Introduction

This talk is intended to give a sense of history, so new developments in memory systems make more sense. It is my view that there is little new in Computer Science -- there is a lot of reinvention, and much that is claimed to be new is just packaging.

"Why memories?" you may ask. Because they are an increasingly difficult part of the overall system to do right -- and the history of memories is fascinating.

In the rest of the talk, I cover early days before the invention of electronics, followed by the first electronic computer. Still with the pioneers, I look at some of the more bizarre early inventions, many of which first appeared in British designs. Today, John von Neumann is credited with inventing the major design features of modern computers, so it would be remiss to neglect taking a look at the first true "Von Neumann" machine. From there, I go to looking more explicitly at memories, starting with historic generations. I then look at some issues that won't die. Having seen the past, I look at where we are now, and trends which lead to speculation about where things will go next.

To conclude, I offer a list of net resources, which I found helpful in putting the talk together.

Updates November 2002: I have fixed a few broken links (or replaced some where the original material had vanished) to outside sites. I haven't adjusted any numbers, which represent a snapshot of computer costs and speeds in 1996. If you spot any more broken links, please mail me, and I will try to fix them. Unfortunately, the net is a highly mutable medium (even this site had to move, when I moved from South Africa to Australia).

Memories are Made of This 17:30 Thurs 12 September 1996 SHB5, Senate House, University of the Witwatersrand, Johannesburg, South Africa