[ih] Where's Multics now, was Internet-history Digest

[Clem Cole]

Sorry for so many typos. Take a dyslexic guy to begin with and give an
iPhone as his keyboard.  I hope that last message was parsable. Please feel
free to send me off-list questions particularly if my wording / typing was
confusing.

Sent from a handheld expect more typos than usual


On Mon, Aug 18, 2025 at 7:55 AM Clem Cole <clemc at ccc.com> wrote:

> First remember Johnny knows a lot about RSX-11M sources so final state I
> trust his comment, but he was never a DECie or worked in the Mill or ZKO
> like some us so he never lived it and I suggest you take many of comments
> with a bit of care.
>
> What I can state for sure is that he’s right that the Culter lead -11M and
> later StarOS aka VMS (the later when he worked for my former boss the late
> Roger Gourd aka “Fossil” in many DEC histories).
>
> I also know that at the time, DC was said to have made it clear that his
> PDP real system was “better” than the early versions RSX which was one of
> reasons he got the job to create the rewrite - ie “prove it.”  Knowing
> Dave a bit later, I believe those reports.
>
> So how much of -11M came from DEX-15/RSX and how came from DCs system from
> DuPont I can not say.  But I can try to ask a few of my
> friends and coworkers his were there that might know (I was doing Unix
> stuff in those days elsewhere and was not paying attention)
>
> The other point is NT-windows is based on DEC’s which was the new
> microkernel DC was writing at DECwest with a VMS and a Tru64 server that he
> took to Microsoft’s where IBM, NCR and Microsoft started to write an OS/2
> and Presentation manager server. Microsoft was
> Also writing a W95 server.
>
> After the divorce,  it became NT-4 net NT-WIN.   IBM move to there
> original kernel from Whiteplain and released Warp and NT-windows was
> released
>
>
> Again, Multics only came into any influence by the later generations.  But
> that was all indirect and by people that joined these projects that
> Might have studied it in school.  As I said, I could be Mistaken but I’m
> not sure Dave has ever read the book.
>
>
> Sent from a handheld expect more typos than usual
>
>
> On Mon, Aug 18, 2025 at 1:01 AM Jim Carpenter <jim at deitygraveyard.com>
> wrote:
>
>> On Sun, Aug 17, 2025 at 4:11 PM Clem Cole via Internet-history
>> <internet-history at elists.isoc.org> wrote:
>> > Ouch... VMS's parent was RSX, and that's parent was a real-time system
>> Dave
>> > wrote for the PDP-10 when he was at Dupont before DEC hired him.   To
>> > my knowledge, Dave had not read Organick's book when he did those
>> systems
>> > (I'm not sure he has even today).   He was purely a real-time
>> guy/process
>> > control guy - not a multi-user/mult-tasking.
>>
>> According to Johnny Billquist, Dave Cutler did a
>> "redesign/reimplementation" of RSX-11D, which became RSX-11M. That
>> then became the starting point for VMS. RSX started at DEC with Dan
>> Brevik on the PDP-15 as DEX-15/RSX-15, then was brought over to the
>> PDP-11.
>>
>> This is part of a post from Dan Brevik on comp.os.vms dated 7/27/2003
>> (
>> http://groups.google.com/groups?selm=g11Va.162679%24ye4.109589%40sccrnsc01&output=gplain
>> ):
>> > RSX did not come from a DEC operating system, as you observed. It's
>> intellectual
>> > precedants were a realtime executive writen by John Neblett (now
>> retired in Ashville, NC)
>> > for the RW-300 process control computer. Thence to "The Synchronous
>> Executive" by
>> > me about 1963 for the TRW-330 process control computer. Then I headed
>> the project
>> > to write "Ops Control" for the BR-340 in 1964 (Dupont loved it).
>> Bunker-Ramo left the
>> > process control business and I joined Honeywell where I managed the
>> project for OLERT
>> > on the DDP-516 (John Haynie and Walt Duncan, chief architechs and
>> developers). I joined
>> > DEC in 1969 in marketing (all marketeers were engineers at that time)
>> and hired Sam Reese
>> > part time to help me write a real time executive for the PDP-15. (Paid
>> him $15,000). I had
>> > known Sam at Honeywell where he dashed off a small RT exec called
>> Samtran. This gave me the
>> > time to just sit back and think. The basic specifications and
>> architecture came primarily out of me based
>> > on experience.
>>
>> The Ramo-Wooldridge RW-300 was announced in 1957 and available in 1959.
>>
>> Visit
>> https://web.archive.org/web/20050404121912/http://www.demillar.com:80/RSX/
>> and read his outline/notes for more tidbits. And FYI, he does write
>> "MULTICS was not an influence."
>>
>> Jim
>>
>