Upcoming Industry Conferences
Streaming Media West [13-14 November 2018]
Live Streaming Summit [13-14 November 2018]
Streaming Forum [26 February 2019]
Past Conferences
Streaming Media East 2018 [8-9 May 2018]
Live Streaming Summit [8-9 May 2018]
Content Delivery Summit [7 May 2018]
Streaming Forum [27 February 2018]

In Defense of Autoplay, an Unloved and Unmourned Feature
While most people cheered when some browsers deactivated autoplay on videos, our columnist lamented. Here's why this is actually a step backward for the web.

I am familiar with being the only person in the room holding a certain viewpoint—I have been banging on about things like multicast, virtualisation, and cellular bonding for many years. I often find myself in situations where I love a new, emerging, and obviously disruptive technology at its very start, and I butt heads with intransigent “incumbents” and Luddites for years before enjoying the satisfaction of being right all along. So I am pretty thick-skinned when my instinct tells me something is right or wrong, regardless of the popular consensus of the day.

That said, am I the only person out there who was frustrated by the recent deactivation of autoplay in most web browsers? I am very surprised by the silence in our industry when it comes to the introduction of an extra click between the users’ search for content and our ability to deliver it to them.

In the first decade of streaming, each and every extra click or key press to enable a stream was considered bad. Those were the days when the internet wasn’t yet synonymous with “the web.” The internet was an uncountable number of telecom networks connected together, and the telecom networks were being used not for words and images, but for sending spoken word audio around the world.

Full duplex had transformed communications nearly half a century before. Push-to-talk was seen as primitive—that extra click might make the technology simpler, but it reduced the “usability” of the technology. That was bad.

And now I have to lean forward to click “play” once I open a YouTube link, or open my favourite internet radio station’s webpage? This is extra effort. This is simply less usable.

Oh, I can hear the objections already. “But did you not get annoyed by all those autoplaying adverts?” No, I didn’t. I was quite capable of muting my browser, and anyway, for years, most adverts have been playing mute unless hovered over.

“But they still play when you are not expecting them, and they disturb other people in the office when they do.” Frankly, if you have a sound card on your machine, then plug some headphones in. Why would an advert you fail to control disturb someone in the office any more than a professional training video? It should be office policy to use headphones, not web browser policy to prevent videos from playing until you actively start them.

Autoplay is essential for many online media channels. Internet radio is a great case in point. What else do you expect when you open an internet radio station’s webpage? I fully expect their audio to stream directly at me. But now that is no longer an option.

We have fallen into a hole of making everything that we access through the web into a “pull” technology. Still, people talk about traditional live TV and online on-demand TV. There are those who think that all internet-accessed media is purely on-demand, as if the net could not be a synchronous/live technology. As if we could not deliver real-time data over a telecommunications network.

Back in the earliest days of the internet there was a great browser called PointCast. It was a client for a push technology. It was the Twitter of the day. The screensaver could be configured with your choice of real-time information, and it would sit passively listening for updates and display them as soon as they were received. Today, Twitter constantly queries for updates, as do so many similar “live updating” services. Because PointCast listened passively for updates, it was very scalable and ripe for wider IP multicast deployment to really allow it to demonstrate that scalability.

Consider that we are only now seeing organisations looking seriously at multicast some 25 years later. That’s 25 years of innovation we lost because something that was clearly going to become important wasn’t considered important at the time. After all, Teletext systems were widespread in the traditional television infrastructures, and so work focusing on using the internet as a push-capable medium was left to dry up once PointCast itself failed (due to commercial, not technical, issues).

One of the last bastions of that push-enabled internet to survive, until recently, was autoplay of media in the web browser. It was, until this year, always possible to simply open a webpage and have multimedia streams sent to video players and to audio players. Synchronised media was possible, simply by the act of opening a webpage.

Now if I want my separate audio stream to line up with my video and my score app on my webpage, and I want to stream them all roughly in sync with each other, I either have to converge all that data at the remote end before I send a composited stream to the end user’s single application (which she only has to press play once on), or I am faced with trying to convince my user to press play on three different media streams at the same time. Yes of course there are ways to converge that data back to a simple single “play” button, but why has the option to source several streams and have them play at the same time been taken away from me simply because browser consortiums are afraid that the noise of “push adverts” might annoy someone who couldn’t find the mute button on his machine?

I say it is a careless move, and it is unnecessarily making something important harder to do. It is an avenue of innovation we are rushing to close, and for unclear reasons.

I would go so far as to suggest that now some folks who have homescreen bookmarks on their phones for directly opening their favourite internet radio station as they drive to work are now at risk of trying to find and press a play button on the webpage while driving, and that could go so far as to cause fatal road accidents.

That alone bothers me more than someone not able to press mute on their device. I fully expect to be flamed for that point of view, but this is the View From the Edge!

[This article appears in the Autumn 2018 issue of Streaming Media European Edition as "In Defense of Autoplay."]

Related Articles
Google's attempt to clarify its decision to drop H.264 from Chrome in favor of WebM creates even more questions than it answers
Pundits have pounced on Google for dropping H.264 support in favor of WebM in the Chrome browser. But what if an all-H.264 world isn't all it's cracked up to be?
Google is phasing out support for the Netscape Plugin API in the coming months, so Smooth Streaming providers need to come up with a strategy soon.