A piggyback-slurp terminal tapped in between two legitimate terminals, monitoring the back and forth transmissions. After a time, the piggyback operator knew enough to begin making transmissions on line, slurping out data from both ends, pretending to be GSFC to Houston, and Houston to GSFC. The piggyback terminal could continue to function until one or both legitimate terminals realized that they were being slurped.
Now the question was: how much data had been slurped out in the last seventy-two hours?
He had called for twenty-four-hour scanner checks, and the readings were disheartening. It looked as though the ERTS computer had yielded up not only original database elements, but also data-transformation histories - the sequence of operations performed on the data by ERTS over the last four weeks.
If that was true, it meant that the Euro-Japanese consortium piggyback knew what transformations ERTS had carried out on the Mukenko data - and therefore they knew where the lost city was located, with pinpoint accuracy. They now knew the location of the city as precisely as Ross did.
Timelines had to be adjusted, unfavorably to the ERTS team. And the updated computer projections were unequivocal - Ross or no Ross, the likelihood of the ERTS team reaching the site ahead of the Japanese and Germans was flow almost nil.
From Travis's viewpoint, the entire ERTS expedition was mow a futile exercise, and a waste of time. There was no hope of success. The only unfactorable element was the gorilla Amy, and Travis's instincts told him that a gorilla named Amy would not prove decisive in the discovery of mineral deposits in the northeastern Congo.
It was hopeless.
Should he recall the ERTS team? He stared at the console by his desk. "Call cost-time," he said.
The computer blinked COST - TIME AVAILABLE.
"Congo Field Survey," he said.
The screen printed out numbers for the Congo Field Survey: expenditures by the hour, accumulated costs, committed future costs, cutoff points, future branch-point deletions. . . . The project was now just outside Nairobi, and was running at an accumulated cost of slightly over
$189,000.
Cancellation would cost $227,455.
"Factor BF," he said.
The screen changed. B F. He now saw a series of probabilities. "Factor BF" was bona fortuna, good luck - the imponderable in all expeditions, especially remote, dangerous expeditions.
THINKING A MOMENT, the computer flashed.
Travis waited. He knew that the computer would require several seconds to perform the computations to assign weights to random factors that might influence the expedition, still five or more days from the target site.
His beeper buzzed. Rogers, the tap dancer, said, "We've traced the piggyback slurp. It's in Norman, Oklahoma, nominally at the North Central Insurance Corporation of America. NCIC is fifty-one percent owned by a Hawaiian holding company, Halekuli, Inc., which is in turn owned by mainland Japanese interests. What do you want?"
"I want a very bad fire," Travis said.
"Got you," Rogers said. He hung up the phone.
The screen flashed ASSESSED FACTOR B F and a probability: .449. He was surprised: that figure meant that ERTS had an almost even chance of attaining the target site befure the consortium. Travis didn't question the mathematics; .449 was good enough.
The ERTS expedition would continue to the Congo, at least for the time being. And in the meantime he would do whatever he could to slow down the consortium. Off the top of his head, Travis could think of one or two ideas to accomplish that.
3.Additional Data
THE JET WAS MOVING SOUTH OVER LAKE RUDOLPH in northern Kenya when Tom Seamans called Elliot.
Seamails had finished his computer analysis to discriminate gorillas from other apes, principally chimpanzees. He had then obtained from Houston a videotape of three seconds of a garbled video transmission which seemed to show a gorilla smashing a dish antenna and staring into a camera.
"Well?" Elliot said, looking at the computer screen. The data flashed up:
DISCRIMINANT FUNCTION GORILLA/CHIMP
FUN CT IONAL GROUPINGS DISTRIBUTED AS:
GORILLA: .9934
CHIMP: .1132
TEST VIDEOTAPE (HOUSTON): .3349
"Hell," Elliot said. At those figures, the study was equivocal, useless.
"Sorry about that," Seamans said over the phone. "But part of the trouble comes from the test material itself. We had to factor in the computer derivation of that image. The image has been cleaned up, and that means it's been regularized; the critical stuff has been lost. I'd like to work with the original digitized matrix. Can you get me that?"
Karen Ross was nodding yes. "Sure," Elliot said.
"I'll go another round with it," Seamans said. "But if you want my gut opinion, it is never going to turn out. The fact is that gorillas show a considerable individual variation in facial structure, just as people do. If we increase our sample base, we're going to get more variation, and a larger population interval. I think you're stuck. You can never prove it's not a gorilla - but for my money, it's not."
"Meaning what?" Elliot asked.
"It's something new," Seamans said. "I'm telling you, if this was really a gorilla, it would have showed up .89 or .94, somewhere in there, on this function. But the image comes out at .39. That's just not good enough. It's not a gorilla, Peter."
"Then what is it?"
"It's a transitional form. I ran a function to measure where the variation was. You know what was the major differential? Skin color. Even in black-and-white, it's not dark enough to be a gorilla, Peter. This is a whole new animal, I promise you."
Elliot looked at Ross. "What does this do to your timeline?"