Commentary: Will Flash Player Survive?
Recent news that Adobe has finally launched Primetime—which we've covered in its beta form several times—comes with confirmation of several long-lived teasers: most notably the planned inclusion of HTTP Live Streaming (HLS) and Dynamic Adaptive Streaming over HTTP (DASH) in the Primetime Player.
Jan Ozer's excellent recap of Adobe's plans—at least the plans they're revealing before early May's Adobe MAX conference—highlights some of the reasons Adobe chose to add HLS and DASH adaptive bitrate (ABR) streaming capabilities to the overall Primetime ecosystem, alongside its own proprietary ABR technology, HTTP Dynamic Streaming.
The Primetime news, though, coupled with Ozer's additional "story behind the story" solidifies four major points in my mind, which I'd like to share here:
Will Flash Player Survive?
Many major player apps and plug-in architectures have plans to support both HLS and DASH, just as Adobe's Primetime Player will. But curiously, Flash Player will not be one of them.
Ozer notes twice in his article that DASH and HLS will not be added to Flash Player, an odd stance from Adobe as it has demonstrated prototypes of both technologies in Flash Player at the last few NAB shows. Adobe Media Server (AMS) also supports HLS and is slated to support DASH, which begs the question as to why Flash Player will not support these two ABR technologies.
The Player War has Shifted to a Delivery Protocol War
Throughout streaming's 15-year history, the battle has raged between major players, from RealPlayer and Windows Media Player to QuickTime and Flash Player. The fact, though, that Adobe will now join the ranks of other major player applications in embracing both HLS and DASH means that we now have a truce in the player war at least when it comes to which delivery protocol to use (HTTP) and which adaptive bitrate (ABR) technology to use.
Adobe telegraphed the end of the player war in a bit of a roundabout way in its Primetime release press release, stating: "By incorporating HLS into the Flash Player for desktops, TV content owners and distributors will be able to efficiently reach more of their audience by deploying one consistent player."
In other words, since all major players will support HLS, it's time to think beyond the player's ability to deliver bits and more towards the things that truly matter: viewership tracking, interactivity, rights management, etc.
Google Can't Get any Media Implementations Right on Android
Speaking of player implementations, has anyone seen Google get an implementation of any streaming protocol or technology right?
Having tested multiple Android-based devices, I've been disappointed in Google's implementation—at the core operating system level—of RTSP and HLS. In addition, since HLS doesn't reach back to pre 3.x Android OS devices, there's an open invitation for a player that handles basic streaming functionality.
Adobe had a very good thing going with Flash Player for Mobile, as it consistently delivered quality streaming in multiple formats on a handful of Android handsets and tablets. But faced with the massive fragmentation in the Android handset market, Adobe wisely chose to move its efforts towards a browser-based architecture as well as AIR-based apps that also contain the Flash Player architecture.
Real's newer Helix player architecture also has the capability to play HLS content on Android OS 2.x devices. It's almost like you want to tell Google just to leave the media implementations to the professionals, or at least get some help from the YouTube team.
DASH is Getting the WebM treatment, but not on its Technical Merits
One of the major issues surrounding WebM, Google's open-source version of the On2 VP8 codec, was not a technical issue at all. Fear, uncertainty, and doubt, or FUD for short, were raised as valid concerns about the liability and royalty aspects of WebM. So, too, DASH is undergoing a level of FUD as we wait for players.
The number of DASH-compatible players in the marketplace has grown, but the major limitation—as I understand it—is locking down a specific subset of the DASH standard on which to test interoperability. True we're not going to get both MPEG-2 Transport Stream and Fragmented MP4 (fMP4) compatibility in players anytime soon, but we are seeing testing being done around H.264 fMP4 implementations, and the results have been promising.
Now if we can only hack our way through the whole licensing labyrinth—if the labyrinth even exists. I suspect we're seeing a mirage of a licensing maze, and reacting to the mirage in a number of ways that we've been trained to do with past licensing issues such as MPEG-2, H.264, and WebM. DASH is none of these, even though it encapsulates two out of the three, but MPEG LA needs to publicly state this post haste.
To wrap up, Adobe's choice of which ABR technologies to support in only one of its players is a curious one.
I'll be listening carefully in Los Angeles to the Adobe MAX keynotes, and listening even more intently at upcoming Creative Suite events, to find out exactly what WILL be included in future versions of Flash Player and Flash Professional.
As the company balances between its support for HTML 5 in many of its Creative Suite products (Dreamweaver, Flash Pro, the whole Edge sub-suite) and the need to propel sales of Primetime's proprietary technologies (Primetime DRM and HDS), it should make for an interesting summer for Flash Player enthusiasts.
[Note: An earlier version of this article asserted that Primetime has been discussed publicly for two years; in fact it was first introduced in February 2012.]
Everybody's using HTML5 and MPEG-DASH these days, right? Not so fast. Reports of Flash's death have been greatly exaggerated.
Was Adobe's decision to remove FLV and F4V export from Creative Cloud the last gasp for Flash? No, and there are better tools around, anyway.
What the online video industry needs is simple standards for reaching all viewers. But when have standards ever simplified online video?
Adobe officially changes all the Creative Suite products to Creative Cloud. While that might make logical sense, the shift could create problems for end users and, ultimately, for Adobe, too.