I was reading Marguerite Reardon’s piece on CNET just now, which made me think about what it will take for Google to win.

1) Permission to believe on distribution. The list of early partners was obviously newsworthy, but Google is going to have to show continuing momentum here. They don’t have to deliver the entire universe of operators, or even the US, as some have said. But continuing drumbeats of momentum will create permission to believe and free up capital to build new apps.

2) Killer app(s). Every new platform needs at least one user experience that was not available before elsewhere, a reason for customers to buy something. For DOS is was Lotus 1-2-3. For RIM it was email anywhere. This customer desire for an application is what forces customer-owning gatekeepers (in this case the carriers) to forego their natural greed and appropriate desire to maintain proprietary differentiation. Developers, developers, and more developers (i.e. having a range of different offerings for the long tail of humanity) are important, but a very small number of killer apps makes the platform. If you want to know if something has, or is, a killer app, ask yourself whether you would buy it for someone as a present.

3) No friction, developer-friendly distribution and deployment. (i.e. use the Web). Why are there so many more brains writing code for the web than for mobile devices, when the mobile world is potentially so much bigger? Working through carriers to get a mobile app in front of users is a nightmare. If Google can use its leverage and capital to make this simpler (maybe a hosted virtual service exchange? Google’s long awaited answer to EC2?) Android wins, big time.

4) The Right Stack. If the alien stuff supports web/mashup development and ties into Google’s impressive collection of gadgetry this will appeal to the larger webdev audience, if it is also Java based, it will suck in the current mobile app code base which is significantly smaller but more leveraged. Thierry Brethes of Unyverse does a good job of laying out the run time options in his comment to the CNET story. I think it’s likely Google gets this right, and Ajax and Java are both supported, to bring in both the new world and the old.

The CNET piece comes coincident with the “leak” of screenshots of “What’s Open” this morning. I am even more impressed by Andy Rubin’s slow reveal. Again all this for an SDK…it’s really masterful. SteveB declaring Microsoft the incumbent is an appeal to calculated reason, but note that you don’t see Apple (the perceived thought leader in mobile platforms) even acknowledging Google’s efforts at the moment.

Maybe Apple gets that it’s not about facts, but about hearts, minds, permission to believe, and most of all hope. And as most hetero males know so well, few things stoke the fires of hope like a slow reveal.