Photo

Monitors tracking Internet traffic patterns at the offices of Dyn, an Internet performance company.Credit Nathaniel Brooks for The New York Times

It is easy to believe, and often said, that the Internet is everywhere and nowhere — that it is a kind of data-delivery system based on programming logic, not physical hardware. But the data always originates somewhere in the enormous mass of equipment that makes up the global Internet.

“Every single part of the Internet is real and expressed physically,” said Bill Woodcock, executive director of Packet Clearing House, a nonprofit organization focused on Internet infrastructure and security.“If you’re logging into something, there is a real server that can be touched,” Mr. Woodcock said.Two websites that reporters for The New York Times tracked as test cases were BetCRIS.com, which has been named in multiple federal indictments, and BetOnline.com, a popular site in the United States, according to traffic data.Going directly from a website name to a physical location is usually not possible on the Internet, with a logical addressing system that makes no reference to geography.Continue reading the main story Wired for Profit

In 2006, Congress tried to crack down on illegal online sports betting. Today, Internet wagering is thriving, and a new business that resembles gambling, fantasy sports, is winning millions of players and stoking controversy. The Times, with the PBS series “Frontline,” investigated illegal gambling in the Internet age.

So The Times used a two-step process: First, reporters used commercially available tools, and Internet routing data from several sources, to learn what entities are delivering content for the websites. The second step was to locate those entities — often they are companies that operate equipment within the Internet — at a street address.

Major sources for the Internet data included Mr. Woodcock and Doug Madory, director of Internet analysis at Dyn, an Internet performance company.

In addition, reporters consulted with two academic research groups, the Center for Applied Internet Data Analysis at the University of California, San Diego, and the Duke University department of computer science, to confirm some broad findings — for example, whether content for a website was delivered from inside or outside the United States.

The key to pinpointing physical locations was something called an “autonomous system number,” often referred to in shorthand as an A.S. This acts as a kind of unique serial number that helps companies recognize one another across the Internet.

When a user types in a domain name corresponding to a website, the Internet addressing system provides the A.S. that is hosting or routing the site or otherwise delivering its content. The A.S. usually corresponds to a specific company — say, one of the so-called content delivery networks that operate servers around the world.

Identifying the A.S. with the routing data “is something you can scientifically test and verify,” Mr. Madory said.

From there, The Times used a variety of techniques to find physical addresses. For example, reporters examined publicly available databases in which companies list the locations of their equipment in various data centers.

Graphic The Deception Behind an Illegal Bet A reporter placed 15 bets on baseball, football and other sports with a Panamanian gambling website, BetOnline. Here’s how it worked.