When I was a kid, I had a friend, Abbie, who collected old hardware. Her bedroom was full of the stuff. When I first saw it, I couldn’t see the appeal. Those old boxes were slow and the software was glitchy and you couldn’t play the latest games. And most of them smelled funny – a combination of old plastic and burning dust when you turned them on. Sometimes there were sparks and small fires. But Abbie liked to tinker with them and I learned a few things from watching her. Like the fact that old things often work like simpler versions of new things and helped you to learn how stuff worked. And old boxes didn’t have the same safeguards as new ones, so you could often do things that new tech didn’t allow you to. That’s an idea I’ve used a few times in the years since then. But most significantly I learned that some people just enjoyed the challenge of keeping old stuff working. Computers, operating systems, software, watches, handguns, or whatever.
Abbie would have had a field day on the Celestia but sadly I’d lost touch with her long ago. Given time I could probably have tracked her down – most people are locatable unless they take active measures not to be. But I didn’t need Abbie. There were a lot of people like her out there and I was sure some of them had a thing for forty-year-old military technology.
As far as anyone knew, the Celestia had been lost in battle and no one was sure where. But she had a sister ship that had survived the conflict. The Achilles had probably ended up as a floating museum or a restaurant. At least the Celestia had been spared that indignity. If I could find an enthusiast who had maintained or salvaged the Achilles’s hardware and software, I could get an insight into how the Celestia was set up. I explained my plan to Trixie. I don’t think she was impressed.
“I need you to find me a geek,” I said. “I mean an enthusiast.”
“I know what a geek is.”
Was she looking at me when she said that?
Trixie made a connection to one of the satellites orbiting Saphira and in relatively short order had managed to piece together much of the information I needed. She showed me a three-dimensional plan of the Celestia’s systems, all wrapped around a blank spot that was the lack of details about the vault itself.
There was an impressive refrigeration system that fed ice-cold fluid into the vault. The Navigator obviously needed cooling – and a lot of it when working at full capacity. Though as it was now, with the ship mostly in standby mode, the Navigator could probably get by on passive cooling alone. But an idea was forming in my head that would put the refrigeration system to use.
The Navigator was effectively a self-contained computer system, isolated from the rest of the ship’s network. Data was passed into the Navigator by the ship and the Navigator then passed data out again – and that was it. The volumes of data shared were massive, but there was no interaction beyond that. Try to pass anything other than authorised data streams to the Navigator and the firewalls – either the ship’s or the Navigator’s or probably both – would quarantine the intrusive code and dissect it with lethal force.
The Celestia’s own computer system was a relatively unsophisticated ancestor of a modern system that I was familiar with. The network that connected the ship’s day-to-day systems was more or less isolated from the ship’s security network. This wasn’t unusual forty years ago and it’s still fairly common now. Information technology departments don’t trust security departments and vice versa. Each sees the other as a potential source of risk and each wants to be the one with ultimate control of network security – so they have their own systems that barely talk to one another. While this can enhance security, it can also introduce vulnerabilities. People like me delight in discovering and exploiting these vulnerabilities.
The Navigator, the ship’s computer network, and the security system all passed data to one another, but lacking the ability to rigorously test this data, they effectively accepted it on trust. If security said something was secure, the other two systems wouldn’t – and couldn’t – argue, because they had no way of checking for themselves. This was something I was sure I could exploit. Somehow.
Vaults are specifically designed to keep people out. They are created by people whose job it is to think up possible ways of entry and then make sure that these ways won’t work. Manufacturers even employ thieves to try and break into their newly designed vaults to test them. I’ve never been asked to do this. Which is a good thing, because if people know you’re a thief and know how to contact you to offer you a job, you’re probably in jail.
Prisons are the opposite of vaults. They are specifically designed to stop people from breaking out. Getting out of a prison can be tricky. On the other hand, breaking into a prison is relatively easy. I can say this based on experience, having sneaked into penitentiaries on two occasions on different planets. Prison designers don’t expect people to try and break in. And this set me thinking...
People who build vaults aren’t going to put a lot of effort into preventing someone from breaking out of them. Why would they? In fact, many vaults have systems in place to allow people to get out if they are accidentally locked in. Having a bank employee slowly suffocate in your vault is not the sort of publicity you want as a manufacturer. Authorised persons can get out of a vault relatively easily. I wasn’t sure where I was going with this idea – I was
