Twenty years ago, shortly after I began a career teaching journalism, a colleague and mentor at New York University took me aside to lament the lack of shared historical or literary tradition among our students.

Without these common threads, he reasoned, not only were efforts to teach journalism hampered by missing context, but social discourse had no frame of reference from which to begin.

'Most of our students haven't even heard of 'War and Peace,'' he blurted out in frustration. 'It should be required reading.'

I had neither the heart nor the courage to confess that I hadn't read Tolstoy's classic either. His words seemed arrogant and a bit condescending. Who was he to say what we all needed to know? Yet his longing for some common ground, some sense of cultural connectedness, was a sentiment I shared.

This month, a couple of events reminded me of my ambivalent reaction during that conversation.

The first was a visit to Emerson College's journalism department by Walter Bender, principal investigator of MIT's News in the Future consortium and a passionate believer, as his home page attests, that 'modern telecommunications is leading us inevitably to the smallest news product imaginable: the personalized newspaper, or Daily Me.'

For Bender, no fan of the gatekeepers of the elite press, that day apparently can't come soon enough.

The second was the release of a new Stanford University study on the Internet and society that faced a firestorm of criticism even before the presses had cooled.

The study, carried out under the direction of Dr. Norman Nie of Stanford's Institute for the Quantitative Study of Society, suggested that 'the more people spend using the Internet ? the more they lose contact with their social environment.' Critics, many of them avid Netizens, cried foul. They called the report methodologically flawed, biased and selectively interpretive. But above all, they seemed to suggest that the authors were clueless about new communities forming online to replace some of those common bonds increasingly hard to find today.

I'd be less than honest if I didn't acknowledge that I find the implications of both Bender's vision and Nie's critique somewhat unsettling.

Today, in an era when the verb, 'to nichify,' has found currency in the vocabulary of any media-savvy marketeer, I'm still looking for a coffee shop (a cyber caf? will do) to call home.

And, shallow as it can be, I still find the mix of information, sports and entertainment that passes as our shared news to be among the last common experiences I can talk about with strangers and acquaintances day to day. The Daily Me? As the concept moves in the theatre of news from the wings to center stage, will the self-absorption so prevalent in society be thrust to new heights? Or will personalized news merely give us the freedom to choose the journalistic equivalent of our own favorite novel rather than the 'War and Peace' of a Pulitzer tome we don't wish to read? Will it enhance societal exchange by piquing our curiosity to learn more about the stories that careen by each day? Or will it pull us further apart?

'We're always looking for ways to get around the gatekeepers with their narrow gates,' my friend Mitch Stephens, a journalism historian and author of 'A History of News' reminded me.

Perhaps.

David Shenk, who has written two books on information overload and the challenges it poses, has a less benign view.

'I feel strongly that (customized news) is inevitable, but I think it's very dangerous,' Shenk said. 'The problem is that it is too efficient. Being able to tell a machine what's relevant to you is inherently restrictive in the kind of information you are going to get. It tunnels for all the future your view of the world.'

What happens to the broader issues of community, diversity and social consensus when one neighbor asks only for stories about high-technology research and high-end cars while another limits his requests to stories about gardening and fitness?

'I think one definition of an enlightened culture,' Shenk says, 'is one in which we are aware of other people's perspectives and see the points of view of people who aren't like us.'

Jack Driscoll, former Boston Globe editor and a colleague of Bender in working on MIT's personalized news project, says the choice needn't be either/or.

'The criticism of personalization is that people become parochial,' Driscoll, now an editor in residence at MIT. 'My experience is that people don't just stay with their personalized news.'

And, if they do, he added, what's the difference between them and newspaper readers who zero in on their favorite column or comic and ignore the rest?

The answer, I suspect, will depend on how personalized news sources evolve in the years ahead. I can think of many stories I'd never have read unless my eye had casually crossed the headline or first paragraph. There was this year's six-part series on an infant with Down's Syndrome in The Boston Globe, a New York Times feature on a kid growing up in trailer park, another story on life in the Silicon Valley. None would have appeared in the parameters of my personalized news selections -- unless, that is, my computer had given me options to ask for stories someone else found interesting.

'There are tradeoffs,' cautions Shenk, who spent four years evaluating the pressures of information glut on our lives.

Who has time to read in breadth and in depth? What do we lose when technology helps us find more specialized information? Part of that answer clearly is time, one thing technology has failed miserably to give us more of.

Recognizing this, it is my hope that customized news, when it comes, will be designed in ways that give readers ready access to depth and breadth. Drawing on MIT's experience, here are a few ways personalization can be something other than self-isolation:

1) Make personalized news a value-added component of what we read rather than a substitute for general news.

At MIT, participants in a customized newspaper project called Fishwrap can select specialized areas of interest. But everyone still gets news under the broad categories of international, national, state and hometown. (Whichever category the participant reads more of appears at the top of the listing, Driscoll says).

Fishwrap's 'subscribers' also get to share their favorite stories with everyone else simply by adding them to a community page. In a sense, suddenly the remote, elite gatekeeper of a daily newspaper becomes the multiple gatekeepers of 'neighbors' in a shared enterprise.

2) Find better ways for readers to 'pull' depth out of stories.

During his visit to our campus, Bender told students that some of the most promising applications of personalized news allow readers to augment information in the story to give them a clearer context. Readers, for example, could be assisted by computer databanks capable of providing local analogies that make the news more relevant. For Bostonians, a story on a flood in India might describe the acreage destroyed in terms equivalent to all the land within the inner ring surrounded by Route 128, a well known artery.

3) Remember that earliest piece of technological wisdom: Garbage In, Garbage Out.

Publishers intent on building intricate personalized news systems need to remain vigilant about investing in good reporting. Time-stressed readers interested in depth will get nothing but a headache if they're sent five equally shallow reports from competitors who've forgotten news is found, synthesized and written by intelligent people rather than intelligent agents.

In the end, perhaps the best personalization will be done by those who become their own publishers. At MIT, Bender has worked with senior citizens in the town of Melrose to help them publish their own newspaper. Three years later, the Melrose Mirror is still putting out hometown news of the most local sort. The consumer has become producer -- and community builder.

I wonder: Do you think they have a coffee shop?