[At this point, the interview switches gears and locations and moves into Steve White's office for his perspective as chief technology officer. ]
ECONTENT: You came in a year after MSNBC.com launched. What were some of the pressing issues when you came on board?
STEVE WHITE: I came in as a program manager. It was really fascinating back then because, first of all, we had no other model, really, than the Microsoft model of product development, and the tradition at Microsoft of what we called the strong PM model, which was basically PMs drive everything through.
There was a lot of debate back then about news. That news is not software and that running software and the development of all aspects of the news business on the Web can't operate like a traditional software company. Interestingly, when I arrived, that kind of internal epiphany had resulted in a shift towards very rapid development-actually managing a situation where we're developing software as part of the news publishing cycle.
So we-being very analytical software guys used to two-year dev cycles-would go through waterfall analyses and realized that we had to basically create a new environment for shipping the back-end product, which was a rapid, "Do it now" approach. We, in all the years since then, have kind of come off that precipice, that constant crisis mode. Now it seems pretty routine to us.
ECONTENT: So things have leveled off. You've got some of your systems in place, but it seems like there's still this constant innovation happening where there's a lot that still can be done as far as developing these in-house kinds of tools to shrink the process down more-make it more efficient and also at the same time, bring up the level of presentation and interactivity.
WHITE: I tend to focus more on the latter because I'm not involved in the day-to-day production workflow process. My mind tends to wrap around [issues like] what will the broadband model at home be in three years? One set of statistics we tend to enjoy is the fact that we used to think six months ago that broadband was a strategic thing. With the help of Nielsen data, we've kind of come to the startling revelation that our audience is now 60% broadband at work and 25 at home, and they're roughly equal audiences, so you're up near 40-45% broadband connectivity today. So what's strategic about that? That's kind of tactical.
ECONTENT: How does the percentage of users who have broadband access map out against the broader geographic numbers-like the number of users in the United States who have that kind of access? Lately, there's been a lot of bad press about broadband, which has created this "hurry up and wait" mentality.
WHITE: An interesting pattern has emerged. The adoption rate is at the high end of all the projections [from] Jupiter, Forrester, everyone. That almost never happens. Those guys are always optimistic. So if the adoption rate is 25% today for us-25% of our home audience is broadband-and 15% of the United States households are broadband, I believe that's the right number, so we're already an increment above that. It's a categorical thing. It's not just Microsoft. It's the news business. It's the product. We are right in the thick of, "It's here. It's now. It's basically half our audience."
ECONTENT: How about the distribution side? We just visited Akamai last week. They partner with you, don't they?
WHITE: Yes, in a business sense. We use them pretty extensively.
ECONTENT: You traffic a lot of very high-bandwidth content. Is that the model you use to physically distribute that kind of stuff?
WHITE: Yes, on the serving side, absolutely. We totally believe their business model. We do not want to be in that business, so they're a great partner in that sense. It's very important to distribute this stuff out and not be trying to serve it all out ourselves. We have to keep Akamai involved for big news, elections and things like that. Those business models are really challenging, and I'm glad they're taking on that challenge because we don't want to have to scale out for maximum capacity, and we also don't want the users to have to go through all the hoops to get to MSNBC. com. So for both of those purposes, the Akamai relationship serves us well.
ECONTENT: Then there's the other distribution side. Obviously, MSN is a good partner for you guys. How else are you getting your content distributed? Who else are you partnered with?
WHITE: MSN is clearly the largest. We've had a relationship with and actually expect to continue a good relationship with the NBCi property as it evolves and figures out where it's going next. The argument we use with our partners-and it's been a successful argument for MSN-is, MSNBC is great for them. They get good content, [they get to] work with our editorial team, and we give them live feeds of news and breaking news instantaneously. [We] make it really easy, kind of a no-brainer for our partners. It enhances their site. MSN embraced that wholeheartedly a little over a year ago and had incredible growth even at their size. So getting the user engaged wherever they are [through] news, entertainment, or gaming-there are a lot of great things to get people pulled in there, so then you can upsell them. That's been really successful with MSN, as I've said. We've done a number of deals with other properties-Prodigy, for instance-and some of the broadband people to get content on their site, especially broadband content. That's been successful, but as you know, the business models are very fluid. They're not nailed down at all. But our philosophy is MSNBC.com should be free, and that this is a good thing for other Internet properties. It's a good thing for us, obviously, because we get traffic. We can brand things. We can cut all kinds of deals to generate revenue, but in the end it's about getting the audience hooked in there, and so that distribution model has worked really well.
ECONTENT: One last question. What are some of the ongoing technological challenges?
WHITE: The big technical challenge, the one thing about the Web that is unique, especially in the news business, is scalability. [When] every new person comes, it's more bids, more servers, more ASP queues, more everything. It isn't like TV where you go "click" and the whole system doesn't even notice. Scalability in this medium is going to be a big issue. There are so many really fast lanes in the infrastructure out there, and there are so many really annoying little bottlenecks along those high-speed lanes. That's probably the biggest challenge, and it starts at our server farm. It even actually starts back at the publishing system. So everybody's learning that, and we're kind of at the front lines because MSN doesn't get these big spikes. MSN gets big traffic everyday, but we're not going to build out an MSN-size farm to manage the one day we get the same traffic they do. So scalability is a constant challenge.