Re: More on the “Parallel Sort” Data Sorting Program.

[Austin Obyrne <austin.obyrne@hotmail.com>]

On Tuesday, July 3, 2012 3:43:25 PM UTC+1, Brian Drummond wrote:
> On Tue, 03 Jul 2012 04:22:43 -0700, Austin Obyrne wrote:
> 
> > I ... would like to point out that “Counting “ or “Count sort”
> > was invented in 1954.
> > 
> > I quote from Wikipedia,
> > 
> > “Although radix sorting itself dates back far longer, counting sort, and
> > its application to radix sorting, were both invented by Harold H. Seward
> > in 1954.”  - Unquote
> > 
> > Given the state of progress in computer development at that time this
> > invention must surely have been more of a longhand algorithm than a
> > computer driven program concept.
> 
> What do you mean by this? Certainly not that he couldn't have implemented 
> it and run it on real hardware. Wikipedia places him at MIT where he had 
> access to a computer for three years by 1954.
> 
> - Brian

Yes, I agree with that.

Nothing less than the dynamics of computer computations would have inspired him to invent his sort method that indexed arrays according to magnitude.
“Although radix sorting itself dates back far longer, counting sort, and its application to radix sorting, were both invented by Harold H. Seward in 1954.[1][4][8]    “ 
Looking at this quote again,  I hazard a guess that with this man’s serendipity and extraordinary intelligence he saw the enormous potential of computing a la 1954 albeit he was probably working in assembly language.  The quote herewith suggests that he was interested in sorting data for a long time prior to inventing “Counting Sort”.  He was a part of the space program also and did work for them on instrumentation.  Further guesses by me is that he never implemented “Counting Sort “ as a working program in any of the low level languages when these came along later.
Maybe the archives of MIT may have something  (In passing, he died only recently  – 19th June 2012 - RIP.)

I claim that there is no evidence of any up and running software in a modern language anywhere in the market place in recent times that uses his algorithm.  I am claiming that my implementation in Ada-95 is the first ever.

I will be quite happy to be proved wrong but it is a case of delivery – let me see it advertised or officially recorded in recent times before I will believe it.

Many thanks for your input.

Cheers – Austin.