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

Groups > comp.misc > #9421 > unrolled thread

[CM] Coding with dad on the Dragon 32

Back to article view | Back to comp.misc

Contents

  [CM] Coding with dad on the Dragon 32 RS Wood  <rsw@therandymon.com> - 2015-11-08 13:49 +0000
    Re: [CM] Coding with dad on the Dragon 32 RS Wood <rsw@therandymon.com> - 2015-11-09 09:22 +0300
      Re: [CM] Coding with dad on the Dragon 32 Marko Rauhamaa <marko@pacujo.net> - 2015-11-09 09:22 +0200
        Re: [CM] Coding with dad on the Dragon 32 "gareth" <no.spam@thank.you.invalid> - 2015-11-09 11:10 +0000
          Re: [CM] Coding with dad on the Dragon 32 Ahem A Rivet's Shot <steveo@eircom.net> - 2015-11-09 12:07 +0000
        Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-09 12:08 +0100
          Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-09 14:13 +0000
            Re: [CM] Coding with dad on the Dragon 32 "gareth" <no.spam@thank.you.invalid> - 2015-11-09 14:23 +0000
              Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-10 14:22 +0000
                Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-10 18:18 +0100
                  Re: [CM] Coding with dad on the Dragon 32 Ahem A Rivet's Shot <steveo@eircom.net> - 2015-11-10 18:25 +0000
                    Re: [CM] Coding with dad on the Dragon 32 "gareth" <no.spam@thank.you.invalid> - 2015-11-10 19:35 +0000
                      Re: [CM] Coding with dad on the Dragon 32 "John Jackson" <jj@nospam.com> - 2015-11-11 06:47 +1100
                        Re: [CM] Coding with dad on the Dragon 32 "gareth" <no.spam@thank.you.invalid> - 2015-11-10 21:15 +0000
                          Re: [CM] Coding with dad on the Dragon 32 "John Jackson" <jj@nospam.com> - 2015-11-11 12:15 +1100
                            Re: [CM] Coding with dad on the Dragon 32 "gareth" <no.spam@thank.you.invalid> - 2015-11-11 10:19 +0000
                      Re: [CM] Coding with dad on the Dragon 32 Ahem A Rivet's Shot <steveo@eircom.net> - 2015-11-11 06:23 +0000
                    Re: [CM] Coding with dad on the Dragon 32 Mike Tomlinson <mike@jasper.org.uk> - 2015-11-10 19:51 +0000
                    Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-11 00:31 +0100
                      Re: [CM] Coding with dad on the Dragon 32 Stan Barr <plan.b@bluesomatic.org> - 2015-11-11 08:01 +0000
                        Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-11 12:23 +0100
                          Re: [CM] Coding with dad on the Dragon 32 pechter@pechter.dyndns.org (William Pechter) - 2015-11-12 03:27 +0000
                            Re: [CM] Coding with dad on the Dragon 32 Ahem A Rivet's Shot <steveo@eircom.net> - 2015-11-12 13:23 +0000
                  Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-11 14:39 +0000
                    Re: [CM] Coding with dad on the Dragon 32 "Rod Speed" <rod.speed.aaa@gmail.com> - 2015-11-12 05:51 +1100
                Re: [CM] Coding with dad on the Dragon 32 "Rod Speed" <rod.speed.aaa@gmail.com> - 2015-11-11 06:28 +1100
            Re: [CM] Coding with dad on the Dragon 32 Bob Eager <news0005@eager.cx> - 2015-11-09 16:24 +0000
              Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-10 14:22 +0000
                Re: [CM] Coding with dad on the Dragon 32 scott@slp53.sl.home (Scott Lurndal) - 2015-11-10 16:01 +0000
                  Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-10 18:20 +0100
                    Re: [CM] Coding with dad on the Dragon 32 scott@slp53.sl.home (Scott Lurndal) - 2015-11-10 18:49 +0000
                  Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-11 14:39 +0000
                    Re: [CM] Coding with dad on the Dragon 32 scott@slp53.sl.home (Scott Lurndal) - 2015-11-11 16:31 +0000
                      Re: [CM] Coding with dad on the Dragon 32 Bob Eager <news0005@eager.cx> - 2015-11-11 18:05 +0000
                        Re: [CM] Coding with dad on the Dragon 32 scott@slp53.sl.home (Scott Lurndal) - 2015-11-11 18:20 +0000
                          Re: [CM] Coding with dad on the Dragon 32 Bob Eager <news0005@eager.cx> - 2015-11-11 23:24 +0000
                      Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-12 14:04 +0000
                        Re: [CM] Coding with dad on the Dragon 32 pechter@pechter.dyndns.org (William Pechter) - 2015-11-12 14:47 +0000
                          Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-13 13:58 +0000
                            Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-13 15:44 +0100
                              Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-14 13:56 +0000
                                Re: [CM] Coding with dad on the Dragon 32 "John Jackson" <jj@nospam.com> - 2015-11-15 04:44 +1100
                              Re: [CM] Coding with dad on the Dragon 32 Anne & Lynn Wheeler <lynn@garlic.com> - 2015-11-14 10:15 -0800
                                Re: [CM] Coding with dad on the Dragon 32 Anne & Lynn Wheeler <lynn@garlic.com> - 2015-11-14 10:53 -0800
                              Re: [CM] Coding with dad on the Dragon 32 scott@slp53.sl.home (Scott Lurndal) - 2015-11-16 14:47 +0000
                        Re: [CM] Coding with dad on the Dragon 32 scott@slp53.sl.home (Scott Lurndal) - 2015-11-12 15:26 +0000
                Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-10 17:52 +0100
            Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-09 20:17 +0100
            Re: [CM] Coding with dad on the Dragon 32 hankvc@blackhole.lostwells.org (Hank) - 2015-11-10 02:21 +0000
              Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-10 14:22 +0000
        Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-09 14:13 +0000
          Re: [CM] Coding with dad on the Dragon 32 "Rod Speed" <rod.speed.aaa@gmail.com> - 2015-11-10 05:31 +1100
            Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-10 14:22 +0000
              Re: [CM] Coding with dad on the Dragon 32 Morten Reistad <first@last.name.invalid> - 2015-11-10 18:05 +0100
                Re: [CM] Coding with dad on the Dragon 32 Ahem A Rivet's Shot <steveo@eircom.net> - 2015-11-11 10:52 +0000
                  Re: [CM] Coding with dad on the Dragon 32 Mike Tomlinson <mike@jasper.org.uk> - 2015-11-11 11:40 +0000
                    Re: [CM] Coding with dad on the Dragon 32 Ahem A Rivet's Shot <steveo@eircom.net> - 2015-11-11 14:01 +0000
              Re: [CM] Coding with dad on the Dragon 32 "Rod Speed" <rod.speed.aaa@gmail.com> - 2015-11-11 06:25 +1100
                Re: [CM] Coding with dad on the Dragon 32 Johnny B Good <johnny-b-good@invalid.ntlworld.com> - 2015-11-10 23:19 +0000
                  Re: [CM] Coding with dad on the Dragon 32 "Sam Thatch" <st342@gmail.com> - 2015-11-11 12:23 +1100
                  Re: [CM] Coding with dad on the Dragon 32 Ahem A Rivet's Shot <steveo@eircom.net> - 2015-11-11 07:34 +0000
                    Re: [CM] Coding with dad on the Dragon 32 Johnny B Good <johnny-b-good@invalid.ntlworld.com> - 2015-11-11 17:24 +0000
                      Re: [CM] Coding with dad on the Dragon 32 RS Wood <rsw@therandymon.com> - 2015-11-11 17:33 +0000
                  Re: [CM] Coding with dad on the Dragon 32 Mike Tomlinson <mike@jasper.org.uk> - 2015-11-11 09:51 +0000
              Re: [CM] Coding with dad on the Dragon 32 pechter@pechter.dyndns.org (William Pechter) - 2015-11-12 03:17 +0000
                Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-12 14:04 +0000
                  Re: [CM] Coding with dad on the Dragon 32 Nyssa <Nyssa@flawlesslogic.com> - 2015-11-12 12:24 -0500
                    Re: [CM] Coding with dad on the Dragon 32 Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2015-11-12 17:45 +0000
                    Re: [CM] Coding with dad on the Dragon 32 jmfbahciv <See.above@aol.com> - 2015-11-13 13:58 +0000
                      Re: [CM] Coding with dad on the Dragon 32 Rich <rich@example.invalid> - 2015-11-13 15:51 +0000
                        Re: [CM] Coding with dad on the Dragon 32 Nyssa <Nyssa@flawlesslogic.com> - 2015-11-13 14:51 -0500
                        Re: [CM] Coding with dad on the Dragon 32 pechter@pechter.dyndns.org (William Pechter) - 2015-11-15 16:44 +0000
                      Re: [CM] Coding with dad on the Dragon 32 Andrew Swallow <am.swallow@btinternet.com> - 2015-11-13 18:11 +0000
                  Re: [CM] Coding with dad on the Dragon 32 "Rod Speed" <rod.speed.aaa@gmail.com> - 2015-11-13 12:37 +1100
        Re: [CM] Coding with dad on the Dragon 32 Paul Sture <nospam@sture.ch> - 2015-11-11 18:32 +0100

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

#9526

Morten Reistad wrote:
> In article <PM0005246CA6348F23@aca411e9.ipt.aol.com>,
> jmfbahciv  <See.above@aol.com> wrote:
>>William Pechter wrote:
>>> In article <PM0005245889FBE0A0@aca40f93.ipt.aol.com>,
>>> jmfbahciv  <See.above@aol.com> wrote:
>>>>Scott Lurndal wrote:
>>>>> jmfbahciv <See.above@aol.com> writes:
>>>>>>Scott Lurndal wrote:
>>>>>>> jmfbahciv <See.above@aol.com> writes:
>>>>>
>>>>>>>>Both were shit products.  VMS got better after the PDP-10 developers
>>>>>>>>got hired.
>>>>>>>
>>>>>>> While I agree that DECTape wasn't the brightest idea,
>>>>>>> your hatred of the VAX is personal; it was obviously a
>>>>>>> great product if you gauge by shipments, revenue and
>>>>>>> user testimonials.
>>>>>>
>>>>>>Can you see the difference between VMS and VAX?
>>>>>>
>>>>>
>>>>> Given that one didn't exist without the other (at least
>>>>> until the BSD port), no.
>>>>
>>>>Ah, I see.  You're a customer.  On my side of the biz,
>>>>they were two very different entities.
>>>>
>>>>/BAH
>>>
>>> There were differences between what LCG development was like and what the
>>> rest of DEC was like.  In the real world VAX wasn't VMS although VMS was
>>> VAX.
>>>
>>> As far as your view on VMS -- it was pretty distorted compared with the
>>views
>>> of customers in the field.  The major colleges and universities which used
>>> TOPS-10 and TOPS-20 weren't as pleased since the VAX was more a
departmental
>>> box than a mainframe.  It wasn't designed to support thousands of users.
>>
>>It's not distorted.  10 and 20 customers were told that VAX/VMS was going
>>to replace the PDP-10 business.  No choice in the matter.  That's what I,
>>and a lot of other people, especially including customers, objected to.
>>VAX/VMS was fine for a particular niche, but not a viable replacement for
>>the work that the PDP-10s did.  It was insane for Bell to insist that
happen.
> [snip]
>>Originally, OSF/1 was not supposed to be yet another Unix.  JMF was part
>>of the original design group and got extrememly despressed when he
>>firgured out that the politics had determined that it would be another unix.
>>The original goal was to design an OS which had the strengths of all
>>the OSes and avoid the problems of those same OSes.
>
> The OSF/1 designers nevers saw the possibility to have multiple
> personalites. These have a bad reputation after the "unix" retrofits
> for VMS(Eunice?), Primos (Primix), etc, but if this were designed
> in from the start, like AIX paritally has.
>
> OSF/1 could have been both a multics-like OS and have a unix
> top layer, but with some hypervisor-like functions for the extra
> functionality that is unknown to the unix api and ui.

Just think what the OS could have been with designers who had
Unix, TOPS-10, TOPS-20, VMS and PDP-11 and -8 OS experience.
If the development process had acquired the Multics' methods,
security would not be an afterthought.  OSF/1 blew it; not
even the project leader could change the minds of the
managers who wanted yet another unix.
>
> QNX has done this, with a "just another unix" interface and a
> "real QNX" interface.

The interface could have been pluggable so that moving to an
OSF/1 system did not require redoing the entire computing
of a customer site.


>
> Well, perhaps some OSF/1 designers did.

they did and fought but lost.

/BAH

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

#9528

"jmfbahciv" <See.above@aol.com> wrote in message 
news:PM00052480CDC79C0E@aca429dc.ipt.aol.com...
> Morten Reistad wrote:
>> In article <PM0005246CA6348F23@aca411e9.ipt.aol.com>,
>> jmfbahciv  <See.above@aol.com> wrote:
>>>William Pechter wrote:
>>>> In article <PM0005245889FBE0A0@aca40f93.ipt.aol.com>,
>>>> jmfbahciv  <See.above@aol.com> wrote:
>>>>>Scott Lurndal wrote:
>>>>>> jmfbahciv <See.above@aol.com> writes:
>>>>>>>Scott Lurndal wrote:
>>>>>>>> jmfbahciv <See.above@aol.com> writes:
>>>>>>
>>>>>>>>>Both were shit products.  VMS got better after the PDP-10 
>>>>>>>>>developers
>>>>>>>>>got hired.
>>>>>>>>
>>>>>>>> While I agree that DECTape wasn't the brightest idea,
>>>>>>>> your hatred of the VAX is personal; it was obviously a
>>>>>>>> great product if you gauge by shipments, revenue and
>>>>>>>> user testimonials.
>>>>>>>
>>>>>>>Can you see the difference between VMS and VAX?
>>>>>>>
>>>>>>
>>>>>> Given that one didn't exist without the other (at least
>>>>>> until the BSD port), no.
>>>>>
>>>>>Ah, I see.  You're a customer.  On my side of the biz,
>>>>>they were two very different entities.
>>>>>
>>>>>/BAH
>>>>
>>>> There were differences between what LCG development was like and what 
>>>> the
>>>> rest of DEC was like.  In the real world VAX wasn't VMS although VMS 
>>>> was
>>>> VAX.
>>>>
>>>> As far as your view on VMS -- it was pretty distorted compared with the
>>>views
>>>> of customers in the field.  The major colleges and universities which 
>>>> used
>>>> TOPS-10 and TOPS-20 weren't as pleased since the VAX was more a
> departmental
>>>> box than a mainframe.  It wasn't designed to support thousands of 
>>>> users.
>>>
>>>It's not distorted.  10 and 20 customers were told that VAX/VMS was going
>>>to replace the PDP-10 business.  No choice in the matter.  That's what I,
>>>and a lot of other people, especially including customers, objected to.
>>>VAX/VMS was fine for a particular niche, but not a viable replacement for
>>>the work that the PDP-10s did.  It was insane for Bell to insist that
> happen.
>> [snip]
>>>Originally, OSF/1 was not supposed to be yet another Unix.  JMF was part
>>>of the original design group and got extrememly despressed when he
>>>firgured out that the politics had determined that it would be another 
>>>unix.
>>>The original goal was to design an OS which had the strengths of all
>>>the OSes and avoid the problems of those same OSes.
>>
>> The OSF/1 designers nevers saw the possibility to have multiple
>> personalites. These have a bad reputation after the "unix" retrofits
>> for VMS(Eunice?), Primos (Primix), etc, but if this were designed
>> in from the start, like AIX paritally has.
>>
>> OSF/1 could have been both a multics-like OS and have a unix
>> top layer, but with some hypervisor-like functions for the extra
>> functionality that is unknown to the unix api and ui.
>
> Just think what the OS could have been with designers who had
> Unix, TOPS-10, TOPS-20, VMS and PDP-11 and -8 OS experience.

Doesn’t necessarily mean that it would have ended up with the
best of all of those. There are plenty of situations with OS where
its just not possible to have all of the best of the implementations
because most things come with downsides and any real design
has to decide what pros and cons matter most.

For example, sandboxing is an excellent approach for ensuring
that rogue apps can't snoop on any data that doesn’t belong
to them, but has the massive downside that that means that
simple file ops aren't available either.

> If the development process had acquired the Multics'
> methods, security would not be an afterthought.

But that has real downsides with useability.

> OSF/1 blew it; not even the project leader could change
> the minds of the managers who wanted yet another unix.
>>
>> QNX has done this, with a "just another unix" interface and a
>> "real QNX" interface.
>
> The interface could have been pluggable so that moving to an
> OSF/1 system did not require redoing the entire computing
> of a customer site.

But pluggable interfaces have their own real downsides.

>> Well, perhaps some OSF/1 designers did.
>
> they did and fought but lost.
 

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

#9529

Morten Reistad <first@last.name.invalid> writes:
> The OSF/1 designers nevers saw the possibility to have multiple
> personalites. These have a bad reputation after the "unix" retrofits
> for VMS(Eunice?), Primos (Primix), etc, but if this were designed
> in from the start, like AIX paritally has. 
>
> OSF/1 could have been both a multics-like OS and have a unix
> top layer, but with some hypervisor-like functions for the extra
> functionality that is unknown to the unix api and ui.
>
> QNX has done this, with a "just another unix" interface and a
> "real QNX" interface.
>
> Well, perhaps some OSF/1 designers did. 

the politics I saw was that it was response to AT&T/SUN tieup
https://en.wikipedia.org/wiki/Unix_wars

While this decision was applauded by customers and the trade press,
certain other Unix licensees feared Sun would be unduly advantaged. They
formed the Open Software Foundation (OSF) in 1988. The same year, AT&T
and another group of licensees responded by forming UNIX International
(UI). Technical issues soon took a back seat to vicious and public
commercial competition between the two "open" versions of Unix, with
X/Open holding the middle ground. A 1990 study of various Unix versions'
reliability found that on each version, between a quarter and a third of
operating system utilities could be made to crash by fuzzing; the
researchers attributed this, in part, to the "race for features, power,
and performance" resulting from BSD-System V rivalry, which left
developers little time to worry about reliability.[2]

....

there was focus on picking up technologies from unix work-alikes
... stuff from UCB (BSD), MIT (x-windows), CMU (MACH, Andrew), UCLA
(LOCUS), etc.

IBM Palo Alto Science Center had been working with UCB to do port to
mainframe 370 ... when they got redirected to do the port to the PC/RT
(which came out as AOS, alternative to AIX). They had also been working
with UCLA on LOCUS ... and they eventually did ports to both 370 & 386
... which was released as AIX/370 & AIX/386.
https://en.wikipedia.org/wiki/LOCUS_%28operating_system%29
and
https://en.wikipedia.org/wiki/Locus_Computing_Corporation

-- 
virtualization experience starting Jan1968, online at home since Mar1970

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

#9530

more OSF (and other) topic drift & trivia: I remember one OSF meeting
where the official DEC rep I knew from the 70s and had been at the
science center and vm370 development group.

the science center ... some past posts
http://www.garlic.com/~lynn/subtopic.html#545tech

had done (virtual machine) cp67. As the cp67/cms group grew, it split
off from the science center (on 4th flr, multics was on 5th flr) and
took over the IBM Boston Programming center (on the 3rd flr). As the
group expanded ... especially with work on cp67 morph to vm370 ...  they
moved out to the vacant former SBC bldg in Burlington Mall.

During the FS period
http://www.garlic.com/~lynn/subtopic.html#futuresys

lots of groups were directed to refocus on FS stuff and suspend 370
activity. When FS imploded, there was mad rush to get stuff back into
the 370 product pipelines. As part of that, the head of POK eventually
convinced corporate to kill the vm370 product, shutdown the Burlington
Mall group and move everybody to POK to work on MVS/XA (or otherwise
MVS/XA wouldn't ship on time some 6-7 years later).

They weren't going to tell the VM370 people until the various last
minute to minimize the number that could escape the move to POK, however
the information managed to leak and some number managed to escape (there
was joke that head of POK was one of the largest contributors to DEC
VMS, some others showed up at Prime). There was also a witch hunt for
the source of the leak ...  fortunately for me, nobody gave up the
source.

-- 
virtualization experience starting Jan1968, online at home since Mar1970

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

#9535

Morten Reistad <first@last.name.invalid> writes:
>In article <PM0005246CA6348F23@aca411e9.ipt.aol.com>,
>jmfbahciv  <See.above@aol.com> wrote:
>>William Pechter wrote:
>>> In article <PM0005245889FBE0A0@aca40f93.ipt.aol.com>,
>>> jmfbahciv  <See.above@aol.com> wrote:
>>>>Scott Lurndal wrote:
>>>>> jmfbahciv <See.above@aol.com> writes:
>>>>>>Scott Lurndal wrote:
>>>>>>> jmfbahciv <See.above@aol.com> writes:
>>>>>
>>>>>>>>Both were shit products.  VMS got better after the PDP-10 developers
>>>>>>>>got hired.
>>>>>>>
>>>>>>> While I agree that DECTape wasn't the brightest idea,
>>>>>>> your hatred of the VAX is personal; it was obviously a
>>>>>>> great product if you gauge by shipments, revenue and
>>>>>>> user testimonials.
>>>>>>
>>>>>>Can you see the difference between VMS and VAX?
>>>>>>
>>>>>
>>>>> Given that one didn't exist without the other (at least
>>>>> until the BSD port), no.
>>>>
>>>>Ah, I see.  You're a customer.  On my side of the biz,
>>>>they were two very different entities.
>>>>
>>>>/BAH
>>>
>>> There were differences between what LCG development was like and what the
>>> rest of DEC was like.  In the real world VAX wasn't VMS although VMS was
>>> VAX.
>>>
>>> As far as your view on VMS -- it was pretty distorted compared with the
>>views
>>> of customers in the field.  The major colleges and universities which used
>>> TOPS-10 and TOPS-20 weren't as pleased since the VAX was more a departmental
>>> box than a mainframe.  It wasn't designed to support thousands of users.
>>
>>It's not distorted.  10 and 20 customers were told that VAX/VMS was going
>>to replace the PDP-10 business.  No choice in the matter.  That's what I,
>>and a lot of other people, especially including customers, objected to.
>>VAX/VMS was fine for a particular niche, but not a viable replacement for
>>the work that the PDP-10s did.  It was insane for Bell to insist that happen.
>[snip]
>>Originally, OSF/1 was not supposed to be yet another Unix.  JMF was part
>>of the original design group and got extrememly despressed when he
>>firgured out that the politics had determined that it would be another unix.
>>The original goal was to design an OS which had the strengths of all
>>the OSes and avoid the problems of those same OSes.
>
>The OSF/1 designers nevers saw the possibility to have multiple
>personalites. These have a bad reputation after the "unix" retrofits
>for VMS(Eunice?), Primos (Primix), etc, but if this were designed
>in from the start, like AIX paritally has. 
>
>OSF/1 could have been both a multics-like OS and have a unix
>top layer, but with some hypervisor-like functions for the extra
>functionality that is unknown to the unix api and ui.
>
>QNX has done this, with a "just another unix" interface and a
>"real QNX" interface.
>
>Well, perhaps some OSF/1 designers did. 

At the same time OSF/1 was being designed, the Amadeus[*] project
was designing a competitive system based on the Chorus Systemes
microkernel.   A Unix subsystem was derived from SVR4.2ES/MP
to run on the microkernel, but there was nothing to preclude any
other subsystem from being implemented - the microkernel was
completely agnostic.

Amadeus team included Unisys, Chorus, USL, Fujitsu, ICL and
a couple others.

The Unix subsystem was completely distributed and ran on multiple
workstations and high-end MPP systems (e.g. the Unisys OPUS).

[*] An EU initiative.

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

#9487

jmfbahciv <See.above@aol.com> writes:
>Scott Lurndal wrote:
>> jmfbahciv <See.above@aol.com> writes:
>>>Scott Lurndal wrote:
>>>> jmfbahciv <See.above@aol.com> writes:
>>
>>>>>Both were shit products.  VMS got better after the PDP-10 developers
>>>>>got hired.
>>>>
>>>> While I agree that DECTape wasn't the brightest idea,
>>>> your hatred of the VAX is personal; it was obviously a
>>>> great product if you gauge by shipments, revenue and
>>>> user testimonials.
>>>
>>>Can you see the difference between VMS and VAX?
>>>
>>
>> Given that one didn't exist without the other (at least
>> until the BSD port), no.
>
>Ah, I see.  You're a customer.  On my side of the biz,
>they were two very different entities.

Clearly I wasn't a DEC employee.  I was, however, an operating
system designer for a large mainframe company.

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

#9439

In article <PM00052430AAEE24F6@aca41c5c.ipt.aol.com>,
jmfbahciv  <See.above@aol.com> wrote:
>Bob Eager wrote:
>> On Mon, 09 Nov 2015 14:13:41 +0000, jmfbahciv wrote:
>>
>>> Morten Reistad wrote:
>>>> In article <877flrwsa6.fsf@elektro.pacujo.net>,
>>>> Marko Rauhamaa  <marko@pacujo.net> wrote:
>>>>>RS Wood <rsw@therandymon.com>:
>>>>>
>>>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>>>>
>>>>>I don't know. Who boots a computer nowadays?
>>>>
>>>> I do, the ones I develop on. And they are up within a minute,
>>>> all of them. With all subsystems like apache, mysql etc.
>>>>
>>>> The laptops come out of hibernation usually as fast as I can open the
>>>> lid, but sometimes there is a ~2s delay, which may be some swapping
>>>> taking place.
>>>>
>>>> The annoying bit is the dhcp stuff for ipv4, which takes around 10
>>>> seconds. ipv6 is up within a second.
>>>>
>>>> I contrast this with the minis at various PPOEs, which all took around
>>>> 30 minutes to come up completely.
>>>
>>> That's (30 minutes boot-up) is horrid.  DEC would consider that a bug.
>>> Note that none of the OS developers would wait that long for a restart
>>> so they would work on getting a system up quickly.
>>
>> Never booted VMS off a DECtape II then?  :)
>
>Both were shit products.  VMS got better after the PDP-10 developers
>got hired.
>
>IIRC, most of VMS' thrashing at boot time was to figure out which
>products were "allowed" to be loaded.  IOW, if the site/system
>had the proper licenses.

Not quite. Even our home-built daemons (in fortran and C, requiring
only RMS and the basic OS as libraries) took more than a quarter of
an hour to load in a 4M 11/750, assigning 28 async ports and setting
up a huge "poll()" on them. 

Also, the same 28 internally scheduled threads had one file pointer
into the same file, an RMS indexed file. That contributed to this
taking ages to start.

But once it was running it ran pretty fast. (for an 11/750, at least).

The very same program on a Prime 50-series took around 2 minutes to
start, but needed around 10 minutes worth of preceding segment sharing, 
so they were both as slow. (on a 550-II).

The program in question is the Z00 driver for disseminating stock
ticker data in real time; which at that time meant within 100 ms of
the actual trade, and with a max skew of 50ms randomly distributed
between recipients. We went to great lengths to ensure these criteria, 
which added around 10 lines of code, but very carefully placed code.

The VMS implementation has N workers internally scheduled in a 
realtime process (because the VMS schedulers stunk); the Primos one
had individual phantoms with lots of shared memory, and a large use
of the recycl() os call, to give up the rest of the time slice to
the next process in line. 

-- mrr

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

#9432

In article <PM0005241C7B957C4D@aca432c8.ipt.aol.com>,
jmfbahciv  <See.above@aol.com> wrote:
>Morten Reistad wrote:
>> In article <877flrwsa6.fsf@elektro.pacujo.net>,
>> Marko Rauhamaa  <marko@pacujo.net> wrote:
>>>RS Wood <rsw@therandymon.com>:
>>>
>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>>
>>>I don't know. Who boots a computer nowadays?
>>
>> I do, the ones I develop on. And they are up within a minute,
>> all of them. With all subsystems like apache, mysql etc.
>>
>> The laptops come out of hibernation usually as fast as I can
>> open the lid, but sometimes there is a ~2s delay, which may
>> be some swapping taking place.
>>
>> The annoying bit is the dhcp stuff for ipv4, which takes
>> around 10 seconds. ipv6 is up within a second.
>>
>> I contrast this with the minis at various PPOEs, which
>> all took around 30 minutes to come up completely.
>
>That's (30 minutes boot-up) is horrid.  DEC would consider
>that a bug.  Note that none of the OS developers would
>wait that long for a restart so they would work on
>getting a system up quickly.

Several VAXen are among the set of 30-minute-boots.

They were more or less identical to the Primes I got
to know so well.

One PPOE had a time sheet for this. One minute power
sequencing and basic diag from the VCP. 6 minutes vcp loading
of microcode into main processor. 3 minute OS load and 
initialisation. 4 minutes starting disk partitions with
roll forward/roll back transactions. 5 minutes sharing
code. Then startup of daemons for printers, databases, etc.
12 minutes (really 20, but we opened for logins at the 30
min mark).

My laptop does a lot of this startup too. It takes all
of 6 seconds to load and initialise the kernel, and a 
further 12 before all is up and logins can start. Before
this there is a 4-5 second bios delay.

So, in this respect things have gotten around 100 times
better. 

-- mrr

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

#9433

In article <PM0005241C7B957C4D@aca432c8.ipt.aol.com>,
jmfbahciv  <See.above@aol.com> wrote:
>Morten Reistad wrote:
>> In article <877flrwsa6.fsf@elektro.pacujo.net>,
>> Marko Rauhamaa  <marko@pacujo.net> wrote:
>>>RS Wood <rsw@therandymon.com>:
>>>
>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>>
>>>I don't know. Who boots a computer nowadays?
>>
>> I do, the ones I develop on. And they are up within a minute,
>> all of them. With all subsystems like apache, mysql etc.
>>
>> The laptops come out of hibernation usually as fast as I can
>> open the lid, but sometimes there is a ~2s delay, which may
>> be some swapping taking place.
>>
>> The annoying bit is the dhcp stuff for ipv4, which takes
>> around 10 seconds. ipv6 is up within a second.
>>
>> I contrast this with the minis at various PPOEs, which
>> all took around 30 minutes to come up completely.
>
>That's (30 minutes boot-up) is horrid.  DEC would consider
>that a bug.  Note that none of the OS developers would
>wait that long for a restart so they would work on
>getting a system up quickly.
>
>/BAH

Oh?  May I acquaint you with the boot up time on a Microvax II running
Ultrix?  

Hank

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

#9436

Hank wrote:
> In article <PM0005241C7B957C4D@aca432c8.ipt.aol.com>,
> jmfbahciv  <See.above@aol.com> wrote:
>>Morten Reistad wrote:
>>> In article <877flrwsa6.fsf@elektro.pacujo.net>,
>>> Marko Rauhamaa  <marko@pacujo.net> wrote:
>>>>RS Wood <rsw@therandymon.com>:
>>>>
>>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>>>
>>>>I don't know. Who boots a computer nowadays?
>>>
>>> I do, the ones I develop on. And they are up within a minute,
>>> all of them. With all subsystems like apache, mysql etc.
>>>
>>> The laptops come out of hibernation usually as fast as I can
>>> open the lid, but sometimes there is a ~2s delay, which may
>>> be some swapping taking place.
>>>
>>> The annoying bit is the dhcp stuff for ipv4, which takes
>>> around 10 seconds. ipv6 is up within a second.
>>>
>>> I contrast this with the minis at various PPOEs, which
>>> all took around 30 minutes to come up completely.
>>
>>That's (30 minutes boot-up) is horrid.  DEC would consider
>>that a bug.  Note that none of the OS developers would
>>wait that long for a restart so they would work on
>>getting a system up quickly.
>>
>>/BAH
>
> Oh?  May I acquaint you with the boot up time on a Microvax II running
> Ultrix?

I never did one of those.  Both products happened after DEC became
Digital ;-).  Did it really take that long?  I'm surprised any "old"
customer bought the product.

/BAH

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

#9428

Marko Rauhamaa wrote:
> RS Wood <rsw@therandymon.com>:
>
>> (That said, I /do/ wish boot times were 0.5 seconds again).
>
> I don't know. Who boots a computer nowadays?

I do.

/BAH

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

#9431

"jmfbahciv" <See.above@aol.com> wrote in message 
news:PM0005241C72B624CC@aca432c8.ipt.aol.com...
> Marko Rauhamaa wrote:
>> RS Wood <rsw@therandymon.com>:
>>
>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>
>> I don't know. Who boots a computer nowadays?
>
> I do.

Yeah, but you have always been a fucking dinosaur
and are actually stupid enough to run down to the
library to look something up when everyone else
uses this funky system called the net that has been
available for quite some time now. 

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

#9437

Rod Speed wrote:
>
>
> "jmfbahciv" <See.above@aol.com> wrote in message
> news:PM0005241C72B624CC@aca432c8.ipt.aol.com...
>> Marko Rauhamaa wrote:
>>> RS Wood <rsw@therandymon.com>:
>>>
>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>>
>>> I don't know. Who boots a computer nowadays?
>>
>> I do.
>
> Yeah, but you have always been a fucking dinosaur

Which makes me an excellent field test site.

> and are actually stupid enough to run down to the
> library to look something up when everyone else
> uses this funky system called the net that has been
> available for quite some time now.
>
Which keeps me from spending more than $1200/year.

/BAH

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

#9441

In article <PM000524309B41BCAE@aca41c5c.ipt.aol.com>,
jmfbahciv  <See.above@aol.com> wrote:
>Rod Speed wrote:
>>
>>
>> "jmfbahciv" <See.above@aol.com> wrote in message
>> news:PM0005241C72B624CC@aca432c8.ipt.aol.com...
>>> Marko Rauhamaa wrote:
>>>> RS Wood <rsw@therandymon.com>:
>>>>
>>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>>>
>>>> I don't know. Who boots a computer nowadays?
>>>
>>> I do.
>>
>> Yeah, but you have always been a fucking dinosaur
>
>Which makes me an excellent field test site.
>
>> and are actually stupid enough to run down to the
>> library to look something up when everyone else
>> uses this funky system called the net that has been
>> available for quite some time now.
>>
>Which keeps me from spending more than $1200/year.

Eh?

I live in a not-too-cheap-at-all Internet country. We (in .no)
pay typically a third more for our Internet than people
in .se or .dk.

I went for fiber in the "new" house, with a gigabit up/downlink, 
and pay for a CIR of 50, NOK 449/month. Or NOK 5388/year. Which
is almost exactly $600/year. In addition I pay around $40 for
some dyndns and secondary mail/dns servers, so I can access
the home from anywhere in the world via a VPN. Or via ipv6; 
i have a /56 assigned. (2001:16d8:ee00:8100::/56)

I have servers permanenty on as firewall, fiber modem etc, and the
power meter says it uses 78 watts in semi-idle. The fiber modem
uses 18 alone, and the mandatory access box another 22.  So, add another
$78 to cover the electricity. I run everything on 12V, and
have two large boat batteries of 120Ah as UPS. Which should
give me a day and a half to fire up a generator. 

Yes, the power stability around here is very bad right now, 
living within 200 meters of two large residential developments
of >200 units each; having them mess with the 21kV network
almost daily. 

This is the total of what I pay, and I get orders of magnitude
better Internet connectivity than what you have, for a little
more than half the cost. And there are nothing metered except
the electricity. I really have "an internet of things" here, now
I have 30 ipv4 and 22 ipv6 addresses assigned. Two of these are
for my tops20/klh10 "virtual machines" .

-- mrr

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

#9460

On Tue, 10 Nov 2015 18:05:59 +0100
Morten Reistad <first@last.name.invalid> wrote:

> I went for fiber in the "new" house, with a gigabit up/downlink, 
> and pay for a CIR of 50, NOK 449/month. Or NOK 5388/year. Which
> is almost exactly $600/year. 

	Hmm over 500 Euro a year for 6Mb down 1Mb up here, we *might* get
fibre before the 2020s but I doubt it.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:>WIN                                      | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

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

#9462

En el artículo <20151111105218.11f857ab074958716a223643@eircom.net>,
Ahem A Rivet's Shot <steveo@eircom.net> escribió:

>       Hmm over 500 Euro a year for 6Mb down 1Mb up here

Ouch.

>, we *might* get
>fibre before the 2020s but I doubt it.

No FTTC?

I thought Eircom was one of the more clued-up telecoms providers.

-- 
 (\_/)
(='.'=) Bunny says: Windows 10? Nein danke!
(")_(")

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

#9463

On Wed, 11 Nov 2015 11:40:56 +0000
Mike Tomlinson <mike@jasper.org.uk> wrote:

> En el artículo <20151111105218.11f857ab074958716a223643@eircom.net>,
> Ahem A Rivet's Shot <steveo@eircom.net> escribió:
> 
> >       Hmm over 500 Euro a year for 6Mb down 1Mb up here
> 
> Ouch.

	The joys of rural Ireland. Although to be fair I am looking at an
alternative service which offers 20Mb down 5Mb up for about 20% more.

> >, we *might* get
> >fibre before the 2020s but I doubt it.
> 
> No FTTC?

	Not even a C to put the F in - as this place got near to being ready
to move in we took up an offer for a landline phone and 24Mb DSL feed for
a good price and free installation. Engineer turned up and noticed lack of
pole (we had told them) and called in an order for poles. Eircom engineer
turns up a few days later works out where to hook us up and marks locations
for two poles. Time passes, more time passes, yet more time passes, even
m... ah you get the idea. After several months of "it'll be next week for
sure" we told them where to stick their poles should they ever find them
and went with a fixed wireless service which was up and running in less
than two weeks.

> I thought Eircom was one of the more clued-up telecoms providers.

	There's great service in (some parts) of the major cities, OTOH if
you live in a cul-de-sac a couple of miles outside of a medium sized
village some 20 minutes drive from the nearest largish town then it's not
so great - in fact if you live in a new build in such a place it can be
essentially non existent.

	Should I hear about fibre turning up round here I will start round
two of getting poles, probably direct with Eircon in the (faint) hope that
they fill their own orders more effectively.

	Back when they were Telecom Eireann I used to call them Telecon
Airhead, I'm not sure they've improved.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:>WIN                                      | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

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

#9445

jmfbahciv <See.above@aol.com> wrote
> Rod Speed wrote
>> jmfbahciv <See.above@aol.com> wrote
>>> Marko Rauhamaa wrote
>>>> RS Wood <rsw@therandymon.com>:

>>>>> (That said, I /do/ wish boot times were 0.5 seconds again).

>>>> I don't know. Who boots a computer nowadays?

>>> I do.

>> Yeah, but you have always been a fucking dinosaur

> Which makes me an excellent field test site.

Like hell it does when on one with even
half a clue operates like you do anymore.

>> and are actually stupid enough to run down to
>> the library to look something up when everyone
>> else uses this funky system called the net that
>> has been available for quite some time now.

> Which keeps me from spending more than $1200/year.

Doesn’t cost anything even remotely like that. 

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

#9451

On Wed, 11 Nov 2015 06:25:54 +1100, Rod Speed wrote:

> jmfbahciv <See.above@aol.com> wrote
>> Rod Speed wrote
>>> jmfbahciv <See.above@aol.com> wrote
>>>> Marko Rauhamaa wrote
>>>>> RS Wood <rsw@therandymon.com>:
> 
>>>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
> 
>>>>> I don't know. Who boots a computer nowadays?
> 
>>>> I do.
> 
>>> Yeah, but you have always been a fucking dinosaur
> 
>> Which makes me an excellent field test site.
> 
> Like hell it does when on one with even half a clue operates like you do
> anymore.
> 
>>> and are actually stupid enough to run down to the library to look
>>> something up when everyone else uses this funky system called the net
>>> that has been available for quite some time now.
> 
>> Which keeps me from spending more than $1200/year.
> 
> Doesn’t cost anything even remotely like that.

 Well, that's three *actual* posts that I've seen from this troll. I'm 
used to seeing ghost sightings of him in others postings in the other NGs 
he trolls but I rather suspect I'll not be seeing many such ghosts once 
I've killfiled him in this NG.

-- 
Johnny B Good

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

#9453

"Johnny B Good" <johnny-b-good@invalid.ntlworld.com> wrote in message 
news:GXu0y.123070$mn4.55497@fx42.am4...
> On Wed, 11 Nov 2015 06:25:54 +1100, Rod Speed wrote:
>
>> jmfbahciv <See.above@aol.com> wrote
>>> Rod Speed wrote
>>>> jmfbahciv <See.above@aol.com> wrote
>>>>> Marko Rauhamaa wrote
>>>>>> RS Wood <rsw@therandymon.com>:
>>
>>>>>>> (That said, I /do/ wish boot times were 0.5 seconds again).
>>
>>>>>> I don't know. Who boots a computer nowadays?
>>
>>>>> I do.
>>
>>>> Yeah, but you have always been a fucking dinosaur
>>
>>> Which makes me an excellent field test site.
>>
>> Like hell it does when on one with even half a clue operates like you do
>> anymore.
>>
>>>> and are actually stupid enough to run down to the library to look
>>>> something up when everyone else uses this funky system called the net
>>>> that has been available for quite some time now.
>>
>>> Which keeps me from spending more than $1200/year.
>>
>> Doesn’t cost anything even remotely like that.
>
> Well, that's three *actual* posts that I've seen from this troll. I'm
> used to seeing ghost sightings of him in others postings in the other NGs
> he trolls but I rather suspect I'll not be seeing many such ghosts once
> I've killfiled him in this NG.

No one actually gives a flying red fuck what you do or do not read. 

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

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

Back to top | Article view | comp.misc

csiph-web