someone smarter than Dave Cutler
in [Device Drivers]
|
From: Morten Reistad on 1 Apr 2010 07:22 In article <alpine.OSX.2.00.1003310830370.366(a)hsinghsing.panda.com>, Mark Crispin <mrc(a)panda.com> wrote: >On Wed, 31 Mar 2010, Morten Reistad posted: >> Also, remember that the PDP10's and some '11s succeeded on >> the nascent Internet very much despite DEC. They ran non-dec >> software more often than not. Tenex, unix, ITS were popular. > >Yup. > >> 1979 >> saw a transition from Tops10 to Tops20, Should have been a little clearer; this is the view from upper rightopondia. We didn't have many ITS or Tenex installations; tops10 was the workhorse up until ca 1979. Most systems weren't directly visible on the arpanet either, until the tcp/ip transition we had to do login hopping through weeeird interfaces. Mail connectivity was established pretty early though. >More accurately, 1979 saw a transition from Tenex to TOPS-20. > >There were never more than a handful of TOPS-10 systems on the network. >IIRC only one (Rutgers) switched from TOPS-10 to TOPS-20. CMU added a >TOPS-20 system to their farm and it stayed online for a while after the >demise of the TOPS-10 systems. > >Most ARPAnet TOPS-10 systems went offline after the TCP/IP transition on >January 1, 1983. Sometime later, a few reappeared after a crash project >to implement TCP/IP for TOPS-10, but others stayed gone for good. All >TOPS-10 systems vanished by 1990; and the plug was pulled on WAITS and ITS >at about that time. The last tops10 system around here was the Oslo University one, shut down in 1986. The console printer is still there, though, with the last printout still attached. >TOPS-20 has never totally vanished from the Internet, but it's an oddball >today. > >> and unix came in use. > >Yes. Unix was very much an oddball prior to that. > >> Some odd machines like cybers, primes, etc were also on the >> net after the tcp/ip transition. > >I don't recall any Cybers or Primes offering services, although there may >have been one or two oddballs. IIRC, there was a Cyber that you could >access by connecting to a Tenex system on the ARPAnet that had RJE >capability to the Cyber. Again, the view from here. The national data company, ND had a disastrously bad TCP/IP stack, probably contributed a lot to it's demise. >> We saw better performance from a 3-head QNX (with arcnet) 80x86 machine >> than from a VAX 785 with unix. Running usenet, conferencing systems, >> ftp archines, telnet logins etc. This was early 1984. > >I am not the slightest bit surprised. VAX's goose was cooked even as the >PDP-10 was killed, but the PHBs at Digital didn't realize that. > >> Sun's also had a brief career as routers, using VME boards for >> E1 and T1 lines, and their built-in ethernet cards. I still have >> some of those VME boards somewhere. > >Talking about the earliest routers as SUN vs. cisco is rather misleading, >as they have a common ancestor. There's a long long story about this, >which others have already told many times and is too long for this email. I didn't mention Cisco at all. I never saw a cisco until late 1985, until that time it was all unix based machines, mostly suns. By 1988 Cisco had taken over. -- mrr
From: jmfbahciv on 1 Apr 2010 08:35 Morten Reistad wrote: > In article <alpine.OSX.2.00.1003302142210.366(a)hsinghsing.panda.com>, > Mark Crispin <mrc(a)panda.com> wrote: >> On Tue, 30 Mar 2010, Pat Farrell posted: >>> Jim Stewart wrote: >>>> Setting that aside, and it's a big set-aside, I question >>>> how much the PDP-10 was responsible for building the >>>> internet. My understanding is that PDP-11's, Vaxen and >>>> IMP's built the early internet. > > Also, remember that the PDP10's and some '11s succeeded on > the nascent Internet very much despite DEC. They ran non-dec > software more often than not. Tenex, unix, ITS were popular. <snip> Again, you all are forgetting about the network software which became ANF-10. /BAH
From: jmfbahciv on 1 Apr 2010 08:37 Mark Crispin wrote: > On Wed, 31 Mar 2010, jmfbahciv posted: >> PDP-8s were the first CPU when a user would use to get at the PDP-10. >> Think about dial-ups and TTYs which were "far away" from the DC-10s. >> Users didn't see the 8s but those systems were used to answer the >> phones. > > Yes, indeed; particularly on the KA10. IIRC, the most notable of these > PDP-8 based front ends was called X680/I using a PDP-8/i. > > However, this was never how any system was connected to the ARPAnet. I don't remember what the config was for the system that JMF worked on at ORNL. I would suspect that they had an 8/I on their KA back then. > Although I have little doubt that the same hackers who implemented > Kermit on the PDP-8 could figure out how to do an NCP, AFAIK nobody ever > did. An ARPAnet connection also required a special hardware interface > (described in BBN Report 1822, hence an "1822 interface") but once again > AFAIK nobody ever did that for a PDP-8. An 8 would be a tad small for NCP. But I'll bet that the developers accessed the system they worked on through an 8 ;-). /BAH
From: jmfbahciv on 1 Apr 2010 08:39 Charles Richmond wrote: > jmfbahciv wrote: >> Mark Crispin wrote: >>> On Tue, 30 Mar 2010, Pat Farrell posted: >>>> >>>> [snip...] [snip...] [snip...] >>>> >>>> The very first IMPs were specialized, then PDP-8s, and later PDP-11. >>> >>> No. >>> >>> PDP-8s were never used as IMPs; nor to my knowledge was a PDP-8 ever >>> connected to the ARPAnet. >>> >>> PDP-11s were (briefly) used as Internet routers before being replaced >>> by microcomputer based devices (e.g., early cisco routers). They >>> were never IMPs. >>> >> PDP-8s were the first CPU when a user would use to get at the PDP-10. >> Think about dial-ups and TTYs which were "far away" from the DC-10s. >> Users didn't see the 8s but those systems were used to answer the >> phones. >> > > I remember devices called "terminal concentrators" that were attached to > our IBM 370/155 at college. For our DEC-20/50, I do *not* know how the > terminals were supported. Someone posted that the PDP-11 *inside* the > DEC-20 handled terminal I/O for the -20. > > For a 20/50...probably through the 11 which had RSX-20F. /BAH
From: Mark Crispin on 1 Apr 2010 11:40
On Thu, 1 Apr 2010, jmfbahciv posted: >> Although I have little doubt that the same hackers who implemented Kermit >> on the PDP-8 could figure out how to do an NCP, AFAIK nobody ever did. An >> ARPAnet connection also required a special hardware interface (described in >> BBN Report 1822, hence an "1822 interface") but once again AFAIK nobody >> ever did that for a PDP-8. > An 8 would be a tad small for NCP. But I'll bet that the developers > accessed the system they worked on through an 8 ;-). I don't remember any of the ARPAnet-connected KA10s using X680/i or any of other PDP-8 based front ends. By the way, "NCP" in the context of ARPAnet refers to "Network Control Protocol", a predecessor to TCP/IP that was specific to 1822-format networks. NCP was a very simple protocol, with a 40-bit header of which 32-bits was the destination socket number. Transmitting sockets were always odd, receiving sockets were always even; and a socket uniquely identified the connection on the system (there could be only one connection to a socket). The connection protocol (ICP) involved connecting to a well-known socket, reading 32-bits for a new socket to actually use, closing the connection to the well-known socket (so others could use it), then opening a pair of connections to get a bidirectional link. TCP's design was highly influenced by lessons learned from NCP, and especially NCP's complex and fragile ICP. -- Mark -- http://panda.com/mrc Democracy is two wolves and a sheep deciding what to eat for lunch. Liberty is a well-armed sheep contesting the vote. |