Incremental updates to a failed (sales-wise) product is not a viable business strategy. You don't release products and then measure to see if that's what the market wants, and then continue to make changes in the hope you eventually figure it out. It's like a war. If you send in your jets and they get knocked out of the sky by AA fire, do you say, "Let's send in another squad from the left... now the right... let's try higher this time... maybe lower...." Pretty soon you have no aircraft left. If you get hammered on the field you have to go back to the drawing board, figure out what you did wrong, and try something different - quite possibly very different. Incremental changes are for achieving incremental improvements. When there's a cost involved in time, money, market share, and mindset, you don't set out to go from flop to success in 20 incremental products. You won't survive. If a network sees a popular show beginning to sag, it changes a cast member or a writer. If the show premieres as the lowest rated series on the network, they yank it after three episodes and try putting something else on in its place.

I always say that all I ever needed to know in life I learned from horse racing. Think of a coin toss game. Heads - you get a dollar, tails you lose a dollar. In the long run you can expect to break even. Horseplayers (and businesses) are not in it to break even. Now let's say the rules are heads - you get two dollars, tails you lose a dollar. Now you have a positive edge. You want a minimum boldness strategy - bet small amounts for as many plays as possible. You don't want to bet too much - if you have a $3 bankroll and you bet $1 a flip, you have a 1/8 chance of tapping out despite the edge. You want to be like a casino and grind out a profit flip after flip. That's Apple, like I discussed above. They have the positive edge and want to milk it with safe updates. Now let's say it's heads - you win 50 cents, tails you lose a dollar. Now the game is against you. You have a negative edge and a maximum boldness strategy is advisable. Your edge is negative 25 cents per flip. In the long run you're going to lose 25 cents for every dollar bet. That's what Nokia's looking at now. A large sequence of small bets is suicide. In this instance, the smartest thing to do (if you have to play) is bet all the money you're going to play on one flip. That way you have a 50% change of making a profit, vs. a 100% of losing money in the long run. (With one or two possible exceptions, every bet in a casino has a negative edge, so your best strategy is to put all your money on one bet. Yes, I'm no fun bringing to a casino. )

This is what Elop is saying. If Nokia keeps dallying and playing around and releasing experiments like the N900 and fiddling with Meego, they're going to lose market share and completely lose the U.S. market and the smartphone market. He's suggesting it's wiser to try something bold then to do nothing. His metaphor had more flames and less coins than mine, but it's the same point. It's wishful thinking that they can afford to continue on the same path. By the time they come up with something viable on that path, they'll have lost capital and market share to the point where it might not matter; they might be too weak in relation to their competitors to be able to successfully execute it. I think it's all but certain that, at least for the U.S. market and probably beyond, Mr. Elop will view partnering the expert hardware maker struggling for a software response to market challenges and the expert software maker struggling to find a powerful and unique phone to showcase its software as a good idea.
I'd love to have a phone that can run desktop apps, but that's not going to happen. In four years it probably will happen when the hardware is there, but I don't think it's going to happen soon. People want their widgets and their pinching and zooming and the bragging rights associated with having to swipe three fingers 30 degrees, place their noses against the screen and blow, jump up and down to set off the accelerometer and touch the screen to their right nipple to install an application. And they want an "app store" so they can browse lots of software and install it in one click... except if it's their desktops and its Linux; then they don't want it. In four or five years we'll have desktop-grade CPUs in portable packages and people will know better and not be wowed by giant cell phones minus the phone (tablets) just as they're hungover from netbooks and trying to figure out what was so awesome about an Atom and a tiny screen that they had to have one of those in the first place. The N900 will be looked back on as the grandfather of those new phones just like the Apple Newton and "handheld electronic organizers" unsuccessfully preceded smartphones. When Nokia's no longer in danger of being left behind then it will begin to create more experiments like N900 again and hopefully be ready to lead the PC-in-your-pocket revolution when its comes.