I’m torn between being impressed and scratch­ing my head. While I admire the devel­op­ing skills behind those pro­jects, I doubt that real life user beha­viour played a role in the cre­ation process. From a video per­spect­ive, many of these pro­jects will simply sink in user testing. 

Watch­ing an elab­or­ately edited video piece with dense nar­ra­tion while being dis­trac­ted by emer­ging maps, wiki­pe­dia content or rapidly dis­played weblinks is painful. Human atten­tion is limited by nature. That’s why we’re so bad at multi-tasking. It’s not helpful to demand more from viewers than they could pos­sibly accom­plish; even more – you’ll risk lower­ing the impact of your core story being told by video. 

Two import­ant addi­tions here. First, audio is a dif­fer­ent story. Check out Sound­cloud and how they display com­ments for single tracks. As long as there aren’t too many of them, com­ments on the timeline can be fun to watch. There’s no other visual stim­u­lus around, so no com­pet­ing for attention. 

Second, don’t confuse what popcorn.js is doing with the idea of a second screen for live tv/video events. Olympics, Red Bull space jump or US elec­tion night have one thing in common: People can afford to let go of focused watch­ing for a moment and check Twitter, Face­book, apps, whatever. Even more – they’re thank­ful for dis­trac­tions as soon as a lengthy live broad­cast has its less excit­ing moments. 

2) Don’t confuse our audi­ence with people watch­ing video on our site in a desktop setting

If you haven’t done it yet I’d encour­age you to check out the popcorn.js demo pro­jects with an iPhone. You will see that mobile is not a top pri­or­ity for the popcorn.js dev com­munity. No sur­prise here. 

Popcorn.js is a power­ful tool for build­ing inter­act­ive stages and exper­i­ences around video – it needs room to unfold. That’s why the desktop browser is its natural home (at least that’s how most developers use popcorn.js so far). There’s nothing wrong with that, of course, at least as long as you’re not aiming for maximum reach. If you do, everything that’s not hap­pen­ing within the video is a problem. 

Don’t get me wrong – from a the­or­et­ical per­spect­ive I love the idea of unbund­ling ele­ments like text, sub­titles, maps etc. from a video. Reality, however, is a beast. We’re deliv­er­ing video (with dynamic ads) on our desktop website, with flash and HTML5 players from Bright­cove depend­ing on your browser choice; we’re addi­tion­ally deliv­er­ing video to mobile and tablet browsers, to our iPhone app and to our YouTube Channel. 

Even this straight­for­ward deliv­ery with nothing but video files is tech­nic­ally chal­len­ging at times; we’re not sat­is­fy­ing every of our users on every device and we’ve still got a long way to go. Would we put future resources in a tech­no­logy that doesn’t fully work on all of our deliv­ery chan­nels and doesn’t speak with the plat­form we’re mon­et­iz­ing with? I doubt it. 

You might see popcorn.js at Zeit Online at some point, however, since we like to exper­i­ment. But I don’t see it in pro­jects where we strive for seam­less, maximum impact. 

I’d love to hear your thoughts about popcorn.js and whether you agree or you don’t. I think the fun­da­mental idea behind popcorn.js is smart and forward-looking, so do me a favor and debunk this post. If you reply to this tweet or to this post on app.net, I will gladly quote you here.