It means no longer scheduling his school day around teachers' availability to supervise -- and no more mockery from classmates.
"They may laugh at some of the hobbies that you like," Willis said. "I didn't really feel comfortable looking at sites I really wanted to."
Willis, 13, exemplifies the difference having Internet access at home can make. If people without home access are classified as disadvantaged, the "digital divide" is larger than recent studies suggest.
According to the Commerce Department, 54 percent of Americans used the Internet in September, up from 44 percent in 2000 and 22 percent in 1997 with increases among all races, income levels and educational backgrounds.
Though there's disagreement over how much work remains and what role government should play, the numbers show that when it comes to basic access, the online population is looking more like America in its diversity.
But those numbers can be deceiving.
Much of the focus so far has been on getting Internet access to schools, libraries and community centers. Federal programs like e-rate, funded through a telephone surcharge, have helped get more than 95 percent of public libraries and 98 percent of public schools wired.
New research from the Pew Internet & American Life Project finds, however, that 12 percent of Internet users can only log on from work, a library or some other place away from home.
While overall home access reached 44 percent of the U.S. population in 2001, minorities and lower-income Americans were less likely to have it, according to Commerce.
For example, half of blacks and Hispanics who use the Internet at public libraries can't log on from home, compared with only 30 percent of whites and 22 percent of Asians.
Among all kids ages 10 to 17, less than one-third of blacks and Hispanics have access at home while at least two-thirds of whites and Asians had it.
Schools, libraries and community centers can be important for training people so they feel comfortable enough to eventually own a computer.
But public spaces alone aren't adequate, given their limited hours and terminals. As well, some schools discourage nonacademic use, hobbling students who might otherwise develop interests online that could sprout into careers.
Then there is privacy. A pregnant teen-ager who wants to get information online before confiding in a parent is less inclined to seek it at a public terminal.
The Willises, a close-knit black family in Brooklyn making less than $25,000, relied on a nonprofit program, Computers For Youth, for a free computer and access.
With the computer centrally located in the living room, Willis can pursue his interests in drums online and help younger siblings with homework, while other family members read and play games nearby. That bonding is difficult outside the home.
His other, Lori, studies the Bible and looks up recipes and poems. She found the small chunks of computer time at the library disruptive.
"Whenever I started getting into something, I had to quickly save that information," she said. "You lose your train of thought."
Policy-makers and researchers who study the digital divide are beginning to pay more attention to location inequities and other factors beyond mere access.
And that just makes the problem more difficult to tackle.
"If it is in the schools, libraries or community technology centers, then this is something for the community," said Mark Lloyd, executive director for the Civil Rights Forum on Communications Policy. "We have a great deal of difficulty making the argument that we ought to subsidize poor people's access" at home.
Richard Akeroyd, who runs library grant programs at the Bill & Melinda Gates Foundation, agrees that access at home is better. But a computer in a library can benefit 20 to 30 times as many people each day.
Inequities also exist in skill. Without training, minorities are less able to plumb the Internet for relevant and empowering content.
There are also inequities in how people get access.
New York University professor Anthony Townsend found that in urban areas, lower-income neighborhoods are less likely to have access to high-speed services and must settle for dial-up connections.
Rural areas are also less likely than cities to have high-speed services.
Those differences are likely to become more pronounced as more online services are developed with high-speed access in mind.
The current criteria of measuring the digital divide through access alone grew out of comparisons between the Internet and the telephone - a device with fewer functions, noted Andrew Blau, a technology consultant who advises nonprofit groups.
A better approach, he said, is to compare Internet use with literacy.
Comparing access "does a real disservice to understanding and seeing the real issues," he said. "We don't think, 'If everyone had a book, they would be literate.'"