Charles Ergen
Analyst · Craig Moffett with Sanford Bernstein
First, I'll take the first one. Again, the way I analyze SAC, this is me personally, I'm probably the only person who does it this way. I look at and what I give our guys ability to do is based on this is to look at what the total cash SAC is and then what the programming discounts are or the other discounts we give to give to customers. So when we give away $200 of programming, I'd look at our SAC as being $1,000. And then you look at your ARPU and you look at all your churn, you look at all your numbers and say, "Is this an economical model?" When we first started in DBS, that was pretty -- you didn't really do the analysis because the customer was paying for his equipment so you didn't really have any SAC to begin with. So it was all profit. Now as you move closer to that economic analysis, you're not sure if some customers are actually economic for you. You have to think ahead a little bit here. But if you're going to get somebody two years of programming discounts and a two-year agreement, at the end of two years, they're just going to ask for more money. They're going to leave because they have three or four other choices to go to, not the least of which might be the Internet, or they're going to ask for more discounts. So you can't really run a model now where you say, "I get a customer with a $1000 SAC, but don't worry about it, two years from now, he's going to paying me $90 ARPU and for the next 10 years." Because that's not going to be realistic at the marketplace. So when you look at that, you say, "Where do I want to spend my money?" And what I spend my time on it, I want to spend money on something I'm going to get a return on. And so when you look at that, there's less of a customer pie probably for everybody in the industry but certainly for us to go after. It's got to be a little better credit score, got to be a little bit a customer that might be strategically a little bit positioned for your particular company to look at. And so that's why -- so we probably less aggressive than other people in the industry, particularly on discounting programming for any length of time because we're not sure that, that type of customer is going to be a long-term -- be the best long-term investment for us. I'm not saying that you can't make money in that customer. You probably can. But whether it would be the best long-term investment for us, no. Others have a high ARPU or lower churn that an economic analysis will be different for them. But it's not a given that as you go after every type of customer who would actually want your service today. So I think the dynamics are changing in our industry in that we have new competitors from the phone companies and we have new competitors from the Internet. And virtually, anybody can enter the Internet business to be competitive, not the least of which some of our programming partners in sales and a lot of piracies. So we've done a pretty good job of controlling piracy on our end. But the Internet piracy is off the charts free for people. Programmers put stuff on the Internet but they haven't really protected it or don't realize the extent of what piracy can be on the Internet. So we take all those things into consideration and say, we're fortunate we have a large critical mass of customers, a stable business for us, a stable customer base. That was awful lot of cash flow and we're able to hopefully, we're able to maneuver our company to what I see on the opposite side of that, which is less opportunity both in satellite and our current core business, and lots of other opportunities are out there. As far as -- do you have a quick follow-up on that question?
Craig Moffett - Sanford C. Bernstein & Co., Inc.: Yes, reading between the lines then, is it your intent that you're more selective and does that explain some of the change in gross additions?