Some of my notes:
Steve is being very Steve – emphatic, in selling mode, drinking a massive Starbucks iced coffee. Discussing why Bing, why now he says Microsoft decided search needed a new brand, but: “Building a brand is a step, step, STEP (almost yelling)!” Bing is a brand that consumers can get their mind around. We have to grow significantly relatively to where we are. We have to build brand equity as well as technology equity.
Walt Mossberg asks why “Bing”? Not Bing Crosby?! No, Steve says, it’s short, it works globally, they could get the URLs, makes for a good brand.
On to the demo. Yusuf Mehdi joins Steve on the stage. 
Live version goes up June 3.
He shows the Bing home page. Steve interrupts: “This is not something that pushes us to 80%” market share. But it’s got some good stuff.
New features:
They point out that 25% of clicks on search is on the back button – IE, folks are not getting what they want. So they are focusing on fixing that.
“Best Match” – the one that is most authoritative. Does a search on Cannes film festival and the results are very good, including better answers instead of “following a million links”.
The snippet is far more advanced.
A search for UPS shows just what Bing thinks most folks are looking for. Including customer service numbers. Live search for Microsoft shows their customer service number. That gets applause from the crowd.
The focus is on head end and service related answers like for flights, etc. It’s going over well.
Another focus is helping with research and tasks.
“Taking a bet that search engines are not designed well to help with research.”
The left nav bar is a big change. It uses a lot of key space to do navigation – essentially taking core search and laying an application interface over it. This is an important new approach.
I like this – the idea of seeing search as a core dataset upon which a more sophisticated navigation interface is overlaid. I’ve been talking about that for a while….
The big difference is that Bing brings most of the web woven into the search interface, instead of the way most of us search now, which is to go back and forth, back and forth. 
Universal search and integration of images, videos, verticals is quite good.
Key verticals supported so far are travel, commerce, and I think health…though have to check on that.
Advertising as a business model on the internet is less proven that conventional wisdom, says Steve, talking about new models of making money with search, including doing deals with content providers to integrate into Bing. This raises the spectre of paid inclusion, worth pushing into on that front…
We’re not trying to live off other people’s work….says Steve, dig at Google.
Yusuf leaves stage, Steve sits down with Walt again. Walt asks will this really change things? Steve says his timeframe is lots of years.
Google is market leader. But Bing has “real differentiators” that will bring new folks to the engine. Walt asks about Ask spending money on marketing but then dropping off after showing some share gain. Steve says they plan to stay in the game long term (in terms of supporting brand with spending), and keep it up. Steve says it was a big budget. He had to gulp when he approved the budget. A gulp at a $60billion company is a big gulp.
Distribution is key (they are buying distro, more on that later – see my post predicting Microsoft would gain share back in january).
Steve called it “the little engine that could.” Walt points out it’s a pretty big train….