Being immersed in computing, from microcode to Unix/linux type OSes and X11 windowing system across 5 different processor architectures (CISC, RISC) I have seen a lot of very cool innovation in the last 30 years. The most famous categories usually lie in the performance domain: CPU speeds, Memory Size and speeds, Storage capacities have all been at the forefront of ECT technology and associated marketing. However, without turning down the immense sum of skills that have been put together in the industry to reach such impressive goals, today’s talk is more about trend changing creations than incremental improvements, how impressive they may look.
Today I would like to open with you my own little computing technology hall of fame, things that have really impressed me instantly, or on the contrary, did not seem poised to take on the whole world but are still around after multiple decades and seem to want to last even longer.
Number 1 Choice in Innovation, Longevity and Scope : The Spreadsheet.
As a young engineer, I remember being frowned upon for repeatedly accessing the accounting department sole PC-AT machine to work on very simple LOTUS 1-2-3 tables. They saved me hours of tedious use of my old pocket calculator. To use what was one of the first spreadsheets, I never had to read a manual and was instantly seduced with this wonderfully innovative interface. At the time, it was the easiest way to use the power of a microprocessor without any programming skills. 30 years later, I run complex product sales tables, interfaced through pivot tables with the main company ERP. This is still the same interface. And even if today’s versions embark very sophisticated features, the first contact for any newbie will look a lot like my 30 years old first experience. So kudos to Visicalc(tm) and Lotus 1-2-3(tm) designers to have opened an era that has survived many generations and will outlive me for sure.
Number 1 Choice for Innovation and Scope: Virtualization.
I will always remember my first attempt at playing with virtualization. Running a linux server within a virtual machine on my WinTel PC allowed me to access again all the nice tools associated with UNIX. Later, I used virtualization to create a clone from my dying laptop to run it in a virtual machine on my brand new personal computer. Gone were the days of careful file transfers and software migration. Wherever and whenever I wanted, I could travel back in time and resuscitate my old PC to access my old files using my legacy applications. Even the desktop background was there. How could a complete, unmodified OS execute on a completely different machine?
Of course, this was not made possible overnight. First, some silicon architectures included an extra hardware protected execution Environment in addition to the privileged mode the OS was using to manage processes. Then more and more sophisticated techniques to emulate devices were also brought into the solution. Lately, specific, hypervisor-friendly pseudo devices where added, in the form of add-ons for the virtualized OS, to bring virtualized performance at the same level of regular, single OS computing.
There seems to be no limit to what virtualization can do. Travelling in time is not the least. In an industry where hardware innovation is also the main weakness for a stable software industry undermined by endless silicon obsolescence, running old binaries on new silicon is a welcome feature. This is nothing new. Sun workstations had to use some of these techniques when moving from 68K to SPARC, as well as the MAC’s when leaving 68K to PowerPC to the current silicon architecture. But there was this pivotal point when the implementation finally reached the ease of use allowing non computer gurus to use it and became mainstream. And the number of new use cases for virtualized enabled computers was large enough to help the creation of a complete new industry: Cloud computing. In embedded computing so far, virtualization has not yet been at the core of the action, but even in tiny machines, virtualization is at the heart of disruptive solutions for data security, safety critical architectures and the solutions for the Internet of Things. So we will see more and more of it. We’d better be prepared.
How about you, what are your best candidates for the computer technology hall of fame?