Why Adobe Should Change its Mobile Strategy (again)

[Where is Adobe really heading with Flash in mobile? Guest blogger Guilhem Ensuque deconstructs Adobe’s recent AIR and Flash mobile strategy and argues why Adobe should go back to the drawing board]
The article is also available in Chinese.

Seen from the outside, Adobe’s mobile game plan is an extension of the same strategy that took them to near-ubiquity in the desktop browser. It’s about putting the Flash Player everywhere for free and cashing-in on the designer and developer tools – plus distribution and analytics services (see the Omniture acquisition). Adobe bets its mobile future on taking the Flash runtime to a forecasted 50% of smartphones by 2012, according to the company.

This strategy has worked well in the past for Adobe in the browser and desktop space. The mobile business is however a completely different animal – which is why Adobe’s strategy will fail. Here’s why.

The two iterations of Adobe’s mobile strategy
Adobe’s mobile strategy v1 was Flash Lite. It has enjoyed massive deployments – more than 1.2 billion devices to date according to VisionMobile’s 100 million club. From a financial standpoint however, Flash Lite royalties represent less than 1.5% of Adobe’s overall revenue.

More importantly, based on discussion with people familiar with the matter, I would estimate that only ~3% of Adobe’s 1million+ mainstream Flash developers customers have been creating Flash Lite content (although no public data is available).

What’s the lesson here ? It’s that subsidizing the Flash Lite runtime penetration into 40-50% of devices did not translate automatically in developers adoption. From the developer’s point of view, Flash Lite indeed lacked a direct content/apps distribution channel in the pre-App Store and “walled gardens” era. It also had different APIs compared to the “full” Flash, and integrations in OEMs handsets were fragmented.

Adobe’s Mobile Strategy v2 was announced in May 2008 as a complete reset of their Flash Lite strategy, aiming to address these obstacles. With the Open Screen Project (OSP), the mainstream Flash Player (v10) and its sibling the AIR runtime are now at the center of the Flash Platform “galaxy” across all types of terminals – desktop, smartphones, TVs, and more.

With this strategy reset, Adobe is going back to square zero to infiltrate the mobile device market with a consistent runtime. Adobe pledges to waive royalty fees for partner OEMs who are collaborating in the Flash/AIR integration effort on their platforms, ensuring over-the-air updateability and consistency. In addition, OSP partners allow distribution and monetisation of Flash content and AIR apps through their app stores (and also through Adobe’s own Distribution service).

Adobe v2 strategy is in essence a pledge to its key customers – organisations like digital agencies paying for design tools and media outlets paying for flash video delivery servers. A pledge that the Open Screen Project will extend the reach of their current technology and people skills investments to the mobile masses – and succeed where Flash Lite hadn’t before.

Sounds good on paper, but …

Time for a reality check
Almost two years after the launch of the Open Screen Project, the results are somewhat lukewarm:

• Flash Player 10.1 is still not released publicly on any mobile device (expected some time in 2010 for some platforms).
• Flash Player 10.1 will not run on existing Symbian S60 devices (reading between the lines of this Symbian Foundation announcement and that Nokia whitepaper).
• Flash Player 10.1 will not run on existing Windows Mobile 6.5 devices (see here).
• Flash Player 10.1 will only run on devices that have the very-latest mobile chipset architecture (Cortex-A8 or above – see here).
• And finally, Apple has slammed the door shut in Adobe’s face with the iPhone and now iPad staying decidedly Flash-free.

In short, it is not yet possible to deploy “full” Flash content to any mobile device. And even when Flash Player 10.1 is finally released, it will only address a minority of smartphones. The promises to OSP partners and Adobe customers have yet to live up to their expectations. At the same time, Adobe is planning (hoping?) that Flash will be deployed on 50% of the smartphone base by 2012.

Adobe developers looking for alternatives.
While waiting for the OSP to bear fruit, developer mindshare has been drifting towards alternatives with proven market traction – most notably the iPhone’s XCode/Objective-C environment, developing apps on demand from brands or for sale on Apple’s App Store.

Bridge technologies have also emerged to fill the void. For example, OpenPlug’s ELIPS Studio allows Actionscript developers to build native cross-platform mobile apps, while Appcelerator’s Titanium and ANSCA Mobile’s Corona are trying to lure Flash developers away from Actionscript to Javascript and Lua programming, respectively.

The Google-Adobe “co-opetition”
Adobe places high hopes on its OSP partners, first and foremost Google, who demo’ed Flash on the NexusOne in Eric Schmidt’s keynote at Mobile World Congress. Adobe is banking on the success of Android as the vehicle to deploy Flash far and wide into smartphone land.

Yet betting on Google’s helping hand is far more risky than it sounds. Adobe is in fact competing with Google in many areas:

• Advertising & Analytics: The more ‘paid search’ dollars go into Google’s AdWords/AdSense, the less dollars go into premium ‘display advertising’ campaigns designed by digital agencies that buy Adobe technologies. Google Analytics also competes directly with Omniture, Adobe’s latest USD 1.8 Billion acquisition.
• Online services: Adobe has launched the Photoshop.com online service, betting on its brand to attract consumers who want to edit and share photos. No luck, Google has just acquired Picnik – another image editing service. The two also compete in other online services like document sharing or web conferencing.
• Web standards: Google is a heavy backer of HTML5 (in W3C and with Chrome). HTML5 is seen as a long-term alternative for Flash on the web, including for video delivery.
• Video: Google’s YouTube is already repurposing its content to the MPEG4/H.264 format instead of Flash Video so that it can play on the 50+ million iPhones/iPods. And an HTML5 version of YouTube has just been launched. Google has also acquired On2 technologies, developers of the VP series of video codecs and holders of a significant patent potfolio. Some, like the Free Software Foundation, have called upon Google to opensource this codec and free it from patent royalties. Such a move would resolve the current HTML5 video codec dispute and make the proprietary Flash Video redundant.

Other Adobe partners hedging their bets ?
Beyond Google, I would argue that Adobe’s strategy with the Open Screen Project is putting it on a collision course with too many other players for it to succeed.

As Francisco Kattan (former Adobe exec) puts it on his blog:

“The strategy to differentiate with applications is not limited to Apple. RIM and Samsung have made recent moves that point to their aspiration to differentiate their devices with applications (although neither can afford to pick the Flash battle at this time; their positions are under attack by Apple and Google and are too busy playing defense)”.

Indeed Samsung’s Bada and RIM’s SuperApps are vertical propositions that attempt to create “stickiness” with application developers through proprietary APIs and distribution systems, going against Adobe’s ambitions for a consistent horizontal environment.

To second Francisco’s analysis, I would also add Nokia to the list of Adobe Open Screen Project partners who are at the same time buidling up a vertical applications and developer ecosystem with OVI and its associated runtimes, Qt and WebKit.

The major OEMs are not sitting idle and waiting for Adobe to deploy Flash/AIR and own the service delivery platform. They are trying to replicate Apple’s formidable hardware+software+services model while grappling for developer and consumer mindshare.

The fallacy of a unified mobile user experience
Adobe’s unified Flash Platform strategy is built on the premise that designers and developers (i.e. Adobe’s key customers) can create content that spans all terminals, form-factors and interaction experiences – thanks to a ubiquitous consistent runtime.

Yet, the freedoms enjoyed by Flash developers in the context of windowed desktop user interfaces and multi-tabbed browsers simply cannot apply in mobile phones environments.

There are driving forces in the mobile space beyond the developer or designer’s control that constrain what is displayed “on the glass”. For example: limited screen real-estate, end-user need for coherence between different applications, interaction with the native device UI for priority events like incoming calls, OEM or operator branding, design guidelines pertaining to most application stores, the list goes on and on.

What needs to be on Adobe’s drawing board
Instead of coaxing its OSP partners into adopting the Flash Player and AIR in all their designs (a “cathedral” vision), Adobe should embrace the diversity that they offer (a “bazaar” vision) – by using its tools to expose the strengths and idiosyncrasies that make each mobile device different.

Tools are Adobe’s core strength (and a massive proportion of its revenue). Photoshop and Dreamweaver are the de-facto tools when it comes to graphics and website design. A success that Adobe built without having to put any proprietary image codec or browser on every desktop.

Adobe is already experimenting with a new tools-centric approach rather than a runtime-centric Flash/AIR strategy. Here are two examples:

• The CS5 packager for iPhone, allows Flash designers to create native iPhone apps.
• The joint effort with Nokia allows WRT widget development in Dreamweaver.

Along the same lines, Adobe needs to go back to the drawing board and redesign its mobile strategy around a consistent toolset, rather than a consistent runtime. Here’s how:

1. Open source the Flash Player code, and hand over its governance to an independent organization (ala Eclipse). This would share the costs of Flash Player development, its porting to devices, and increase reliance on Adobe’s tools. Note that Adobe has taken already some timid steps in this direction with the handing over of the Tamarin virtual machine to Mozilla.
2. Massively contribute to open source HTML5 browser implementations (WebKit or Mozilla, or both). This way, Adobe would regain credibility and influence, to balance the dominance of Apple and Google in that area.
3. Make Dreamweaver the best toolchain for mobile web apps by providing extensions for the multiplicity of mobile widget APIs (JIL, Nokia WRT, BONDI, Palm WebOS…) and making sense out of the Javascript/CSS UI and MVC frameworks quagmire at the tools level.
4. Push the envelope with Catalyst. An area currently underserved in mobile is the designer-developer workflow. The new Catalyst tool from Adobe is very promising here, especially if it’s blended into the native user experiences afforded by the underlying mobile platforms.
5. Make Flex the best toolchain for native mobile apps by severing its dependency on the FP/AIR runtimes, giving it the ability to build natively (like the CS5 packager for iPhone) and exposing the specific native services and UI components of each platform (from iPhone to BREW). This is particularly promising as the Flex toolchain (ActionScript / MXML / Flex Framework / Flex Builder IDE) is light-years ahead of the existing mobile C/C++ SDKs in terms productivity and capabilities. It also holds great potential for bridging fragmentation between mobile platforms.

In conclusion, Adobe’s runtime-centric mobile strategy with the Open Screen Project has been based on naïve assumptions: going back to square zero post Flash-Lite, and sowing the smartphone fields with a consistent, heavier runtime across all terminals. But this strategy is suffering from implementation delays and from its dependency on partner co-opetition. More importantly, it proposes to developers a horizontal contents & applications distribution vector at a time when the industry is going vertical. Adobe should instead refocus its resources on a tools-centric strategy allowing “Write Once, Tweak for Any, Build for All”.

- Guilhem

[Guilhem Ensuque is Director of Product Marketing at OpenPlug. He has more than twelve years of experience in the areas of mobile software and mobile telecoms. Guilhem was a speaker at last year’s Adobe MAX conference. His favorite pastimes (beyond mobile software strategy!) include making his newborn daughter smile and sailing his Hobie Cat with his girlfriend. You should follow Guilhem on twitter @gensuque_op]

Guilhem Ensuque

Guilhem Ensuque is Director of Product Marketing at OpenPlug. He has more than twelve years of experience in the areas of mobile software and mobile telecoms. Guilhem was a speaker at last year’s Adobe MAX conference. His favorite pastimes (beyond mobile software strategy!) include making his newborn daughter smile and sailing his Hobie Cat with his girlfriend. You should follow Guilhem on twitter @gensuque_op