When the Committee on House Life instituted the non-ordered choice in early 1990, it was billed as a three-year experiment which would conclude with last spring's housing lottery.

Next month, the committee--made up of house masters, administrators and student representatives--plans to review non-ordered choice. And several house masters hint that the experiment will remain in place for at least one more year.

Dean of the College L. Fred Jewett '57 echoes the sentiment, saying "If we're going to make a change, it will come later than this year. My personal sense right now is that the system has worked well enough that we can live with it."

Non-ordered choice was supposed to be a short-term compromise, born out of a spirited committee fight. Back then, Adams House Master Robert J. Kiely and former Eliot House Master Alan E. Heimert '49 vigorously opposed an attempt by some of their colleagues and Jewett to fully randomize the housing lottery.

"The masters never exactly got at each other's throats," says Mather House Master Jeffrey G. Williamson, who supported the compromise. "But it's fair to say the debate was lively."

Jewett said in an interview last week that while he will continue to support the compromise, he prefers randomizing all housing assignments.

"By and large, the masters feel there's been some progress," says Jewett. "I feel there are some who would like to go with a randomization system."

According to interviews with half a dozen house masters, a majority of the committee members prefer either non-ordered choice or randomization, and many randomization supporters view the compromise system favorably.

"Harvard spends millions of dollars [on financial aid] to get a diversified student body, so each house should get a diverse group," says Quincy House Master Michael Shinagel, who considers himself a supporter both of increased diversity and non-ordered choice.

"If anything, it's softened some of the house stereotypes," Shinagel says.

Despite growing support for some randomization and the departure of Heimert, Kiely maintains his support for ordered choice, under which students could rank their house preferences.

"I'm still in favor of the old system," says Kiely, who is Loker professor of English. "I think if people who have similar interests want to group together, I don't have any problem with that."

Kiely says the system is at fault for the record number of transfers, which the Adams master terms "an administrative headache." Jewett says last spring's 127 transfers was a result of a "relaxing" of transfer procedures.

Still, Jewett concedes that the housing lottery creates considerable stress for first-years. But he says no compromise is without flaws.

In the fall of 1989, it was Jewett--armed with a set of statistics showing a lack of extracurricular, racial and socioeconomic diversity in the houses--who raised the spectre of randomization as a viable method of diversifying the houses.

He was immediately rebuked by a student group calling itself the Committee Against Randomization and headed by James M. Harmon '93-'94. The committee of first-years argued that student choice should not be neglected at the expense of increasing diversity.

"Every system is flawed," says Harmon, whose group proposed the compromise. "But I think the present system is working well enough to preserve it."