Thursday, February 21, 2008

iPhone Flashpoint

Although I no longer read the daily WSJ, Larry Gagnon of ZDNet pointed me to an interesting article today on the WSJ website (by former CNET reporter Ben Charny) about Flash on the iPhone.

Adobe has successfully created a new platform with Flash, Adobe wants the iPhone to ship Flash. Mobile platforms are a strategically important source of growth for Adobe. As of last October, it had 300 million installs on 430 phone models and 140 device models, and mobile/device revenues (for both Flash and Acrobat) were $52.5 million, up 40%.

So far Apple isn’t cooperating, and went so far as to use a work-around to get YouTube on the iPhone without providing Flash. Today Adobe needs Apple’s cooperation because Apple tightly controls third party apps on the iPhone; even on the ISV-friendly OS X, Apple retains some control such as for apps that require low-level interfaces.

Once the new SDK ships (this month?), the iPhone will become a feasible target for a wider range of third-party apps. At that point, Apple’s attitudes could range from

  1. bundle it and pay a royalty
  2. bundle it if it can get out of a royalty (not very likely)
  3. treat Flash as a strategically important application and help Adobe develop a Flash port for the iPhone — which it then can distribute as it sees fit
  4. treat Adobe like any other ISV (which is to say, not go out of its way to support a very popular application)
  5. do what it can to discourage/prevent Flash for the iPhone, such as by changing APIs or allowing it on its redistribution site.

The WSJ makes it seem like Apple is at #4 or #5.

Neither side is talking, but the key issues would appear to be:

  • Processor usage. Flash is a resource pig, although Flash Lite supposedly runs on a 150 MHz ARM chip. The iPhone has a 620 MHz ARM chip, but that chip is severely taxed right now.
  • Royalties. Adobe gets 20¢ per phone for preloaded software, but nothing if the software is downloaded by the user (e.g. for Symbian or Windows Mobile).
  • Platform control. Flash is a classic closed standard — even more closed than Windows — in that Adobe controls the APIs and no third party implementations are available. By comparison, Apple supports Adobe’s PDF standard on OS X but has its own implementation of a PDF reader that lags Adobe’s by several generations but for most purposes is good enough.
IMHO this last point is the most important one. Flash is a prerequisite to using an increasing number of websites, as I found trying to go without for several years. (Still, it was only GrandCentral that caused me to finally install it). Apple is gambling that Adobe needs the iPhone more than the iPhone needs Flash.

John Gruber has a long post about why he believes Apple will not help bring Flash to the iPhone. It appears that we don’t agree on much, but we agree here:
Apple doesn’t control the HTML/CSS/JavaScript web standards, but neither does anyone else. And Apple does control and own WebKit, which is by anyone’s measure the best mobile implementation of these standards today.

Flash, on the other hand, is (from Apple’s perspective) the wrong sort of proprietary — owned and controlled by another company.

Gruber notes (correctly) that Flash may have won on the desktop, but the mobile web is still wide open.

The iPhone has an opportunity to influence those mobile standards far out of proportion to its market share (of about 6% of new smartphone sales). As Google told the Financial Times last week, Google “had seen 50 times more searches on Apple‘s iPhone than any other mobile handset”.

My bet is that for 2008, Apple will continue to promulgate WebKit and work with Google (Ajax etc.) to make websites work well with the iPhone using open standards. I also expect Adobe to ship its own Flash player (with or without Apple’s help) in Q2 or Q3 of this year.

The real issue is whether mobile-oriented websites will develop assuming a Flash player, and then require iPhone users to download a copy of Flash Lite — or whether they will follow the Google-iPhone web application model.

No comments: