Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!ukc!mucs!logitek!hrc63!mrcu!uk.co.gec-mrc!paj From: paj@uk.co.gec-mrc (Paul Johnson) Newsgroups: comp.misc Subject: Software Patents Message-ID: <820@puck.mrcu> Date: 1 Feb 91 11:40:08 GMT Sender: paj@mrcu Reply-To: paj@uk.co.gec-mrc (Paul Johnson) Organization: GEC-Marconi Research Centre, Great Baddow, Essex Lines: 36 I am of the opinion that any non-recursive algorithm can be patented. Here is what you do (note: I have no legal or patent expertise whatsoever). First, write down your algorithm in C or Pascal without using recursion. Second, translate the code into a diagram with every variable represented by a register and every operation represented by a piece of hardware (e.g. adders, multipliers). Write text describing each stage of the computation. You now have a description of a machine which executes your algorithm. Apply for a patent. If someone implements your algorithm in software, you can point to memory locations where variables are stored and alledge that these correspond to the registers, and state that the ALU is doing the work of the adders and multipliers. Of course this is a lot of work, but a program to produce such a set of diagrams and descriptions could be written. It might even be possible as a back end to gcc. Does anyone think that this might be possible? The possibility of producing such diagrams rests on Turing's work on his machines. He proved that any turing machine can be emulated by a universal turing machine. I assume the converse holds, so that any UTM program can be emulated by a specific non-UTM (anyone actually know?) The legal side might require some pushing, particularly when patent applications start being measured by the ton :-) Paul. Paul Johnson UUCP: !mcvax!ukc!gec-mrc!paj --------------------------------!-------------------------|------------------- GEC-Marconi Research is not | Telex: 995016 GECRES G | Tel: +44 245 73331 responsible for my opinions. | Inet: paj@uk.co.gec-mrc | Fax: +44 245 75244