Graham-Cumming founded Plan 28 in 2010, a project which intends to build the Analytical Engine based on Babbage’s papers. Humans Invent spoke to Graham-Cumming to find out more about the Analytical Engine, how far his team have come in building it and why Lord Byron’s daughter, Ada Lovelace, was the only person to truly understand the potential of Babbage’s machine, perhaps even more so than the man himself.
Babbage incorporated this decision-making capacity into his design. He talked about a snake eating its own tail.
That really comes down to answering the question, ‘what is a computer?’ as opposed to, ‘what is a calculator?’ If you think about a pocket calculator, when you ask it to do a sum, it gives you the answer and does nothing else. It has no ability to examine what it has done and make a decision and fundamentally what computers can do is look at the output of the calculation and make a decision. It is this ‘if then’ ability that allows computers to make decisions and therefore do different stuff that can be used for everything from making a decision in a game about when a bad guy will appear to deciding what is going to be displayed in a web browser. It is that decision-making facility that is the fundamental issue.
Babbage incorporated this decision-making capacity into his design. He talked about a snake eating its own tail. The ability for the Analytical Engine to look at what it has calculated and examine it and make a decision where to go next.
How far have you come in building the Analytical Engine?
First of all we are making sure that Babbage’s design was complete, so it really could be built and we think the answer will be yes. As long as the answer is yes, then we will try and build it. We’ve been quite careful about giving an accurate decision on when we will complete it, considering there are a lot of things to discover but I would have thought it was something like a ten year project, to do all of the work, the research on Babbage’s papers, simulation and then actually building the device.
Did he create a lot of papers for you to work with?
Yes, if anything, there are too many papers. Many people will have seen the Difference Engine Number 2 which was reconstructed by the Science Museum and there is also a copy of it in the Computer History Museum in California. With this one Babbage left behind almost one complete set of plans. Unfortunately, when it comes to the Analytical Engine he left behind hundreds of different plans and thousands of pages of notes about ideas he had so a really big first part of Plan 28 is to sift through these to find instructions to build something that Babbage would recognise as his machine.
Lovelace realised that just because you were using numbers inside your device it didn’t mean you were restricted to just thinking about numbers.
Well, I think she is important for two reasons. One, she is important today because a lot of what we know about the Analytical Engine in terms of public exposition of it came about because she translated a paper into English and added a bunch of notes. These notes, which were actually longer than the original paper, really explained a lot about what Babbage was thinking.
I think the other thing is that during Babbage’s lifetime she was the person who most grasped what the Analytical Engine was about and what it was capable of. What is really interesting is that she grasped something that Babbage himself didn’t. Babbage was building the Analytical Engine to do mathematics and what Lovelace realised was that just because you were using numbers inside your device it didn’t mean you were restricted to just thinking about numbers.
A computer that everyone has today is fundamentally doing arithmetic underneath – those numbers are used to represent music and images and text. Ada Lovelace realised this and explicitly said we shouldn’t be limited just to doing mathematics, she even said one of these machines could be taught to compose music.
The intriguing question is how much did Turing know about Babbage when he did his seminal paper? We think the answer is not a lot but he came to learn about Babbage afterwards and fundamentally his machines were different in terms of construction because, of course, by that point we had electronics and he went down that route whereas Babbage’s was completely mechanical. In another sense, they’re identical because, as Turing noted, it doesn’t matter how you build a computer, they all have the same capabilities – that is what a Turing Machine tells you. In this way Babbage’s machine was a Turing machine as much as the latest PC is.