Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > alt.folklore.computers > #211682 > unrolled thread

What ASCII Should Have Been

Started byQuadibloc <jsavard@ecn.ab.ca>
First post2020-06-01 19:25 -0700
Last post2020-07-05 16:35 -0700
Articles 20 on this page of 197 — 32 participants

Back to article view | Back to alt.folklore.computers


Contents

  What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-01 19:25 -0700
    Re: What ASCII Should Have Been Andy Burns <usenet@andyburns.uk> - 2020-06-02 07:49 +0100
      Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 00:26 -0700
        Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 00:35 -0700
    Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-02 08:18 +0000
      Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 02:45 -0700
        Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 02:57 -0700
        Re: What ASCII Should Have Been Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-06-02 10:06 +0000
        Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-02 11:52 +0000
        Re: What ASCII Should Have Been songbird <songbird@anthive.com> - 2020-06-02 12:18 -0400
    Re: What ASCII Should Have Been Theo <theom+news@chiark.greenend.org.uk> - 2020-06-02 11:31 +0100
      Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 05:05 -0700
        Re: What ASCII Should Have Been daphne@panix.com (Daphne Eftychia Arthur) - 2020-06-03 18:09 +0000
          Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 14:01 -0700
          Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 17:54 -0700
            Re: What ASCII actually is John Levine <johnl@taugh.com> - 2020-06-04 02:11 +0000
            Re: What ASCII Should Have Been daphne@panix.com (Daphne Eftychia Arthur) - 2020-06-04 03:05 +0000
              Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 20:26 -0700
                Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-04 07:06 -0700
              Re: What ASCII Should Have Been daphne@panix.com (Daphne Eftychia Arthur) - 2020-06-04 03:27 +0000
                Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 20:31 -0700
                  Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 20:41 -0700
          Re: What ASCII Should Have Been Johann 'Myrkraverk' Oskarsson <johann@myrkraverk.invalid> - 2020-06-06 02:22 +0800
            Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-05 11:33 -0700
              Re: What ASCII Should Have Been Johann 'Myrkraverk' Oskarsson <johann@myrkraverk.invalid> - 2020-06-06 13:58 +0800
                Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-06 11:10 -0700
    Re: What ASCII Should Have Been robin.vowels@gmail.com - 2020-06-02 19:46 -0700
      Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 23:24 -0700
        Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 23:36 -0700
        Re: What ASCII Should Have Been Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-06-03 17:22 +0000
          Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 10:33 -0700
            Re: What ASCII Should Have Been Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-06-03 20:30 +0000
              Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 13:40 -0700
                Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-04 07:06 -0700
                  Re: What ASCII Should Have Been Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-06-04 16:59 +0000
              Re: What ASCII Should Have Been J. Clarke <jclarke.873638@gmail.com> - 2020-06-03 19:50 -0400
                Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 17:47 -0700
                Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-04 07:06 -0700
              Re: What ASCII Should Have Been Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-06-03 20:53 -0300
                Re: What ASCII Should Have Been John Levine <johnl@taugh.com> - 2020-06-04 00:08 +0000
                Re: What ASCII Should Have Been Gerard Schildberger <gerard46@rrt.net> - 2020-06-03 17:10 -0700
                Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 17:48 -0700
                  Re: What ASCII Should Have Been Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-06-04 03:11 -0300
                    Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-04 01:18 -0700
                      Re: What ASCII Should Have Been Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-06-05 03:50 -0300
              Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-04 07:06 -0700
                Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-05 00:44 -0700
                  Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-05 07:53 -0700
                    Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-05 08:04 -0700
                      Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-05 10:03 -0700
                    Re: What ASCII Should Have Been JimP <solosam90@gmail.com> - 2020-06-05 11:05 -0500
                      Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-05 09:27 -0700
                        Re: What ASCII Should Have Been Dallas <dallas@texas.usa> - 2020-06-05 11:45 -0500
                          Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-05 09:55 -0700
                            Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-05 10:16 -0700
                            Re: What ASCII Should Have Been Ahem A Rivet's Shot <steveo@eircom.net> - 2020-06-05 18:23 +0100
                          Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-05 10:03 -0700
                            Re: What ASCII Should Have Been Niklas Karlsson <anksil@yahoo.se> - 2020-06-07 15:05 +0000
                              Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-07 09:28 -0700
                                Re: What ASCII Should Have Been Niklas Karlsson <anksil@yahoo.se> - 2020-06-07 19:56 +0000
                                  Re: What ASCII Should Have Been Dallas <dallas@texas.usa> - 2020-06-07 16:15 -0500
                          Re: What ASCII Should Have Been Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-06-05 17:31 +0000
                            Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-05 11:24 -0700
                          Re: What ASCII Should Have Been robin.vowels@gmail.com - 2020-06-05 15:02 -0700
                            Re: What ASCII Should Have Been Dallas <dallas@texas.usa> - 2020-06-05 17:40 -0500
                          Re: What ASCII Should Have Been John Levine <johnl@taugh.com> - 2020-06-06 00:10 +0000
                            Re: What ASCII Should Have Been Dallas <dallas@texas.usa> - 2020-06-05 19:44 -0500
                              Re: What ASCII Should Have Been John Levine <johnl@taugh.com> - 2020-06-06 01:54 +0000
                        Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-05 10:03 -0700
                          Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-05 11:57 -0700
                        Re: What ASCII Should Have Been JimP <solosam90@gmail.com> - 2020-06-06 11:15 -0500
                          Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-06 11:10 -0700
                            Re: What ASCII Should Have Been JimP <solosam90@gmail.com> - 2020-06-06 16:14 -0500
                              Brest U-boat pens (was: Re: What ASCII Should Have Been "Kerr-Mudd,John" <notsaying@invalid.org> - 2020-06-07 10:01 +0000
                                Re: Brest U-boat pens (was: Re: What ASCII Should Have Been "Kerr-Mudd,John" <notsaying@invalid.org> - 2020-06-07 17:58 +0000
                                  Re: Brest U-boat pens Dallas <dallas@texas.usa> - 2020-06-07 15:54 -0500
                          Re: What ASCII Should Have Been Joy Beeson <jbeeson@invalid.net.invalid> - 2020-06-06 19:12 -0400
                            Re: What ASCII Should Have Been J. Clarke <jclarke.873638@gmail.com> - 2020-06-06 19:40 -0400
                              Re: What ASCII Should Have Been John Levine <johnl@taugh.com> - 2020-06-07 01:22 +0000
                                Re: What ASCII Should Have Been Dallas <dallas@texas.usa> - 2020-06-06 22:22 -0500
                                Re: What ASCII Should Have Been Andy Walker <anw@cuboid.co.uk> - 2020-06-07 10:46 +0100
                                Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-07 07:58 -0700
                                Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-07 15:37 +0000
                              Re: What ASCII Should Have Been Gareth Evans <headstone255@yahoo.com> - 2020-06-07 21:52 +0100
                            Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-06 19:44 -0700
                              Re: What ASCII Should Have Been Alfred Falk <aefalk@telus.net> - 2020-06-07 03:11 +0000
      Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 23:25 -0700
      Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-02 23:27 -0700
      Re: What ASCII Should Have Been Andy Walker <anw@cuboid.co.uk> - 2020-06-03 10:40 +0100
        Re: What ASCII Should Have Been Andy Walker <anw@cuboid.co.uk> - 2020-06-03 12:05 +0100
        Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-03 13:08 +0000
          Re: What ASCII Should Have Been Andy Walker <anw@cuboid.co.uk> - 2020-06-03 15:47 +0100
            Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-03 17:20 +0000
              Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-03 17:26 +0000
                Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 10:35 -0700
                  Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-03 18:03 +0000
              Re: What ASCII Should Have Been Andy Walker <anw@cuboid.co.uk> - 2020-06-03 19:27 +0100
                Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-03 18:38 +0000
                  Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-03 13:41 -0700
                    Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-03 21:15 +0000
                  Re: What ASCII Should Have Been Andy Walker <anw@cuboid.co.uk> - 2020-06-04 00:25 +0100
                    Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-04 07:06 -0700
            Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-04 07:06 -0700
              Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-05 18:27 +0000
                Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-05 12:00 -0700
                  Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-07 16:52 +0000
                    Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-07 10:21 -0700
                    Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-07 11:03 -0700
                      Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-07 19:08 +0000
                        Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-07 13:00 -0700
                        Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-07 16:09 -0700
                          Re: What ASCII Should Have Been usenet@only.tnx (Questor) - 2020-06-09 09:18 +0000
                            Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-09 07:22 -0700
                              Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-09 11:51 -0700
                              Re: What ASCII Should Have Been Rich Alderson <news@alderson.users.panix.com> - 2020-06-09 15:40 -0400
                                Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-09 16:25 -0700
                                  Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-10 06:52 +0000
                                    Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-10 10:22 +0000
                                    Re: What ASCII Should Have Been Rich Alderson <news@alderson.users.panix.com> - 2020-06-10 14:26 -0400
                                      Re: What ASCII Should Have Been Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-06-10 20:54 +0000
                                    Re: What the DEC-20 was, not ASCII Should Have Been John Levine <johnl@taugh.com> - 2020-06-10 18:30 +0000
                                      Re: What the DEC-20 was, not ASCII Should Have Been Rich Alderson <news@alderson.users.panix.com> - 2020-06-11 19:01 -0400
                                    Re: What ASCII Should Have Been Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-06-10 19:35 +0000
                                      Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-11 12:59 +0000
                                        Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-11 07:59 -0700
                                          Re: What ASCII Should Have Been Thomas Koenig <tkoenig@netcologne.de> - 2020-06-11 17:31 +0000
                                            Re: BATTERY RADIOS & STIRLING MOTORS robin.vowels@gmail.com - 2020-06-11 21:29 -0700
                                              Re: BATTERY RADIOS & STIRLING MOTORS Thomas Koenig <tkoenig@netcologne.de> - 2020-06-12 07:04 +0000
                                              Re: BATTERY RADIOS & STIRLING MOTORS Bob Eager <news0073@eager.cx> - 2020-06-12 09:06 +0000
                                                Re: BATTERY RADIOS & STIRLING MOTORS robin.vowels@gmail.com - 2020-06-12 05:32 -0700
                                                Re: BATTERY RADIOS & STIRLING MOTORS Gareth Evans <headstone255@yahoo.com> - 2020-06-12 14:11 +0100
                                                  Re: BATTERY RADIOS & STIRLING MOTORS => Alfred Morgan ted@loft.tnolan.com (Ted Nolan <tednolan>) - 2020-06-12 13:41 +0000
                                                    Re: BATTERY RADIOS & STIRLING MOTORS => Alfred Morgan Bob Eager <news0073@eager.cx> - 2020-06-12 14:33 +0000
                                                      Re: BATTERY RADIOS & STIRLING MOTORS => Alfred Morgan Gareth Evans <headstone255@yahoo.com> - 2020-06-12 16:14 +0100
                                                    Re: BATTERY RADIOS & STIRLING MOTORS => Alfred Morgan Gareth Evans <headstone255@yahoo.com> - 2020-06-12 16:07 +0100
                                                      Re: BATTERY RADIOS & STIRLING MOTORS => Alfred Morgan ted@loft.tnolan.com (Ted Nolan <tednolan>) - 2020-06-12 16:59 +0000
                                                      Re: BATTERY RADIOS & STIRLING MOTORS => Alfred Morgan Peter Flass <peter_flass@yahoo.com> - 2020-06-12 10:31 -0700
                                                        Re: BATTERY RADIOS & STIRLING MOTORS => Alfred Morgan Ahem A Rivet's Shot <steveo@eircom.net> - 2020-06-12 20:33 +0100
                                                      Fun with X-raays (thread drift from  BATTERY RADIOS & STIRLING...) Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-06-12 17:09 -0300
                                                        Re: Fun with X-raays (thread drift from  BATTERY RADIOS & STIRLING...) Quadibloc <jsavard@ecn.ab.ca> - 2020-06-12 14:12 -0700
                                                          Re: Fun with X-raays (thread drift from  BATTERY RADIOS & STIRLING...) J. Clarke <jclarke.873638@gmail.com> - 2020-06-12 19:54 -0400
                                                            Re: Fun with X-raays (thread drift from  BATTERY RADIOS & STIRLING...) J. Clarke <jclarke.873638@gmail.com> - 2020-06-12 19:54 -0400
                                                          Re: Fun with X-raays (thread drift from  BATTERY RADIOS & STIRLING...) Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-06-13 02:57 -0300
                                                            Re: Fun with X-raays (thread drift from  BATTERY RADIOS & STIRLING...) J. Clarke <jclarke.873638@gmail.com> - 2020-06-13 02:15 -0400
                                                        Re: Fun with X-raays (thread drift from BATTERY RADIOS & STIRLING...) Gareth Evans <headstone255@yahoo.com> - 2020-06-14 13:54 +0100
                                                          Re: Fun with X-raays (thread drift from BATTERY RADIOS & STIRLING...) Peter Flass <peter_flass@yahoo.com> - 2020-06-14 15:30 -0700
                                                            Re: Fun with X-raays (thread drift from BATTERY RADIOS & STIRLING...) Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-06-14 22:57 -0300
                                                            Re: Fun with X-raays (thread drift from BATTERY RADIOS & STIRLING...) JimP <solosam90@gmail.com> - 2020-06-14 21:00 -0500
                                                              Re: Fun with X-rays robin.vowels@gmail.com - 2020-06-14 21:12 -0700
                                                                Re: Fun with X-rays JimP <solosam90@gmail.com> - 2020-06-15 10:05 -0500
                                                                  Re: Fun with X-rays Peter Flass <peter_flass@yahoo.com> - 2020-06-15 10:05 -0700
                                                                    Re: Fun with X-rays Dallas <dallas@texas.usa> - 2020-06-15 13:26 -0500
                                                                      Re: Fun with X-rays John Levine <johnl@taugh.com> - 2020-06-15 18:46 +0000
                                                                        Re: Fun with X-rays J. Clarke <jclarke.873638@gmail.com> - 2020-06-15 19:30 -0400
                                                                          Re: Fun with X-rays Dallas <dallas@texas.usa> - 2020-06-15 20:36 -0500
                                                                            Re: Fun with X-rays J. Clarke <jclarke.873638@gmail.com> - 2020-06-15 22:27 -0400
                                                                            Re: Fun with X-rays Bob Eager <news0073@eager.cx> - 2020-06-16 06:40 +0000
                                                                Re: Radium girls, not Fun with X-rays John Levine <johnl@taugh.com> - 2020-06-15 17:29 +0000
                                                  Re: BATTERY RADIOS & STIRLING MOTORS Peter Flass <peter_flass@yahoo.com> - 2020-06-12 07:49 -0700
                                                  Re: BATTERY RADIOS & STIRLING MOTORS robin.vowels@gmail.com - 2020-06-12 08:50 -0700
                                            Re: What ASCII Should Have Been Bob Martin <bob.martin@excite.com> - 2020-06-12 06:07 +0000
                                        Re: What ASCII Should Have Been Niklas Karlsson <anksil@yahoo.se> - 2020-06-14 14:50 +0000
                                          Re: What ASCII Should Have Been "Kerr-Mudd,John" <notsaying@invalid.org> - 2020-06-14 15:01 +0000
                                          Re: What ASCII Should Have Been J. Clarke <jclarke.873638@gmail.com> - 2020-06-14 11:36 -0400
                                            Re: What ASCII Should Have Been Niklas Karlsson <anksil@yahoo.se> - 2020-06-14 18:03 +0000
                              Re: What ASCII Should Have Been usenet@only.tnx (Questor) - 2020-06-18 18:48 +0000
                            Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-09 07:26 -0700
                              Re: What ASCII Should Have Been Dan Espen <dan1espen@gmail.com> - 2020-06-09 12:02 -0400
                                Re: What ASCII Should Have Been Dallas <dallas@texas.usa> - 2020-06-09 11:08 -0500
                                  Subject line evolution Ahem A Rivet's Shot <steveo@eircom.net> - 2020-06-09 17:33 +0100
                                  Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-09 11:51 -0700
                                  Re: What ASCII Should Have Been usenet@only.tnx (Questor) - 2020-06-18 18:48 +0000
                                Re: fast emulators, not What ASCII Should Have Been John Levine <johnl@taugh.com> - 2020-06-09 17:43 +0000
                              Re: What ASCII Should Have Been usenet@only.tnx (Questor) - 2020-06-18 18:48 +0000
                            Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-09 07:32 -0700
                              Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-09 07:47 -0700
                                Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-09 11:51 -0700
                                Re: What ASCII Should Have Been usenet@only.tnx (Questor) - 2020-06-18 18:49 +0000
                              Re: What ASCII Should Have Been usenet@only.tnx (Questor) - 2020-06-18 18:50 +0000
                            Re: What ASCII Should Have Been John Levine <johnl@taugh.com> - 2020-06-09 15:24 +0000
                          Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-09 11:51 -0700
                          Re: What ASCII Should Have Been pechter@lakewoodmicro.com () - 2020-06-16 20:55 +0000
                Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-05 21:27 +0000
                  Re: What ASCII Should Have Been Dallas <dallas@texas.usa> - 2020-06-05 16:54 -0500
                  Re: What ASCII Should Have Been scott@slp53.sl.home (Scott Lurndal) - 2020-06-05 22:44 +0000
                    Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-05 23:18 +0000
          Re: What ASCII Should Have Been robin.vowels@gmail.com - 2020-06-03 16:37 -0700
            Re: What ASCII Should Have Been Bob Eager <news0073@eager.cx> - 2020-06-04 05:23 +0000
        Re: What ASCII Should Have Been robin.vowels@gmail.com - 2020-06-03 16:35 -0700
          Re: What ASCII Should Have Been Andy Walker <anw@cuboid.co.uk> - 2020-06-04 11:31 +0100
            Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-04 07:05 -0700
      Re: What ASCII Should Have Been Peter Flass <peter_flass@yahoo.com> - 2020-06-04 07:06 -0700
        Re: What ASCII Should Have Been scott@slp53.sl.home (Scott Lurndal) - 2020-06-04 15:23 +0000
          Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-06-04 09:38 -0700
        Re: What ASCII Should Have Been Ahem A Rivet's Shot <steveo@eircom.net> - 2020-06-04 17:31 +0100
          Re: What ASCII Should Have Been Andy Burns <usenet@andyburns.uk> - 2020-06-09 10:25 +0100
    Re: What ASCII Should Have Been Quadibloc <jsavard@ecn.ab.ca> - 2020-07-05 16:35 -0700

Page 6 of 10 — ← Prev page 1 … 4 5 [6] 7 8 … 10  Next page →


#211724

FromAndy Walker <anw@cuboid.co.uk>
Date2020-06-04 00:25 +0100
Message-ID<rb9bh0$1mm8$1@gioia.aioe.org>
In reply to#211717
On 03/06/2020 19:38, Bob Eager wrote:
[I wrote:]
>> 	Yes, but when someone says "Algol", do you automatically think
>> of A60 or A68, and if the former, then why?
> Because it was used far more, and implemented on more systems, than ALGOL
> 60 ever was. In my experience, anyway.

	Ah.  Well, my experience corresponds to what you wrote rather than
what you intended to write!  A60 was in substantial use over roughly a
decade, as was A68;  but A68's decade was later, when there were far more
computers around, and when there were far more undergraduate CS courses.
There were [beyond reasonable doubt] more independent A60 implementations,
but many of those were for computers produced in penny packets rather than
the mass production of later years.

-- 
Andy Walker,
Nottingham.

[toc] | [prev] | [next] | [standalone]


#211749

FromPeter Flass <peter_flass@yahoo.com>
Date2020-06-04 07:06 -0700
Message-ID<892833355.612971796.884580.peter_flass-yahoo.com@news.eternal-september.org>
In reply to#211724
Andy Walker <anw@cuboid.co.uk> wrote:
> On 03/06/2020 19:38, Bob Eager wrote:
> [I wrote:]
>>> Yes, but when someone says "Algol", do you automatically think
>>> of A60 or A68, and if the former, then why?
>> Because it was used far more, and implemented on more systems, than ALGOL
>> 60 ever was. In my experience, anyway.
> 
> 	Ah.  Well, my experience corresponds to what you wrote rather than
> what you intended to write!  A60 was in substantial use over roughly a
> decade, as was A68;  but A68's decade was later, when there were far more
> computers around, and when there were far more undergraduate CS courses.
> There were [beyond reasonable doubt] more independent A60 implementations,
> but many of those were for computers produced in penny packets rather than
> the mass production of later years.
> 

In my (limited) experience, I saw ALGOL 60 a few times, and Algol 68 never.
I think it was more widely used in Rightpondia?

-- 
Pete

[toc] | [prev] | [next] | [standalone]


#211746

FromPeter Flass <peter_flass@yahoo.com>
Date2020-06-04 07:06 -0700
Message-ID<1307251810.612971179.385427.peter_flass-yahoo.com@news.eternal-september.org>
In reply to#211706
Andy Walker <anw@cuboid.co.uk> wrote:
> On 03/06/2020 14:08, Bob Eager wrote:
>> But that is specifically ALGOL 68 - a very very different language.
> 
> 	???  Different from Algol 60, certainly;  different from Algol?
> Do you feel the need to say of modern C that it is very very different
> from C, meaning that modern C is not the same as K&R C?  Or that Fortran
> is very very different from Fortran when referring to modern features?
> 

Certainly FORTRAN seems to be a different language, from my perspective of
not knowing very much about it. OTOH perhaps the distinction should be
whether the current compiler can still compile old programs unchanged.

-- 
Pete

[toc] | [prev] | [next] | [standalone]


#211786

FromThomas Koenig <tkoenig@netcologne.de>
Date2020-06-05 18:27 +0000
Message-ID<rbe2qt$2a0$1@newsreader4.netcologne.de>
In reply to#211746
Peter Flass <peter_flass@yahoo.com> schrieb:

> Certainly FORTRAN seems to be a different language, from my perspective of
> not knowing very much about it. OTOH perhaps the distinction should be
> whether the current compiler can still compile old programs unchanged.

There were a few dodgy features that were indeed removed from the
Fortran standards, but most compilers retain them anyway.  But
many people wrote illegal code in the day, and that can be aproblem.

Even today, you can catch flak for pointing out these errors to people.
I'm there, doing that.

[toc] | [prev] | [next] | [standalone]


#211791

FromQuadibloc <jsavard@ecn.ab.ca>
Date2020-06-05 12:00 -0700
Message-ID<659e00ea-d427-42aa-bf0a-160561797c2co@googlegroups.com>
In reply to#211786
On Friday, June 5, 2020 at 12:27:42 PM UTC-6, Thomas Koenig wrote:

> There were a few dodgy features that were indeed removed from the
> Fortran standards, but most compilers retain them anyway.  But
> many people wrote illegal code in the day, and that can be aproblem.

> Even today, you can catch flak for pointing out these errors to people.
> I'm there, doing that.

I'm sorry, but usually what *counts* is getting existing programs to work with 
the least amount of trouble. So anything less than strict upwards compatibility, 
even with the quirks and undocumented features of earlier compilers, is a major 
problem.

Often, the money just isn't there to re-write an old program to bring it up to 
current standards.

John Savard

[toc] | [prev] | [next] | [standalone]


#211833

FromThomas Koenig <tkoenig@netcologne.de>
Date2020-06-07 16:52 +0000
Message-ID<rbj60a$2me$1@newsreader4.netcologne.de>
In reply to#211791
Quadibloc <jsavard@ecn.ab.ca> schrieb:
> On Friday, June 5, 2020 at 12:27:42 PM UTC-6, Thomas Koenig wrote:
>
>> There were a few dodgy features that were indeed removed from the
>> Fortran standards, but most compilers retain them anyway.  But
>> many people wrote illegal code in the day, and that can be aproblem.
>
>> Even today, you can catch flak for pointing out these errors to people.
>> I'm there, doing that.
>
> I'm sorry, but usually what *counts* is getting existing programs to work with 
> the least amount of trouble. So anything less than strict upwards compatibility, 
> even with the quirks and undocumented features of earlier compilers, is a major 
> problem.

People could just stick with old compilers then.

However, they want the performance of new architectures,
vectorization, aggressive optimization, GPU offloading, parallel
processing, allocatable variables, GUIs, and everything else.
However, they want that particular aggressive optimization only
very specifically, where it doesn't happen to break their own
code. Then, it's evil (of course).

> Often, the money just isn't there to re-write an old program to bring it up to 
> current standards.

Problem is, the particular point in question has been illegal since
subroutines and functions were introduced to Fortran, i.e. Fortran II,
the second ever Fortran compiler.

Oh well...

[toc] | [prev] | [next] | [standalone]


#211834

FromQuadibloc <jsavard@ecn.ab.ca>
Date2020-06-07 10:21 -0700
Message-ID<92740b2d-4b14-4e5e-9b29-ae07dd907824o@googlegroups.com>
In reply to#211833
On Sunday, June 7, 2020 at 10:52:27 AM UTC-6, Thomas Koenig wrote:

> However, they want the performance of new architectures,
> vectorization, aggressive optimization, GPU offloading, parallel
> processing, allocatable variables, GUIs, and everything else.
> However, they want that particular aggressive optimization only
> very specifically, where it doesn't happen to break their own
> code. Then, it's evil (of course).

And, what, the people who write the compilers aren't smart enough to write 
perfectly compatible compilers that take advantage of all these performance 
increases?

Of course people are unreasonable and they want to have their cake and eat it 
too. Is this surprising?

John Savard

[toc] | [prev] | [next] | [standalone]


#211836

FromPeter Flass <peter_flass@yahoo.com>
Date2020-06-07 11:03 -0700
Message-ID<438602514.613245430.120860.peter_flass-yahoo.com@news.eternal-september.org>
In reply to#211833
Thomas Koenig <tkoenig@netcologne.de> wrote:
> Quadibloc <jsavard@ecn.ab.ca> schrieb:
>> On Friday, June 5, 2020 at 12:27:42 PM UTC-6, Thomas Koenig wrote:
>> 
>>> There were a few dodgy features that were indeed removed from the
>>> Fortran standards, but most compilers retain them anyway.  But
>>> many people wrote illegal code in the day, and that can be aproblem.
>> 
>>> Even today, you can catch flak for pointing out these errors to people.
>>> I'm there, doing that.
>> 
>> I'm sorry, but usually what *counts* is getting existing programs to work with 
>> the least amount of trouble. So anything less than strict upwards compatibility, 
>> even with the quirks and undocumented features of earlier compilers, is a major 
>> problem.
> 
> People could just stick with old compilers then.

Well, no. Upward compatibility doesn’t imply that new programs have to be
written to the lowest common denominator, only that you shouldn’t have to
rewrite programs to move to a new compiler.

> 
> However, they want the performance of new architectures,
> vectorization, aggressive optimization, GPU offloading, parallel
> processing, allocatable variables, GUIs, and everything else.
> However, they want that particular aggressive optimization only
> very specifically, where it doesn't happen to break their own
> code. Then, it's evil (of course).

Most optimizing compilers offer multiple levels of optimization, including
no optimization, and usually offer ways to turn it on and off for specific
sections of code.

> 
>> Often, the money just isn't there to re-write an old program to bring it up to 
>> current standards.
> 
> Problem is, the particular point in question has been illegal since
> subroutines and functions were introduced to Fortran, i.e. Fortran II,
> the second ever Fortran compiler.
> 

I forgot the “particular point in question”, but often compilers will
support even illegal code from older versions, if it was used widely
enough.


-- 
Pete

[toc] | [prev] | [next] | [standalone]


#211837

FromThomas Koenig <tkoenig@netcologne.de>
Date2020-06-07 19:08 +0000
Message-ID<rbje0b$8ks$1@newsreader4.netcologne.de>
In reply to#211836
Peter Flass <peter_flass@yahoo.com> schrieb:
> Thomas Koenig <tkoenig@netcologne.de> wrote:
>> Quadibloc <jsavard@ecn.ab.ca> schrieb:
>>> On Friday, June 5, 2020 at 12:27:42 PM UTC-6, Thomas Koenig wrote:
>>> 
>>>> There were a few dodgy features that were indeed removed from the
>>>> Fortran standards, but most compilers retain them anyway.  But
>>>> many people wrote illegal code in the day, and that can be aproblem.
>>> 
>>>> Even today, you can catch flak for pointing out these errors to people.
>>>> I'm there, doing that.
>>> 
>>> I'm sorry, but usually what *counts* is getting existing programs to work with 
>>> the least amount of trouble. So anything less than strict upwards compatibility, 
>>> even with the quirks and undocumented features of earlier compilers, is a major 
>>> problem.
>> 
>> People could just stick with old compilers then.

> Well, no. Upward compatibility doesn’t imply that new programs have to be
> written to the lowest common denominator, only that you shouldn’t have to
> rewrite programs to move to a new compiler.

If you used a language construct that was illegal all the time,
sooner or later the compilers will (hopefully) catch up with you,
one way or another.  The nice way is an error message, the
not-so-nice way is silently failing. 

An example:

Let's look at argument size mismatches that "worked" on
little-endian systems like a VAX. Should IBM have changed their
compilers for big-endian MVS systems to match what the VAX did?

Should argument mismatches be caught, in your opinion?  (If you
write C, do you use prototypes?)

> Most optimizing compilers offer multiple levels of optimization, including
> no optimization, and usually offer ways to turn it on and off for specific
> sections of code.

And most people who care about speed use -Ofast (or whatever)
without knowing or caring what it does :-|

[toc] | [prev] | [next] | [standalone]


#211839

FromPeter Flass <peter_flass@yahoo.com>
Date2020-06-07 13:00 -0700
Message-ID<288041464.613252609.452753.peter_flass-yahoo.com@news.eternal-september.org>
In reply to#211837
Thomas Koenig <tkoenig@netcologne.de> wrote:
> Peter Flass <peter_flass@yahoo.com> schrieb:
>> Thomas Koenig <tkoenig@netcologne.de> wrote:
>>> Quadibloc <jsavard@ecn.ab.ca> schrieb:
>>>> On Friday, June 5, 2020 at 12:27:42 PM UTC-6, Thomas Koenig wrote:
>>>> 
>>>>> There were a few dodgy features that were indeed removed from the
>>>>> Fortran standards, but most compilers retain them anyway.  But
>>>>> many people wrote illegal code in the day, and that can be aproblem.
>>>> 
>>>>> Even today, you can catch flak for pointing out these errors to people.
>>>>> I'm there, doing that.
>>>> 
>>>> I'm sorry, but usually what *counts* is getting existing programs to work with 
>>>> the least amount of trouble. So anything less than strict upwards compatibility, 
>>>> even with the quirks and undocumented features of earlier compilers, is a major 
>>>> problem.
>>> 
>>> People could just stick with old compilers then.
> 
>> Well, no. Upward compatibility doesn’t imply that new programs have to be
>> written to the lowest common denominator, only that you shouldn’t have to
>> rewrite programs to move to a new compiler.
> 
> If you used a language construct that was illegal all the time,
> sooner or later the compilers will (hopefully) catch up with you,
> one way or another.  The nice way is an error message, the
> not-so-nice way is silently failing. 
> 
> An example:
> 
> Let's look at argument size mismatches that "worked" on
> little-endian systems like a VAX. Should IBM have changed their
> compilers for big-endian MVS systems to match what the VAX did?

Changing to a different architecture is a whole different problem.

> 
> Should argument mismatches be caught, in your opinion?  (If you
> write C, do you use prototypes?)

I’ve forgotten half my C. A warning message might be appropriate.

> 
>> Most optimizing compilers offer multiple levels of optimization, including
>> no optimization, and usually offer ways to turn it on and off for specific
>> sections of code.
> 
> And most people who care about speed use -Ofast (or whatever)
> without knowing or caring what it does :-|
> 

Rule #1: If it doesn’t work, first turn off optimization.


-- 
Pete

[toc] | [prev] | [next] | [standalone]


#211844

FromQuadibloc <jsavard@ecn.ab.ca>
Date2020-06-07 16:09 -0700
Message-ID<53977137-c62f-40f8-bbdf-49d1bec91dd6o@googlegroups.com>
In reply to#211837
On Sunday, June 7, 2020 at 1:09:00 PM UTC-6, Thomas Koenig wrote:

> Let's look at argument size mismatches that "worked" on
> little-endian systems like a VAX. Should IBM have changed their
> compilers for big-endian MVS systems to match what the VAX did?

I will admit no. For one thing, the 7090 was not a VAX, and it was also big-
endian. Instead, Digital should have made sure that its compilers for the VAX 
would preserve all the quirks of their compilers for the PDP-10, since they 
discontinued the DECsystem-20 in favor of the VAX.

Since the PDP-10 was a 36-bit big-endian system, and the VAX a 32-bit little-
endian system, however, one will have to forgive some limitations. Instead, they 
should simply have done what IBM did with the 360: allow the VAX to run the 
PDP-10 compilers and the code they generate in emulation mode.

John Savard

[toc] | [prev] | [next] | [standalone]


#211846

Fromusenet@only.tnx (Questor)
Date2020-06-09 09:18 +0000
Message-ID<5edf53ba.4570291@news.dslextreme.com>
In reply to#211844
On Sun, 7 Jun 2020 16:09:40 -0700 (PDT), Quadibloc <jsavard@ecn.ab.ca> wrote:
>On Sunday, June 7, 2020 at 1:09:00 PM UTC-6, Thomas Koenig wrote:
>> Let's look at argument size mismatches that "worked" on
>> little-endian systems like a VAX. Should IBM have changed their
>> compilers for big-endian MVS systems to match what the VAX did?
>
>I will admit no. For one thing, the 7090 was not a VAX, and it was also big-
>endian. Instead, Digital should have made sure that its compilers for the VAX 
>would preserve all the quirks of their compilers for the PDP-10, since they 
>discontinued the DECsystem-20 in favor of the VAX.

Uh, no.  They did not "discontinue the DECsystem-20 in favor of the VAX."  The
way you phrased it, it sounds like there was a deliberate, intentional strategy
to move PDP-10 customers (who used TOPS-10 or TOPS-20) to the VAX.  There simply
wasn't.  DEC screwed up so badly in producing a follow-on processor to the KL10
model that they eventually threw in the towel and discontinued the product line
with a sudden end-of-life announcement.  "Migration," as it was known, was an ad
hoc marketing pitch because the company had nothing else to offer but the hope
that enough VAXen and their expected future improvements would make up the
difference.  (I think this was before the VAX 8600 was released.)  After being
told for years that something better was in the works, PDP-10 customers were
understandably furious.  Faced with the necessity of moving to a completely
different system, many ended up with non-DEC solutions.  Others bought VAXen,
but ran some version of Unix instead of VMS in an effort to avoid further vendor
lock-in.


>Since the PDP-10 was a 36-bit big-endian system, and the VAX a 32-bit little-
>endian system, however, one will have to forgive some limitations. Instead, they 
>should simply have done what IBM did with the 360: allow the VAX to run the 
>PDP-10 compilers and the code they generate in emulation mode.

Ha ha ha ha ha ha ha ha.  Uh, no and no.  It sounds like what you are proposing
is running a PDP-10/TOPS-20 emulation on the VAX.  Yes, that was theoretically
possible -- just "a small matter of programming" -- but the performance would
have been terrible.  One might as well write TOPS-20 for the VAX and run it on
the bare metal.

One may say that Fortran is Fortran, but as has been discussed in this forum,
even that isn't true, and then there are many implicit assumptions made about
how files etc. are structured and used which depend on the operating system.
You propose somehow translating the file structure, system calls, et. al. of
TOPS-20 into those for VMS.  Good luck with that.  Again, possible, but not
practical.  It would certainly be a non-trivial programming project, one that
would have taken many months to complete.  There was no plan nor funding for
such a significant effort.

[toc] | [prev] | [next] | [standalone]


#211848

FromQuadibloc <jsavard@ecn.ab.ca>
Date2020-06-09 07:22 -0700
Message-ID<c872bb9b-53bf-4ec2-af02-5aa685997c4fo@googlegroups.com>
In reply to#211846
On Tuesday, June 9, 2020 at 3:17:30 AM UTC-6, Questor wrote:

> Uh, no.  They did not "discontinue the DECsystem-20 in favor of the VAX."  The
> way you phrased it, it sounds like there was a deliberate, intentional strategy
> to move PDP-10 customers (who used TOPS-10 or TOPS-20) to the VAX.  There simply
> wasn't.  DEC screwed up so badly in producing a follow-on processor to the KL10
> model that they eventually threw in the towel and discontinued the product line
> with a sudden end-of-life announcement.

Oh. I did not know that, instead I uncritically accepted the posts here by a 
former DEC employee about the demise of the PDP-10 architecture which was, in 
her eyes, very unfortunate.

John Savard

[toc] | [prev] | [next] | [standalone]


#211858

FromPeter Flass <peter_flass@yahoo.com>
Date2020-06-09 11:51 -0700
Message-ID<2086703136.613421099.831258.peter_flass-yahoo.com@news.eternal-september.org>
In reply to#211848
Quadibloc <jsavard@ecn.ab.ca> wrote:
> On Tuesday, June 9, 2020 at 3:17:30 AM UTC-6, Questor wrote:
> 
>> Uh, no.  They did not "discontinue the DECsystem-20 in favor of the VAX."  The
>> way you phrased it, it sounds like there was a deliberate, intentional strategy
>> to move PDP-10 customers (who used TOPS-10 or TOPS-20) to the VAX.  There simply
>> wasn't.  DEC screwed up so badly in producing a follow-on processor to the KL10
>> model that they eventually threw in the towel and discontinued the product line
>> with a sudden end-of-life announcement.
> 
> Oh. I did not know that, instead I uncritically accepted the posts here by a 
> former DEC employee about the demise of the PDP-10 architecture which was, in 
> her eyes, very unfortunate.
> 
> John Savard
> 

Well, it was, and I wasn’t  even a PDP-10 user at the time. I recall
thinking that it was the beginning of the end, although it took a while.

-- 
Pete

[toc] | [prev] | [next] | [standalone]


#211861

FromRich Alderson <news@alderson.users.panix.com>
Date2020-06-09 15:40 -0400
Message-ID<mddeeqoxc28.fsf@panix5.panix.com>
In reply to#211848
Quadibloc <jsavard@ecn.ab.ca> writes:

> On Tuesday, June 9, 2020 at 3:17:30 AM UTC-6, Questor wrote:

>> Uh, no.  They did not "discontinue the DECsystem-20 in favor of the VAX."
>> The way you phrased it, it sounds like there was a deliberate, intentional
>> strategy to move PDP-10 customers (who used TOPS-10 or TOPS-20) to the VAX.
>> There simply wasn't.  DEC screwed up so badly in producing a follow-on
>> processor to the KL10 model that they eventually threw in the towel and
>> discontinued the product line with a sudden end-of-life announcement.

> Oh. I did not know that, instead I uncritically accepted the posts here by a
> former DEC employee about the demise of the PDP-10 architecture which was, in
> her eyes, very unfortunate.

Then you misunderstood some (many) of Barb's posts.

The Jupiter project (to build the next generation PDP-10, in the form of the
KC-10 CPU and associated devices) was undertaken after "DEC" had begun to
transform into "Digital", with the decision to concentrate on a single system
architecture (VAX) and kill off older lines (the 12- and 18-bit systems to
start, with the 36- and 16-bit systems to follow).

As the red-headed bastard stepchild, it was starved for resources, so that it
could never reach its stated performance goals (which were already far behind
what the market really demanded).  Many of the people who could have
contributed to preventing that failure were moved into other projects, or out
the door.

Her viewpoint is that of an insider who had to watch a great product die.  Mine
is that of a customer who waited with bated breath for the next level machine,
only to be crushed with disappointment at the May 1983 announcement that it was
all going away.

NB:  The customer base for the PDP-10 was large enough, strong enough, that
Digital was forced to agree to continue hardware development (retrofitting the
new peripheral hardware to the KL-10 CPU) for 5 years, to 1988, and software
development/support for 10, to 1993.  An unsuccessful product line would not
have had that kind of leverage.

-- 
Rich Alderson					  news@alderson.users.panix.com
      Audendum est, et veritas investiganda; quam etiamsi non assequamur,
	  omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
									--Galen

[toc] | [prev] | [next] | [standalone]


#211862

FromQuadibloc <jsavard@ecn.ab.ca>
Date2020-06-09 16:25 -0700
Message-ID<842e568c-d489-4baf-9e17-e4c72a788f42o@googlegroups.com>
In reply to#211861
On Tuesday, June 9, 2020 at 1:40:32 PM UTC-6, Rich Alderson wrote:
> Quadibloc <jsavard@ecn.ab.ca> writes:
> 
> > On Tuesday, June 9, 2020 at 3:17:30 AM UTC-6, Questor wrote:
> 
> >> Uh, no.  They did not "discontinue the DECsystem-20 in favor of the VAX."
> >> The way you phrased it, it sounds like there was a deliberate, intentional
> >> strategy to move PDP-10 customers (who used TOPS-10 or TOPS-20) to the VAX.
> >> There simply wasn't.  DEC screwed up so badly in producing a follow-on
> >> processor to the KL10 model that they eventually threw in the towel and
> >> discontinued the product line with a sudden end-of-life announcement.
> 
> > Oh. I did not know that, instead I uncritically accepted the posts here by a
> > former DEC employee about the demise of the PDP-10 architecture which was, in
> > her eyes, very unfortunate.
> 
> Then you misunderstood some (many) of Barb's posts.
> 
> The Jupiter project (to build the next generation PDP-10, in the form of the
> KC-10 CPU and associated devices) was undertaken after "DEC" had begun to
> transform into "Digital", with the decision to concentrate on a single system
> architecture (VAX) and kill off older lines (the 12- and 18-bit systems to
> start, with the 36- and 16-bit systems to follow).
> 
> As the red-headed bastard stepchild, it was starved for resources, so that it
> could never reach its stated performance goals (which were already far behind
> what the market really demanded).  Many of the people who could have
> contributed to preventing that failure were moved into other projects, or out
> the door.
> 
> Her viewpoint is that of an insider who had to watch a great product die.  Mine
> is that of a customer who waited with bated breath for the next level machine,
> only to be crushed with disappointment at the May 1983 announcement that it was
> all going away.
> 
> NB:  The customer base for the PDP-10 was large enough, strong enough, that
> Digital was forced to agree to continue hardware development (retrofitting the
> new peripheral hardware to the KL-10 CPU) for 5 years, to 1988, and software
> development/support for 10, to 1993.  An unsuccessful product line would not
> have had that kind of leverage.

But then I'm misunderstanding your post in the same way.

I see a difference between *failing* in a genuine attempt to produce a new 
computer with the PDP-10 architecture... and not really making the attempt, or 
sabotaging the attempt by starving it of resources. The latter _is_ a genuine 
prioritization of the PDP-11 and its descendants to the exclusion of the PDP-10 
user community, not an _inability_ to design new products for them.

John Savard

[toc] | [prev] | [next] | [standalone]


#211863

FromThomas Koenig <tkoenig@netcologne.de>
Date2020-06-10 06:52 +0000
Message-ID<rbq006$b6b$1@newsreader4.netcologne.de>
In reply to#211862
Quadibloc <jsavard@ecn.ab.ca> schrieb:

> I see a difference between *failing* in a genuine attempt to produce a new 
> computer with the PDP-10 architecture... and not really making the attempt, or 
> sabotaging the attempt by starving it of resources.

There is a middle ground - underestimating the technical difficulty
of producing a new computer, and then facing a hard decision
on how to proceed, because of the economic considerations and
resource limitations.

In my experience, projects never "fail" in the sense that everybody
agrees that they have failed, and even projects that succeed very
often have some people thinking, during the project, that they
have already failed.

On the technical aspects:

While having two 18-bit addresses in a 36-bit word is elegant,
you can only address 1.125 MB of memory with that scheme.
Extending this while maintaining compabibility would be hard.
I think they wanted to extend the address space to 30 bit, but
that would require quite a rewrite of software.

Compared to this, an MVS user program for a 24-bit architecture could
be ported to a 31-bit address space more easily, provided it didn't
fiddle with the upper byte.

[toc] | [prev] | [next] | [standalone]


#211864

FromBob Eager <news0073@eager.cx>
Date2020-06-10 10:22 +0000
Message-ID<hkbqk1Frt4lU4@mid.individual.net>
In reply to#211863
On Wed, 10 Jun 2020 06:52:54 +0000, Thomas Koenig wrote:

> Compared to this, an MVS user program for a 24-bit architecture could be
> ported to a 31-bit address space more easily, provided it didn't fiddle
> with the upper byte.

Indeed. I helped to port an operating system (not an IBM one) from the 24 
bit architecture to XA. Of course, the major part was the channel 
programming.

-- 
Using UNIX since v6 (1975)...

Use the BIG mirror service in the UK:
 http://www.mirrorservice.org

[toc] | [prev] | [next] | [standalone]


#211865

FromRich Alderson <news@alderson.users.panix.com>
Date2020-06-10 14:26 -0400
Message-ID<mdda71ahj4t.fsf@panix5.panix.com>
In reply to#211863
Thomas Koenig <tkoenig@netcologne.de> writes:

> Quadibloc <jsavard@ecn.ab.ca> schrieb:

>> I see a difference between *failing* in a genuine attempt to produce a new
>> computer with the PDP-10 architecture... and not really making the attempt,
>> or sabotaging the attempt by starving it of resources.

> There is a middle ground - underestimating the technical difficulty of
> producing a new computer, and then facing a hard decision on how to proceed,
> because of the economic considerations and resource limitations.

Yes, DEC underestimated the difficulties of expanding the PDP-10 architecture,
but that could have been overcome had there been corporate will to do so.

The "VAX is everything" mindset at Digital after the VAX was defined changed
the priority from "we have to build what our customers want" to "our customers
will want whatever we build"--which in the end killed the company, although
that took another 15 years to take full effect.

> In my experience, projects never "fail" in the sense that everybody agrees
> that they have failed, and even projects that succeed very often have some
> people thinking, during the project, that they have already failed.

> On the technical aspects:

> While having two 18-bit addresses in a 36-bit word is elegant, you can only
> address 1.125 MB of memory with that scheme.

We usually measured things in character equivalency:  Each kiloword on a PDP-10
holds 2560 7-bit ASCII characters, so is equated to 2.5KB on a machine forced
to use 8-bit bytes.  The 256KW memory of the original PDP-10 is equivalent to
1.5MB on a System/360 for byte-level processing.

> Extending this while maintaining compabibility would be hard.  I think they
> wanted to extend the address space to 30 bit, but that would require quite a
> rewrite of software.

The hardware expanded to 22 bits of address on the 2nd generation processor,
introduced in 1971, although user programs were held to 18-bit address spaces.
The physical memory is 4MW on the KI-10, and on the 3rd generation KL-10; the
latter has a microcode implementation of the pager, and allowed 23-bit
addressing.  This was eventually allowed to user programs on both operating
systems (TOPS-20 c. 1981, Tops-10 in 1988).

A third party manufacturer built a full 30-bit addressing clone of the KL-10
which hit the market in late 1995 (missing the upgrade market by roughly a
year).  The programming changes to make use of the full 30 bits of address were
not as drastic as you think--most of the work was done in the early 1970s on
the virtual memory version of Tops-10 for the KI.  SMOP, not a herculean effort.

> Compared to this, an MVS user program for a 24-bit architecture could be
> ported to a 31-bit address space more easily, provided it didn't fiddle with
> the upper byte.

And that was a source of pain for a lot of applications, because they *could*
fiddle with the upper byte, and did.

-- 
Rich Alderson					  news@alderson.users.panix.com
      Audendum est, et veritas investiganda; quam etiamsi non assequamur,
	  omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
									--Galen

[toc] | [prev] | [next] | [standalone]


#211868

FromCharlie Gibbs <cgibbs@kltpzyxm.invalid>
Date2020-06-10 20:54 +0000
Message-ID<rbrhaa21odo@news4.newsguy.com>
In reply to#211865
On 2020-06-10, Rich Alderson <news@alderson.users.panix.com> wrote:

> Thomas Koenig <tkoenig@netcologne.de> writes:
>
>> Compared to this, an MVS user program for a 24-bit architecture could be
>> ported to a 31-bit address space more easily, provided it didn't fiddle with
>> the upper byte.
>
> And that was a source of pain for a lot of applications, because they *could*
> fiddle with the upper byte, and did.

Guilty as charged.  Mea culpa.

-- 
/~\  Charlie Gibbs                  |  Microsoft is a dictatorship.
\ /  <cgibbs@kltpzyxm.invalid>      |  Apple is a cult.
 X   I'm really at ac.dekanfrus     |  Linux is anarchy.
/ \  if you read it the right way.  |  Pick your poison.

[toc] | [prev] | [next] | [standalone]


Page 6 of 10 — ← Prev page 1 … 4 5 [6] 7 8 … 10  Next page →

Back to top | Article view | alt.folklore.computers


csiph-web