The Great Disconnect

FIRST CAME COMPUTER PIONEER JOHN KEMENY. NEXT CAME THE MACINTOSH REVOLUTION, WHICH DARTMOUTH HEARTILY EMBRACED. NOW THE COLLEGE EMBARKS ON ITS THIRD WAVE OF COMPUTING INNOVATION BY BLANKETING THE CAMPUS WITH A FIRST-RATE WIRELESS NETWORK.

AS THE LAST MILLENNIUM TICKED SOUNDLESSLY into the new and Dartmouth's thousands of well-prepared computers slipped bug-free past the predictions of worldwide software failure, a lack of technical innovation nonetheless remained a nagging problem on the Hanover plain. Few were willing to say it, but the College, long known as a leader in academic computing, had fallen behind. In the previous decade Windows had overtaken Dartmouths beloved Macs. The World Wide Web had appeared, then blossomed elsewhere. Both the Clinton White House and Pizza Hut had Web sites up and running before Dartmouth did. The College had launched its own Web site by 1996, to be sure, but was that good enough? Had its early reputation as a digital innovator become part of its history? Had the "Kemeny legacy" been justaone-time bequest, or was it still the active fund of inspiration that John Kemeny, president of the College from 1970 to 1981 and world-renowned computer pioneer, had always intended it to be? And if it were the latter, when and how would it reappear?

The answer came in the spring of 2001, and it came almost unheralded, as if the residents of a small city awoke one morning to find their aging publis transit system suddenly gone anti-grav- ity: The subway system was still there and you could ride it sta- tion-to-station if you wanted to, but why would you when instead you could instantly teleport yourself from any corner to any build- ing, or vice versa. Dartmouth's third computing wave had arrived: The entire campus had gone wireless. Now any Dartmouth user with a laptop computer and a wireless card could get high-speed Internet access anywhere on campus—every building, every space outside—without having to plug into a network port. Anyone could read e-mail from the porch of the Hanover Inn or the sundeck of the Skiway or in between Frisbee tosses on the Green. No other Ivy League school had anything like it. Nor, in fact, did many institutions at all, public or private; most had a "hot spot" or just a few whole buildings with wireless access points. Dartmouth suddenly had 300 Cisco Aironets covering the entire 200 acres (coverage that was complete but subject to localized overload—it now has more than 600 access points and another 400 coming). So quickly had the College adopted this new technology that 57 colleges and universities sent their best information technology people to Dartmouth's "Unleashed" conference on campus last October to learn more about it. With keynote addresses by President James Wright and former FCC Chairman Reed Hundt, the conference was both a hands-on demonstration of the network and a clear signal to all in attendance that Dartmouth had again stepped to the front. Or had it? Some conference attendees weren'tyet ready to commit to the expense. No doubt Dartmouth's wireless network is a grand new feature, they said, but is it one that actually has educational benefits? Or is it something less, a new-tech way for students to send e-mail while walking between classes or to order pizza from a fraternity parking lot?

"Innovation means new, not necessarily better," says Chris Jernstedt, professor of psychological and brain sciences and director of Dartmouth's Center for Educational Outcomes. "You can't just throw it out there. In Michigan they gave every teacher a computer. It was a waste of money; most had no idea how to use it in actual teaching. But this wireless network here has the possibility of making things vastly better."

Jernstedt, who's been teaching at the College since 1967, at first wasn't aware that the new network had any value to him. That changed when he called Larry Levine, the College's director of com- puting, who had sold the idea to then-Provost Susan Prager. "Larry went out on a limb to install the wireless system, making an expensive resource available before many could see the need for it," says Jernstedt. "I found out about it because I went to him with a general need and he pointed me to this specific solution. I started using it immediately."

Jernstedt s "general need" was every teacher's: He wanted more and better responses from the students in his classroom. He equipped each student in his Psychology 22: "Learning" class with a wireless-enabled Handspring Visor PDA (personal digital assistant), donated by the manufacturer. Then, whenever he asked a question, none of the 80 students had to raise a hand and vie for attention—all of them could answer immediately by tapping the responses into their PDAs, which transmitted and projected all the answers to a screen in front of the class,-with instant feedback in both directions. "Not only could they all respond," reports Jernstedt, "but they all did. You go from a 50-minute class with an average of seven responders out of 80 to a 50-minute class with 80 responders out of 80. And not only do they get the feedback, but so do I. A high percentage of wrong answers tells me I haven't made the point clear."

Jernstedt calls this process "brain-based learning," a term that's not redundant to a brain scientist. "When the brain is truly en- gaged," he explains, "when it's actively forming a connection—memorizing or recalling—that process is a physical one. Neuronal connections are formed and real learning takes place. And the brain is much more actively engaged when answering a question than it is when passively listening." With the wireless networks ability to actively connect every student to every question posed during a class, even if that class has gathered on the grass in front of Dartmouth Hall, Jernstedt says that all teachers will have to adapt to it.

But how soon they'll adapt is unanswerable. Because for every Jernstedt, there's probably a Luddite such as the unnamed faculty member (identified only as "a scientist") quoted in the College's 2000 Self-Study Report on the Computing and Information Environmentat Dartmouth who described the Internet as "a vast waste of bandwidth" and who suggested his comments be included in an appendix titled "The Old Fogey Mindset."

The bulk of the faculty certainly falls somewhere between these two extremes, but just where is difficult to measure. What scant evidence there is seems to point somewhere toward the old-school end of the spectrum. Though no one seems to be keeping a tally, it appears that the number of Dartmouth classes with a Web site where students can find a syllabus, updated assignments and an instructor's office hours are outnumbered by those classes without. It's not for lack of tools. Academic Computing makes available to every teacher, with training, a customized off-the-shelf course-management program called Blackboard. But last fall only 160 out of 400 classes were using it. On the other hand, that's up from 75 the previous year.

Why have so few of the Colleges professors taken their courses digital? Some would ask it the other way: Why have so many? In a major liberal arts college with undergraduate degrees awarded in 54 majors, computing for many of its teachers is just another discipline. It may have a national reputation and a rich history, they'll argue, but it's still just that: another field. Not ours.

Most of those who administer computing at Dartmouth to- day would argue that's wrong. Computing at Dartmouth isn't the field, they'll explain, it's the fertilizer, and it's spread over many academic fields. Keep it fresh and spread it thick, they believe, and lots of stuff will grow. It's up to the individual users, they argue, to sort the weeds from the crops; our job is to keep the whole place up-to- date and ready to bear fruit. The new wireless network is as fresh a layer as anyone has anywhere. But of course it's not that simple. Not at a 235-year-old institution with 580 tenure-track faculty, where one professor's welcome layer of technological help is another's un- necessary pile of bovine ordure.

"Look," says Professor Jernstedt, "we knew 25 years ago that the lecture model didn't work. Everybody knew it. Has that changed the way most teachers teach? No. Does that mean we're going to quit working on better models? No."

At Dartmouth it's a good bet that those better models will be, in one form or another, computer-based. It's been that way for almost a half-century.

"YOU GUYS OUGHT TO DO TIME SHARING."

That's exactly how he said it, recalls retired mathematics professor Thomas Kurtz of his 1961 conversation with John McCarthy, the man who five years previous had coined the phrase "artificial intelligence" while he was teaching math at Dartmouth. Now at MIT, McCarthy had led a team in programming a Digital Equipment Corp. PDP-1 computer to simultaneously process the tasks of more than one user, a then-revolutionary idea. Unable to get MlT's full backing for the "time-shared computing" concept, he tossed the idea to his former Dartmouth colleagues Kurtz and John Kemeny. We accepted the idea full bore," says Kurtz, co-inventor of the BASIC computer language with Kemeny and the first director of the Dartmouth Computation Center.

Three years later, in College Hall basement space taken over from the College photographer and armed with a half-million dollar grant from the National Science Foundation, Kemeny and Kurtz installed a GE computer in February. Assisted by a team of student programmers led by John McGeachie '65 and Mike Busch '66 (who "swarmed over the new hardware," according to Kurtz), they set out to teach the machine to run the newly written BASIC language in a time-shared environment.

"In only two-and-a-half months," says Kurtz, "the 'kids' got the time-sharing system to service two teletypes and run a simple BASIC program. The official legend is that this all took place at 4 a.m. on May 1,1964. Mean time to failure was about five minutes. However, by the middle of June the system was 'reliably' servicing 11 teletypes, well enough to allow Kemeny to introduce the new 'baby' to the faculty. And by fall there were 20 terminals."

By the fall of 1968 there was a whole new building, the Kiewit Computation Center, a $650,000 gift from Peter Kiewit '22 and his wife, Evelyn. The 100 simultaneous users of its GE-635 running the Dartmouth Time Sharing System (DTSS) included not only Dartmouth users—who reserved time in advance and signed in to use the terminals in half-hour increments, taking their output with them in rolled-up sheets of yellow teletype paper—but students at seven New England secondary schools: Philips Exeter, St. Paul's, Mount Hermon, Vermont Academy, Philips Andover, Hanover High and Mascoma Valley Regional. It was an initiative that eventually would spread to 30 colleges and 20 schools in New England, 12 colleges in Canada and scores of educational and research organizations throughout North America. Kemeny s three-part vision—user-friendly campus-wide computing, access without charge for faculty and students, encouragement of computer-aided instruction across the entire curriculum—was already in place. The "Kemeny legacy" was born.

General Electric, seeing the potential for a more commercial legacy of its own, renamed DTSS the "GE Mark I" and started offering it to the rest of the world. For computing at Dartmouth, and at General Electric, it was a technological high-water mark, albeit a short-lived one. Individual users sharing time on large comput- ers would eventually be replaced by small computers sharing time on large networks.

BUT IF DTSS WAS DARTMOUTH'S TECHNOLOGICAL peak, it was far from its innovative one. "Dartmouth isn't a technological innovator," says Kurtz today, looking back across his 44 years of involvement, "its a usage innovator."

And the consensus peak of the Colleges next usage innovation happened precisely at the marker everyone either shrinking from or hurrying toward the future had been waiting for: 1984. In one of the most memorable commercials ever aired on television, Apple flung its hammer into the Big Brother face of the IBM personal computer. And Dartmouth joined the revolution. The trustees approved the inclusion of the cost of an Apple Macintosh in the basic fee structure of the College, thus allowing financial aid to pay for it. Between April and November the entire campus was wired with 2,600 connections, a "port for every pillow," as the saying goes. By the end of the year there was an online Dartmouth Mail Directory to send electronic mail to anyone on campus, and a little more than half the card catalog of the library system was searchable over the network by typing "c lib" into any DTSS network terminal, on or off campus. It would take another year and a half to develop, but work was already under way on DarTerminal, a filetransfer program to integrate all those user-friendly Macs into the same system. Dartmouth was by acclaim the most wired, most plugged-in, most computer-soaked liberal arts institution on the planet, and it seemed to have happened overnight.

But swift as the adoption of the Macintosh was, itwas still slow by some others' reckoning. The PC had been available since the hob- by-kitAltair 8800 began shipping in 1975, complete with avariant of Dartmouth's own BASIC as its native programming language. That language wasn't Dartmouth's, though. It wasn't anybody's. Ke- meny and Kurtz, believing fervently in making every aspect of com- puting open and free to every user, had released it into the public domain, a fact not missed by the writers of the Altair variant: Paul Allen and Bill Gates. Within the year Gates dropped out of Harvard, joined Allen in forming Microsoft and rewrote a copyrightable version of BASIC and began licensing it. By 1977 it was everywhere: in Apples, Radio Shack Tandys, Commodores and IBM PCs. Every-where but Dartmouth, that is. By then, computing at Dartmouth had moved on.

"When the Apple lis and IBM PCs came out, we looked at them as retrograde technology," says Kurtz. "Dartmouth students already had full terminal access to mainframe computers using higher-order languages with virtually unlimited capacity, day or night and without charge. I remember picking up my first Apple II—I think it was 1983—and the BASIC it was running took me back 20 years."

But the Mac was different. Not only was its icon-based interface vastly easier to use, with its 32-bit processor and built-in Apple Talk networking it was the first PC that had the power to be hooked into the Dartmouth network. Adoption by students and faculty—with the notable exception of Tuck School, which had ordered 50 IBM PCs the year before—was immediate and universal.

Levine, the current director of computing, arrived with the Mac in 1984. As he tells it: "That Mac was a 128K, crash-all-the-time, two-application masterpiece. Coming from Indiana, I was 'Mr. DOS.' I clucked and snickered at the Mac. But then I saw how quickly it won people over. And it didn't win over just DOS users; much more important, it won over people who had wanted nothing at all to do with computers." Levine became director in 1991, at the height of the Mac monoculture. "Through about 1994," he says, "Dartmouth was essentially all Mac, and arrogant about it."

What broke the marriage? Like all divorces, the answer depends on whom you ask. But without doubt the two main outside factors were the growth of the Internet and the market domination of the Windows operating system. Dartmouth was slow to react to the former, reluctant to accept the latter. By the mid-19905, however, it had no choice.

"If we weren't as quick with a Web presence as some others," says William "Punch" Taylor, Levine's predecessor as director of computing, "it was because we were very enamored of the platform-specific client software that we had written ourselves. Our Macs were seamlessly running everything from DTSS emulation to Blitz Mail on our own network. So when we first looked at those early Web browsers it was partly ego and partly 'why switch to an inferior platform?' that caused us to hold back."

The same attitude—if it's less good, why go there?—was even more firmly entrenched in the Mac monoculture, and with good reason. For 10 years there hadn't been a person—faculty, staff or undergraduate—on campus who wasn't doing everything on a Mac. The problem, of course, was that most of the rest of the world was doing it on something else, 90 percent of them on Microsoft Windows. A great university sends its graduates out into the world with a full box of tools, not a narrow set of prejudices. The Mac monoculture had to go, even if it went kicking. In the fall of 1998, 80 percent of incoming students still chose Macs. By fall of 2002, 80 percent chose Dells, a ratio that has held steady since. And in January 2002 Provost Barry Scherr and Treasurer Win Johnson '67 decreed a three-year Mac-to-Windows migration for all administrative computing at Dartmouth. The king was dead.

Change was, and remains, inevitable. Dominance in comput- ing is subject to overthrow everywhere, almost by definition, and even more so at a place such as Dartmouth. For all its long and rich his- tory of computer innovation, the positive effects have largely fallen outside the College. Gates took BASIC and started Microsoft with it. Apple Computer reinvented itself with the Mac, aided in no small way by Dartmouth's nationally trumpeted early adoption of the ma- chine. Today Cisco's wireless networking business may be about to enjoy the rewards of the same stamp of approval. There is, of course, nothing wrong with this—new knowledge and academic certification are what a great educational institution is supposed to provide to the world around it, free of charge. Kemeny knew that. But he also knew that while those companies fattened their bottom lines, the College's store of intellectual capital would grow even more. a

ED GRAY learned BASIC at Tuck School, where he graduated in 1971, andhas been accessing Dartmouth computers, one way or another, ever since. Helives in Lyme, New Hampshire.