[SIGCIS-Members] More records: oldest hardware and software in production use

David Grier grier at email.gwu.edu
Tue Oct 6 08:59:50 PDT 2020


Tom
	You might want to talk with Don Shafer.  donshafer at mac.com <mailto:donshafer at mac.com>  There was a period when he and I were competing for being the world’s oldest second generation computer scientist.  (As I recall, he won on both metrics.  His father joined the computer field in 52 while mine waited until 54).  At one point, Don came across some of his father’s code and realized that it was still being used.  I believe it was for the AEC and that it was one of the early assemblers like Autocode that was being executed by a simulator.  But the details are lost.  He will tell you the story in all its glory.  And will include pictures of Armadillos in the telling.  

David
  

	

> On Oct 6, 2020, at 11:35 AM, <thomas.haigh at gmail.com> <thomas.haigh at gmail.com> wrote:
> 
> Hello SIGCIS,
>  
> Continuing with the Guinness Records theme, I’d like to nominate records for the systems and code that lasted longest if these can be determined with sufficient confidence.
>  
> Electronic computer system in longest continual operation: Voyager 2
>  
> This has to be a space probe, as they get launched into space and aren’t serviced or upgraded for decades. Voyager 1 and 2 were launched in 1977 and is still going, having left the solar system back in 2012. Their computers are still sending back data. Oddly, Voyager 2 was launched 16 days BEFORE Voyager 1, so even though it hasn’t gone as far it has presumably been in operation slightly longer. 
>  
> Pioneer 11, launched in 1973 also had a long run but has not been heard from since 1995. Pioneer 10 sent back data from 1972 to 2003. So the Voyagers have already lasted far longer.
>  
> “Continual” is there to distinguish this from old computers that have been restored to operation in museums. Are there other systems I might have overlooked?
>  
> Computer code with longest lifespan: IRS Individual Master File???
>  
> This would be a harder record, perhaps too hard to pin down with confidence. You might, for example, say “FORTRAN” but none of the original code from FORTRAN 1 is present in modern versions. So what is the oldest snippet of code that actually still runs on a production basis? 
>  
> It clearly has to be in an organization that computerized early and hasn’t been great at updating. That almost certainly puts it in the US government, and there are many reports of ancient code underpinning the US Tax system so the IRS is a likely candidate. One might also suspect the SSA. Code from the early 1960s is claimed to still lurk at the heart of Individual Master File application, in a mix of Cobol and assembly language. https://en.wikipedia.org/wiki/Individual_Master_File <https://en.wikipedia.org/wiki/Individual_Master_File> sources the retention of 1960s code to https://fcw.com/articles/2016/04/08/taxman-tech-troubles.aspx <https://fcw.com/articles/2016/04/08/taxman-tech-troubles.aspx>. The age of this system got publicity when the it crashed on tax day in 2018.
>  
> But is that THE oldest code running in production? And can we get a more specific date on when the IMF entered operation. According to material from the second edition of Ceruzzi’s A History of Modern Computing(mostly retained for the forthcoming New History) the IRS was an early and very large scale adopter of computers. In 1964 it was running IBM 7070s in a new “national center”, and by 1965 the inclusion of an SSN on all tax returns was required. (Obviously SSNs themselves are much older). I’d guess that this shift went along with the introduction of the master file, though if any assembly is still in operation it is surely from the slightly later System/360 rather than the 7070. System/360 machines had replaced the 7070s by 1967.
>  
> Does anyone have information that might let us pin this down more precisely? Or other ideas on the oldest code still in production use?
>  
> Best wishes,
>  
> Tom
>  
>  
> _______________________________________________
> This email is relayed from members at sigcis.org <http://sigcis.org/>, the email discussion list of SHOT SIGCIS. Opinions expressed here are those of the member posting and are not reviewed, edited, or endorsed by SIGCIS. The list archives are at http://lists.sigcis.org/pipermail/members-sigcis.org/ <http://lists.sigcis.org/pipermail/members-sigcis.org/> and you can change your subscription options at http://lists.sigcis.org/listinfo.cgi/members-sigcis.org <http://lists.sigcis.org/listinfo.cgi/members-sigcis.org>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sigcis.org/pipermail/members-sigcis.org/attachments/20201006/b5838ac1/attachment.html>


More information about the Members mailing list