Steve Gardner	John Middlemas
Date: 22 June 96 To: Steve Gardner <webman> From: john@brain.eu.org (John Middlemas) Subject: webneurons Hi Steve, I am interested in your comments on webneurons. >Yes, things like JAVA have happened very fast, but JAVA is realy only a bolt on >which you can take or leave, that runs on top of existing protocols. But can't a good bolt on migrate downwards and take over. What do you think about Microsoft integrating it into Windows? Whatever happens, a kernel is needed to control webneuron operation. That kernel cannot be written using webneurons. For example, since webneurons are multi-tasking, a stack is required and that stack must be managed by the kernel. So how to write the kernel and with what? and how to make it webwide. What specifically do you suggest for this? CGI, C++, Java, Assembler....? >If you were to try and implement your ideas over the existing http you will >end up with a shadow of your true vision. Your changes are realy at a very low >level, and of a fundamental nature. You mean dump HTML? I agree the changes are low level in nature and fundamental. I had been developing them at assembler level before I came across the webneuron idea. You could build an OS from them. A point to bear in mind is that there are countless URL's. In assembler I was going to use a unique 32 bit code for each webneuron but this is insufficient address space to cover the web. How do you think the address space problem is best overcome? How do you think a workable system could be implemented in a reasonable amount of time? I mean, Java has already sorted the problem of platform dependance and this could save a lot of time in fiddling about with one system for Mac, one for Windows, UNIX etc etc. I would love to rewrite Windows itself (you don't need files with webneurons) but this could take ages. >Either that or you compromise your plans >and end up with "JAVA II" :) which would be much easier to integrate into the >Net, but would defeat the objectives that you have setout to achieve. What else to use that might be practical? You would need a large team to do it in assembler for every computer. In balance, what do you suggest, or indeed, do you want to get involved? Look forward to your reply.