The launch was already running late as it was--almost seven years late. Congress gave the Galileo project its stamp of approval way back in 1977 and the probe, which had been budgeted to cost about $400 million, was scheduled to be launched in 1982. However, things began going wrong almost from the start. In 1979, NASA pushed the flight out to 1984 because of shuttle development problems. Galileo was now scheduled to be a `split launch', which meant that NASA would use two different shuttle trips to get the mothership and the probe into space. By 1981, with costs spiralling upwards, NASA made major changes to the project. It stopped work on Galileo's planned three-stage booster system in favour of a different system and pushed out the launch deadline yet again, this time to 1985. After a federal Budget cut fight in 1981 to save Galileo's booster development program, NASA moved the launch yet again, to May 1986. The 1986 Challenger disaster, however, saw NASA change Galileo's booster system for safety reasons, resulting in yet more delays. The best option seemed to be a two-stage, solid-fuel IUS system. There was only one problem. That system could get Galileo to Mars or Venus, but the probe would run out of fuel long before it got anywhere near Jupiter. Then Roger Diehl of NASA's Jet Propulsion Laboratory had a good idea. Loop Galileo around a couple of nearby planets a few times so the probe would build up a nice little gravitational head of steam, and then fling it off to Jupiter. Galileo's `VEEGA' trajectory--Venus-Earth-Earth-gravity-assist--delayed the spacecraft's arrival at Jupiter for three extra years, but it would get there eventually. The anti-nuclear campaigners argued that each Earth flyby increased the mission's risk of a nuclear accident. But in NASA's view, such was the price of a successful slingshot. Galileo experienced other delays getting off the ground. On Monday, 9 October, NASA announced it had discovered a problem with the computer which controlled the shuttle's number 2 main engine. True, the problem was with Atlantis, not Galileo. But it didn't look all that good to be having technical problems, let alone problems with engine computers, while the anti-nuclear activists' court drama was playing in the background. NASA's engineers debated the computer problem in a cross-country teleconference. Rectifying it would delay blast-off by more than a few hours. It would likely take days. And Galileo didn't have many of those. Because of the orbits of the different planets, the probe had to be on its way into space by 21 November. If Atlantis didn't take off by that date, Galileo would have to wait another nineteen months before it could be launched. The project was already $1 billion over its original $400 million budget. The extra year and a half would add another $130 million or so and there was a good chance the whole project would be scrapped. It was pretty much now or never for Galileo.