That was Megaupload, wasn’t it?

If you look at the Sprint network, it’s a direct burial network. It's not in conduit, it’s old. The quality of the glass isn't there. If you're telling me that you're going to invest in the T-Mobile or Sprint network, and then turn around and sell dark fibers, that both Amazon and Microsoft will look at and say, that stuff is junk. It's so old. Consider the generation of fiber; if I’m using leaf fiber and then I'm going to SMF 28 fibers, it’s a different quality of fiber and it’s a different type of glass. If I’m using the older leaf fiber, I may have to put more amplification in it to perform.

Theoretically, if 10 years in the Microsoft IRU have to be future planning, and Lumen would crank up the cost, if Cogent comes into this market and says, our glass isn’t the best, but we’ll offer you the IRU for $150 million, instead of the $50 million you’re talking about, is that something Microsoft would consider? Or is the state of the glass such that they would never consider it, regardless of cost?

It would be a stretch, to be honest with you. Because if I moved to the Cogent network, which is the old T-Mobile or Sprint network, I’d have to move all those repeater connections, and all my gear. I’ve got to bring in a whole new operational group, because I've been operating under Level 3, Lumen for a decade, and now you're going to ask me to integrate that into my network operation center, repair and maintenance and upkeep O&M. Every seven years you change out your optical gear, you do your servers every three or four years, your network every five to seven years, and then the backup generators every 15 years. With Microsoft, the timing would have to be right. If you caught them in the midst of a change, where they’re going to migrate the gear out anyway, then you can propose an overbuild and kill.