> Linn had finally overreached its finances and the project was clearly at risk, but the finalstraw was a Linn delivery driver called Shug. He reversed into Davids Porsche and Ivorunwisely decided that, since the incident had happened on private ground, Linn werentresponsible and wouldnt pay for the repairs. David quit and "chucked all that I had in theway of bits and bobs of hardware into the Forth and Clyde Canal."
David Harland is a very interesting person. Why would a professor of computer science at St Andrews University suddenly drop out of computing entirely, and move to writing about space exploration?
It turns out he has strong opinions about the state of modern computing: https://news.ycombinator.com/item?id=8217780
Here is the one technical bit, the rest is a very entertaining job story.
> The thing about the REKURSIV was that users could program different instruction sets. James Lothian, who was worked at Edinburgh University microcoding the Prolog instruction set describes it as being "a really interesting and unusual design: the main memory was in effect a persistent object store, with every object having its type, size and position in memory known in hardware, so that (for example) the hardware could prevent you from 'running off the end of an array and corrupting surrounding memory. Paging of objects into and out of main memory was handled by the host machine (generally a Sun 3), and was completely transparent, even at the microcode level. This meant that you could write arbitrarily complex algorithms in microcode, even recursive ones, hence the machine's name. Every object had a unique identifier (a 40-bit number), and the MMU chip would translate that into the object's store address (if it was in main memory). Since only the MMU knew the object's address, an object could be moved around in memory without having to update references to it (since they were in terms of its object number); this made garbage-collection particularly straightforward."
This is, of course, complete hearsay, but Tim Rowledge ( https://news.ycombinator.com/user?id=timrowledge ) told me once that the Recursiv processor was used in a prototype british "smart torpedo." The official story was the CPU locked up during a test and the torpedoes have never been seen since.
> The project originated in an initiative within the hi-fi manufacturer Linn Products to improve its manufacturing automation systems, which at the time ran on a DEC VAX minicomputer. This resulted in the design of Lingo, an object-oriented programming language derived from Smalltalk and ALGOL. Due to the poor performance of Lingo on the VAX, a subsidiary company, Linn Smart Computing Ltd., was formed to develop a new processor to efficiently run Lingo.
This brought to mind the J. Lyons & Co. (a 1950s UK restaurant chain and food manufacturer) who built the LEO computer [1] to run their own business, and ended-up manufacturing them as the first computer designed for business use. A big departure from their main business.
I keep meaning to go back and look at it and see how practical it would be to write up an emulator or an implementation on FPGA. But the absence of any software to run on it for testing would make it tricky.
> the processor instruction set supported recursion
I think that's one of the most interesting parts. It would be interesting to see this implemented in a more general processor's instruction set (like x86). I wonder how much performance boost there would be in some algorithms if you could compile down to a set that included a recursive instruction?
It would be interesting to learn a little bit more about what this means. X86 "supports recursion" in the sense that it has a stack and the push, call, ret, syscall, etc. instructions make use of it.
I read the paper, which includes a forward reference to Figure 1 (the system architecture). When I got to the end of the paper, it says that Figure 1 is "TO BE SENT SEPARATELY BY MAIL, TOGETHER WITH A PHOTO OF ONE OF THE CUSTOM CHIPS". Oh well.
"Due to the poor performance of Lingo on the VAX, a subsidiary company, Linn Smart Computing Ltd., was formed to develop a new processor to efficiently run Lingo."
Ivor is well known for being single minded and opinionated. He'd proved his heretical and opinionated point about turntables against the general industry view in the sixties, and built a successful and highly profitable business on the back of it. From nothing.
From all I've read and heard of him, it's just the sort of thing I might expect. :)
Hah!