Hello Squeakers,
I would like run for the first time for the Squeak Oversight Board election in 2024.
I have been reading quietly on the Squeak mailing lists since early 1997.
Not many people know about me so why would they vote for me?
I have - at considerable expense- been building an archive of Squeak and Smalltalk related science papers, files, images, VM’s, source code and videos that has grown to terabytes.
We have seen that preserving image and virtual machine based software like Smalltalk-72, Smalltalk-78, Smalltalk-80 and Squeak can be kept running for 52 years in the future and I see it as my job to keep it running for another 100 years or so. Serving on the Squeak Oversight Board at some time in the future will serve to complete my preservation efforts. If you feel preserving all past and future Squeak software in a running form than voting me on the Squeak Oversight Board is a good way to help my efforts.
A personal history.
At 17 years old I had been blown away by the elegance, simplicity and scientific insights behind Smalltalk-80 and its GUI ever since I read the Byte Magazine articles in august 1981. Message passing and bytecode virtual machines are still the best architectural principles for building computer systems today, especially if you want to build message passing massively parallel computing systems. In 1983 I started my first attempt to re-implement Smalltalk from the Blue and Green Book as the VM, language and operating system for my 800 core Transputer supercomputer hardware. When Squeak was released in 1996/97 I was on this mailing list again trying to port the VM to the parallel hardware in my bedroom that I ran my first public internet provider on. Since 2008 when David Ungar’s RoarVM showed me how I am trying to rewrite the Squeak VM to run on the wafer scale integration million core microprocessors Jecel and I have been working on. We describe this work in a talk
https://vimeo.com/731037615If elected to the Squeak Board I want to organize that Squeak will run bit-identically on more hardware than Linux does. For example compiling the Squeak virtual machines on modern iOS, WatchOS, MacOS for all Squeak versions 1.1 to 6.1 including Croquet, compile for the 100 core AppleSilicon, Risc-V, several FPGA hardware, Android, Windows, etc. I want to port the parallel RoarVM enhancements into the mainstream VM, maybe re-introduce the object table and introduce the use of tags and distributed VM heaps. I can not promise to release all this work while on the Squeak Oversight board but I can try.
Thank you to the folks who encouraged me to run,
Merik Voswinkel