Ah, Aral posted a few weeks back about Flash Lite failing, putting it down to a failure of "user experience". I've been wanting to write a follow-up to this with some more thoughts, but never quite got round to it. Here we go...

In short, I don't think the issue is one of user experience. Macromedia had Flash Lite running on mobile devices back in 1999 or earlier (I met a company back then who'd been brought in to do the port to Palm, I think), had plenty of time to work through UX issues with the product, and changed most aspects of it many times without success. In my opinion the problem was more a historical and strategic one for them, and it went like this:

1. The Flash player was ubiquitous on the fixed web, but they'd only ever managed to make money out of authoring tools - so felt short-changed;
2. Mobile came along and gave them an opportunity to get revenues derived from player volume: i.e. monitise the razor blades, not the razors;
3. They spent years trying to do this, at the expense of getting the player ubiquitous as fast as possible - ubiquity was the destination, not the journey;

My evidence for this? Mainly, the absolutely batshit business models they pursued: I met a Macromedia guy at the Symbian show sometime before 2005, who told me they'd be charging UK end-users £7 for the player itself - a policy that didn't get promoted too far, and which no-one talks about now. They also approached device manufacturers like Nokia and tried to license the technology to them on a commercial basis. With no installed base on mobile and lots of competition for getting interactive content onto phones (J2ME, Silverlight, Symbian, web) they had little leverage and couldn't negotiate such deals effectively. This meant that the player didn't go far except in markets like Japan, where they did an excellent job of getting in early, working closely with DoCoMo and bundling it onto devices. I'm not sure what the commercial basis for this was, but it let Macromedia claim gazillions of Flash-enabled phones in press releases so I'm guessing terms were favourable to Docomo etc.

To be fair, they were also hamstrung in mobile by the difficulty of upgrading players to handsets (which would help them roll out new versions or fix bugs), by the same sort of fragmentation issues that have affected the Java guys (difficulty of reaching many devices, and quality of implementations), and by a consequent apathy from developers. All this happened whilst the alternatives were out there and doing a better job of getting an audience - J2ME may be flawed as a platform, but it's still a de facto standard, for those that can work with it. Silverlight popped up in the last few years and managed to get some deals - whether or not you think it's a valid competitor, it did compete and probably muddied the waters in licensing negotiations.

Aral touches on the issues of fragmentation and recommends that Macromedia should've delivered a great experience on one device; I actually think they did this with a number of devices in Japan quite successfully. In the West, pre-iPhone there was no one device which could deliver the sort of audience to justify such a strategy, and post-iPhone... well, they're beholden to Mr Jobs on that front.

The end result? A few years down the line there's lots of competition, no real installed base, and few people wanting to produce content. At this stage I'd expect Adobe to be doing things which modify the authoring tools and make them more appropriate for creating mobile content (e.g. compiling Flash into iPhone apps), rather than concentrating on getting the player everywhere - effectively going back to the way they did things on the web, and acknowledging that their attempt to monitise the player in mobile has failed.

Comments

[John Dalziel]

Many moons ago I worked at a company called Pogo. We'd built a mobile browser that ran at a reasonable speed over GSM. More importantly to this post was that the UI was built in Flash and we also supported it in the browser. Back in the early days of Flash 4 there was a SWF specification with no strings attached. Macromedia very quickly altered this to try and monetize players on devices.

Anyways, I took a prototype of the device to Flash Forward in Amsterdam - mainly to gauge developer interest. While I was there I also put it in front of some Macromedia evangelists. At the time they were very excited about their Flash Player for Comapaq so I figured they might help us promote the platform.

The chap I met was weirdly dismissive of what he saw but took all our contact details anyway. A few days later another chap from Macromedia arranged to meet us in London, but he wasn't there to discuss tech stuff or promotion, he was a lawyer sent to shake us down for money. As you said, 'batshit business models'.

[twitter.com/Alfie]

"(e.g. compiling Flash into iPhone apps"

As I understand it they're doing just that with the next CS update - output to iPhone app.

[John Dowdell]

Hi, cross-device work actually preceded Flash... the Macromedia Director engine was actually ported to a variety of computers and consoles.

Biggest reason for the Flash Lite discrepancy was that the emerging class of pocket devices were dramatically underpowered compared to desktop computers. That's why an older SWF profile with fewer features worked for over a billion handsets.

Getting compensated for work is everyone's problem -- finding a business plan in a tumultuous high-growth environment isn't easy. ;-) (That's one of the things Apple did right with iPhone, by cutting out the operators and writing content developers a check.)

I'm still not sure how your title connects, but your final paragraph should be disproven within the year -- Flash's new mandate is to make it easy to deploy to any device, workstation, pocket, or wallscreen. The demand from manufacturers is enormous -- the engineering to *reconcile* those floating APIs is monumental, but proceeding. ;-)

jd/adobe

[Tom Hume]

First off: thanks for commenting, it's great to hear something from the horses mouth to balance my remote gibberings.

Just to be clear: how is the new mandate for Flash (deploy to anywhere) different from the old mandate (run Flash anywhere)? When you say "deploy to any device", do you mean doing this by having a ubiquitous Flash player which plays the same Flash content on many devices? Or are you talking about going down the route that you seem to be going w/iPhone, of taking a great authoring environment and outputting platform-specific binaries?

The latter is what - as Alfie points out - you seem to be doing, but that's a markedly different strategy from the one you've been pursuing to date, and one which doesn't rely on monitising the Flash player (something you've tried and thus far failed to do in mobile). This is the point my last paragraph is trying to make.

[John Dowdell]

Hi Tom, the big change with Player work over the past year-plus has been in making a single engine which works across different devices. There will still be device capability differences (screensize, accelerometer, etc), but the logic and file formats will be consistent across form-factors.

(The "export to iPhone" tack is different, agreed... Apple restricts iPhone to a different runtime, requiring a different file. Adobe's goal in general is to make it easier for creatives to connect with their audiences, wherever they may be, whether the medium is print, film, video, HTML, interactive screen, whatever. Flash is a good means to that end (as most all manufacturers agree! ;-), but is not the only means.)

jd/adobe