HP wants this to become a customer showcase, so what kind of pressure does that put on you? It's certainly pressure to talk about things before they are finished. That's probably the biggest pressure, but on the other side, it actually affords a lot of support and help out of the product and R&D community because they see it as being important.

The typical system administrator-to-server ratio has been described by HP as 1-to-20, but company officials have said 1-to-200 is possible. What kind of ratio will you be running? I don't want to comment on that, but I think we will exceed 200.

In terms of making this a customer showcase, how forthcoming will you be as this is accomplished? Will you also share your lessons learned, things that went wrong, the things you could have done better? I think part of our responsibility as being part of Hewlett-Packard is to be very forthcoming on 'here's what we started with, we decided there were problems with this, we changed course.' There will be certainly lessons learned along the way and I don't know what the right milestones will be. But my guess is it will be multiple times a year. Certainly, one of the things you do by making a public statement like you do in a forum as widely followed as this [means you're] going to have everyone asking: 'Where are you at now?"

When do you see the data center consolidation being completed? Our goal right now is to getting to our end of [fiscal year] '08, which is literally Oct. 31, 2008. A lot of it will happen, quite frankly, into next year. We will get some consolidation done at the end of this year; A lot of the heaviest lifting will happen in FY 07 for us.

HP has said it increased capital spending by 40 percent this year to $2.8 billion. What is this allowing you to do? Capital expenditure really gives us the freedom to move quickly -- and too many times that's where these initiatives tend to fail, because they don't get enough capital approved, [or it's] not approved the next year. And what you do is you end up ramping up the cost because you put in part of the infrastructure you need but you don't actually get to the end state that actually drives you to a lower cost of operations.