| Re: The ICL 2900 [message #338190 is a reply to message #338165] |
Thu, 23 February 2017 09:39   |
Ahem A Rivet's Shot
Messages: 4843
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On Thu, 23 Feb 2017 14:03:50 GMT
scott@slp53.sl.home (Scott Lurndal) wrote:
> Ahem A Rivet's Shot <steveo@eircom.net> writes:
>> On Wed, 22 Feb 2017 21:19:41 GMT
>> scott@slp53.sl.home (Scott Lurndal) wrote:
>>
>>> Morten Reistad <first@last.name.invalid> writes:
>>>> In article <20170222153929.e10603983a258bcb672dd293@eircom.net>,
>>>> Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>>>> >On 22 Feb 2017 14:39:24 GMT
>>>> >jmfbahciv <See.above@aol.com> wrote:
>>>> >
>>>> >> Aren't today's controllers a part of the disk drive? So there's no
>>>> >> longer one controller for many drives.
>>>> >
>>>> > It's layered but certainly there's a controller on the drive
>>>> > that
>>>> >usually supports queuing commands, remapping bad sectors, keeping
>>>> >track of drive health and presenting a standard interface (SATA or
>>>> >SAS these days), then there's a controller that interfaces a system
>>>> >I/O bus (usually some flavour of PCI these days) to one or more SATA
>>>> >or SAS channels, then there's a controller that interfaces the I/O
>>>> >bus to the CPU/memory bus.
>>>>
>>>> Normally CPUs have three main buses, the Front Side bus, the North
>>>> Side bus and the South Side bus.
>>>
>>> 10 years ago, perhaps. But the frontside/northside bus was eliminated
>>> with AMD's original Hypertransport (2001), and followed by Intel when
>>> they introduced Quickpath (QPI) in 2008.
>>
>> This is why I left that bit as just CPU/memory bus - because I
>> know
>> there are a number of variations on the detail here, and you're just
>> looking at x86-64 CPUs - things like the Cavium ThunderX 48 core ARM SOC
>> are different again.
>
> I did mention ThunderX in the part of the article you _didn't_ quote.
Sorry I missed that, it was the variations on the interconnects
that I was referring to with "a number of variations on the detail". I
don't know how similar these interconnects really are but they seem to be a
key part of the distinction between the internal architectures from
different manufacturers.
--
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/
|
|
|
|
| Re: The ICL 2900 [message #338191 is a reply to message #338171] |
Thu, 23 February 2017 09:51 |
Ahem A Rivet's Shot
Messages: 4843
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On 23 Feb 2017 14:22:58 GMT
jmfbahciv <See.above@aol.com> wrote:
> I'm quite proficient learning new OSes. There are no books which
> train new system owners. Part of the problem is that there is no
> common user interface because of the shell concept. I'd like to
> see the command language which is underneath those shells.
There isn't one, beneath the shell there is just the libraries
(section 3 of the manual) and the OS call interface (section 2 of the
manual), oh and C of course which is what most of it is written in.
--
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/
|
|
|
|
| Re: The ICL 2900 [message #338195 is a reply to message #338048] |
Thu, 23 February 2017 10:32 |
Quadibloc
Messages: 4399
Registered: June 2012
Karma: 0
|
Senior Member |
|
|
On Wednesday, February 22, 2017 at 7:40:32 AM UTC-7, jmfbahciv wrote:
> Quadibloc wrote:
>> On Tuesday, February 21, 2017 at 5:55:36 AM UTC-7, jmfbahciv wrote:
>>
>>> When did the A come out? I worked with 8/Is. If A came before the I's,
>>> then it came before the PDP-10s.
>>
>> No, the PDP-8/A was the very last PDP-8. It had a calculator keyboard on the
> front
>> panel instead of switches.
>
> I don't think I met an 8/A. That's a shame.
Here's a page with pictures of one, and some technical details.
http://www.oldcomputers.arcula.co.uk/pdp84.htm
It was indeed made long enough ago for the cost of memory to still be a major
issue - according to another web site, it was first made available in 1974.
John Savard
|
|
|
|
| Re: The ICL 2900 [message #338196 is a reply to message #338190] |
Thu, 23 February 2017 10:57 |
scott
Messages: 4239
Registered: February 2012
Karma: 0
|
Senior Member |
|
|
Ahem A Rivet's Shot <steveo@eircom.net> writes:
> On Thu, 23 Feb 2017 14:03:50 GMT
> scott@slp53.sl.home (Scott Lurndal) wrote:
>
>>>> 10 years ago, perhaps. But the frontside/northside bus was eliminated
>>>> with AMD's original Hypertransport (2001), and followed by Intel when
>>>> they introduced Quickpath (QPI) in 2008.
>>>
>>> This is why I left that bit as just CPU/memory bus - because I
>>> know
>>> there are a number of variations on the detail here, and you're just
>>> looking at x86-64 CPUs - things like the Cavium ThunderX 48 core ARM SOC
>>> are different again.
>>
>> I did mention ThunderX in the part of the article you _didn't_ quote.
>
> Sorry I missed that, it was the variations on the interconnects
> that I was referring to with "a number of variations on the detail". I
> don't know how similar these interconnects really are but they seem to be a
> key part of the distinction between the internal architectures from
> different manufacturers.
The internal architectures fall into one of two general categories:
- They have an internal ring bus that circulates between per-core
caches, the memory controllers and the I/O side. This ring is
wide enough to transfer a cache line (64 bytes) in a single clock.
With larger core counts, there are often two of these that circulate
in opposite directions (to reduce latency).
- They use a non-blocking crossbar switched interconnect between
per-core caches and the memory/IOsystem(s).
These rings/busses are not exposed off-chip.
The external interconnects (HT, QPI, CCPI) are very similar
to each other in effect, but the implementations vary based
on the coherence algorithms implemented by the cache/memory
subsystem. All of them are high-speed serial links up to
some number of lanes in width (the number varies depending
on how many sockets are used). In a two-socket, one can
gang all the lanes for better bandwidth.
QPI, for example, has gone through several generations. The
first generation allowed either home-snooping or source-snooping
behavior. The currently generation only does home-snooping.
http://www.realworldtech.com/qpi-evolved/3/
|
|
|
|
| Re: The ICL 2900 [message #338197 is a reply to message #338191] |
Thu, 23 February 2017 11:03 |
scott
Messages: 4239
Registered: February 2012
Karma: 0
|
Senior Member |
|
|
Ahem A Rivet's Shot <steveo@eircom.net> writes:
> On 23 Feb 2017 14:22:58 GMT
> jmfbahciv <See.above@aol.com> wrote:
>
>> I'm quite proficient learning new OSes. There are no books which
>> train new system owners. Part of the problem is that there is no
>> common user interface because of the shell concept. I'd like to
>> see the command language which is underneath those shells.
>
> There isn't one, beneath the shell there is just the libraries
> (section 3 of the manual) and the OS call interface (section 2 of the
> manual), oh and C of course which is what most of it is written in.
The shell is a command language. The strength of Unix is that
one can easily build a different command language to match ones needs.
There is _no_ command language provided by the operating system itself,
only a general programming services API (system calls for privileged
operations or OS functions and library functions for anything non-privileged).
This is unlike DEC operating systems such as VAX/VMS where DCL was
part of the operating system (and executed in Ring 1),
record management services (RMS) executed in ring 2 and the
kernel (or SYS$CMKRNL) executed in ring 3.
|
|
|
|
| Re: The ICL 2900 Buying a computer in the 1960s [message #338198 is a reply to message #338186] |
Thu, 23 February 2017 11:14 |
Anne & Lynn Wheel
Messages: 3156
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
pechter@pechter.dyndns.org (William Pechter) writes:
> That was the way at DEC, IBM, Bell Labs. All of those companies started
> to lose that in the cutbacks of the 90's to 2000s.
In the late 70s, 3270 terminals were viewed as expensive/scarce resource
.... most places relegated to terminal rooms and budgeted as part of the
fall plan ... even tho some of us done a business case showing the 3yr
amortized cost of 3270 terminal was about the same monthly cost of
business phone that was default on all desks.
There was a moment around 1980 where there was a rapidly spreading rumor
that some of the top IBM executives were starting to communicate via
email. At that point there was a rush by middle management to pre-empt
the annual 3270 terminal allocation, redirecting them to their desks
.... making it appear like they too were computer literate.
After IBM/PCs and 3270 terminal emulation, the biggest and largest
IBM/PCs became status symbols for middle management ... even if it met
intercepting allocation needed for development projects. They typically
were turn on in the morning with the VM370 logo burning into the face of
the screen ... or in some cases, logged on and the PROFS menu burning
itself into the screen (while administrative assistants actually handle
things like email ... frequently even printing paper hard copy so the
middle manager could read them).
recent posts mentioning PROFS:
http://www.garlic.com/~lynn/2017.html#68 Trump White House Senior Staff Have Private RNC Email Accounts
http://www.garlic.com/~lynn/2017.html#98 360 & Series/1
http://www.garlic.com/~lynn/2017b.html#10 IBM 1970s
http://www.garlic.com/~lynn/2017b.html#74 The ICL 2900
http://www.garlic.com/~lynn/2017b.html#78 The ICL 2900
some past posts mentioning rapidly spreading rumor that top executives
communicated by email and middle management started pre-empting 3270
terminal allocation as status symbols:
http://www.garlic.com/~lynn/2006n.html#23 sorting was: The System/360 Model 20 Wasn't As Bad As All That
http://www.garlic.com/~lynn/2006w.html#7 Why these original FORTRAN quirks?
http://www.garlic.com/~lynn/2009l.html#41 another item related to ASCII vs. EBCDIC
http://www.garlic.com/~lynn/2009m.html#34 IBM Poughkeepsie?
http://www.garlic.com/~lynn/2009q.html#43 The 50th Anniversary of the Legendary IBM 1401
http://www.garlic.com/~lynn/2009q.html#49 The 50th Anniversary of the Legendary IBM 1401
http://www.garlic.com/~lynn/2009q.html#51 The 50th Anniversary of the Legendary IBM 1401
http://www.garlic.com/~lynn/2010c.html#88 search engine history, was Happy DEC-10 Day
http://www.garlic.com/~lynn/2011d.html#13 I actually miss working at IBM
http://www.garlic.com/~lynn/2012d.html#37 IBM cuts more than 1,000 U.S. Workers
http://www.garlic.com/~lynn/2012d.html#40 Strategy subsumes culture
http://www.garlic.com/~lynn/2013b.html#58 Dualcase vs monocase. Was: Article for the boss
http://www.garlic.com/~lynn/2015g.html#98 PROFS & GML
http://www.garlic.com/~lynn/2016g.html#89 "I used a real computer at home...and so will you" (Popular Science May 1967)
--
virtualization experience starting Jan1968, online at home since Mar1970
|
|
|
|
| Re: The ICL 2900 Buying a computer in the 1960s [message #338199 is a reply to message #338186] |
Thu, 23 February 2017 11:41 |
Anne & Lynn Wheel
Messages: 3156
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
pechter@pechter.dyndns.org (William Pechter) writes:
> That was the way at DEC, IBM, Bell Labs. All of those companies started
> to lose that in the cutbacks of the 90's to 2000s.
re:
http://www.garlic.com/~lynn/2017b.html#80 The ICL 2900 Buying a computer in the 1960s
I was blamed for online computer conferencing on the internal network
(larger than the arpanet/internet from just about the beginning until
some time mid-80s) in the late 70s and early 80s. Folklore is that
when the corporate executive committee was informed of online computer
conferencing (and the internal network), 5of6 wanted to fire me.
some past posts
http://www.garlic.com/~lynn/subnetwork.html#internalnet
and
http://www.garlic.com/~lynn/subnetwork.html#cmc
a periodically virulent one was a trip report that i distributed about
visiting Jim Gray a few months after he went to tandem ... from IBMJARGON:
Tandem Memos - n. Something constructive but hard to control; a fresh of
breath air (sic). That's another Tandem Memos. A phrase to worry middle
management. It refers to the computer-based conference (widely
distributed in 1981) in which many technical personnel expressed
dissatisfaction with the tools available to them at that time, and also
constructively criticised the way products were are developed. The memos
are required reading for anyone with a serious interest in quality
products. If you have not seen the memos, try reading the November 1981
Datamation summary.
.... snip ...
some number of task forces were kicked off afterwards ... and
also conflated with a memo that Jim wrote on leaving IBM, also
from IBMJARGON:
MIP envy - n. The term, coined by Jim Gray in 1980, that began the
Tandem Memos (q.v.). MIP envy is the coveting of other's facilities -
not just the CPU power available to them, but also the languages,
editors, debuggers, mail systems and networks. MIP envy is a term every
programmer will understand, being another expression of the proverb The
grass is always greener on the other side of the fence.
.... snip ...
and as referenced, it was a trip report that I distributed that kicked
things off ... not Jim's earlier goodby memo ... a version hear:
http://www.garlic.com/~lynn/2007d.html#email800920
in this post
http://www.garlic.com/~lynn/2007d.html#17 Jim Gray Is Missing
slightly different version here
http://research.microsoft.com/en-us/um/people/gray/papers/MI PEnvy.pdf
In any case, one of the task forces kicked off was to visit and report
on other research institutions (including bell labs) ... different
pieces/references in these past posts:
http://www.garlic.com/~lynn/2001l.html#61 MVS History (all parts)
http://www.garlic.com/~lynn/2006n.html#56 AT&T Labs vs. Google Labs - R&D History
http://www.garlic.com/~lynn/2008m.html#11 Yet another squirrel question - Results (very very long post)
include surveys of computer resources, CMU, Stanford, LBL, MIT, Bell
Labs, etc
other recent reference to Jim palming stuff on me when he left for
tandem:
http://www.garlic.com/~lynn/2017b.html#75 The ICL 2900
--
virtualization experience starting Jan1968, online at home since Mar1970
|
|
|
|
| Re: The ICL 2900 [message #338200 is a reply to message #338091] |
Thu, 23 February 2017 12:16 |
|
|
Originally posted by: Lawrence Statton NK1G
Andrew Swallow <am.swallow@btinternet.com> writes:
> On 21/02/2017 16:40, Lawrence Statton NK1G wrote:
>> Morten Reistad <first@last.name.invalid> writes:
>>> You know, *n*x has a file/user protection system that is similar
>>> to tops10/20.
>>>
>>> If you are feeling nervous about destroying stuff, make another
>>> user that has read/excute only to the files, everywhere (except
>>> /tmp; which is normally cleared on every boot).
>>>
>>> -- mrr
>>
>> I was about to respond to another post upthread, but you touched on the
>> topic in a way that gives a better opening.
>>
>> In Unix a non-privileged user can do nothing but destroy his own
>> personal environment. The worst that can happen is that you need a
>> qualified privileged user to restore it.
>>
>> That many people mistakenly believe themselves to be the Qualified
>> Privileged User is their own problem ;)
>>
>
> No but Unix beginners having to be a Qualified Privileged User is a
> major design problem.
I never said they had to *BE* one ... I said they had to *HAVE* one.
Like learning pretty much *anything* in life, it's helpful to have
someone who already knows how to do it to guide you.
In that respect, Unix is entirely unlike every other operating system in
the world. If you don't know what you're doing, and you carelessly
scrag critical system infrastructure, you're hosed. Unix. VMS. CP/M.
Anything.
|
|
|
|
| Re: The ICL 2900 [message #338201 is a reply to message #338189] |
Thu, 23 February 2017 12:21 |
|
|
Originally posted by: Lawrence Statton NK1G
Ahem A Rivet's Shot <steveo@eircom.net> writes:
> On Thu, 23 Feb 2017 14:23:20 -0000 (UTC)
> pechter@pechter.dyndns.org (William Pechter) wrote:
>
>> I have always wanted Pyramid conditional symbolic links so I could rig
>> up FreeBSD with (at least) two universes -- a Linux one and a FreeBSD
>> one.
>
> DragonFlyBSD has conditional symlinks, it's the only modern OS I
> know of that does.
Conditional on what? I'm curious.
|
|
|
|
| Re: The ICL 2900 [message #338205 is a reply to message #338137] |
Thu, 23 February 2017 12:28 |
|
|
Originally posted by: Lawrence Statton NK1G
pechter@pechter.dyndns.org (William Pechter) writes:
> In article <PM0005491EEE3D3A2F@aca480e6.ipt.aol.com>,
> jmfbahciv <See.above@aol.com> wrote:
>> Ahem A Rivet's Shot wrote:
>>> On 21 Feb 2017 14:38:59 GMT
>>> Huge <Huge@nowhere.much.invalid> wrote:
>>>
>>>> On 2017-02-21, jmfbahciv <See.above@aol.com> wrote:
>>>>
>>>> > Unix's user interface is dangerous
>>>> > because I don't know what I'm doing
>>>>
>>>> And that's *Unix*'s problem?
>>>
>>> "If you decide to shoot yourself in the foot it is the job of the
>>> operating system to ensure reliable delivery of the bullet." - I wish I knew
>>> who to attribute that gem to.
>>>
>>> Barb is right though, the standard unix user interface is not
>>> designed to provide a safe learning environment for the user. It is pretty
>>> good at protecting the OS and other users from the consequences of this
>>> except when you are simultaneously learning to use it and to admin it and
>>> thus making misteaks as root.
>>>
>> But that's the bias of Unix. It was developed so that its users
>> could go straight into user-flavored tasks such as editing or data
>> entry. It left the usual sysadmin tasks to the gurus.
>>
>> /BAH
>
> And TOPS10 and TOPS20 and VAX/VMS all had operators and Systems Managers
> to do it. Unless you had a Vax in your house. I know one programmer
> who had me install an 11/750 in his nice suburban colonial living room.
> Believe me. I'm sure it screwed his neighbor's TV and Radio reception.
>
> Bill
> --
Did you know Jim WB6NIL when he lived in wakefield (later Stoneham)? I
hung out with him for a year learning VAX "stuff" and we had an 11/730
in the kitchen for development (1988).
|
|
|
|
| Re: The ICL 2900 [message #338206 is a reply to message #338200] |
Thu, 23 February 2017 12:35 |
|
|
Originally posted by: Lawrence Statton NK1G
Lawrence Statton NK1G <lawrence@senguio.mx> writes:
> Andrew Swallow <am.swallow@btinternet.com> writes:
>
>> On 21/02/2017 16:40, Lawrence Statton NK1G wrote:
>>> Morten Reistad <first@last.name.invalid> writes:
>>>> You know, *n*x has a file/user protection system that is similar
>>>> to tops10/20.
>>>>
>>>> If you are feeling nervous about destroying stuff, make another
>>>> user that has read/excute only to the files, everywhere (except
>>>> /tmp; which is normally cleared on every boot).
>>>>
>>>> -- mrr
>>>
>>> I was about to respond to another post upthread, but you touched on the
>>> topic in a way that gives a better opening.
>>>
>>> In Unix a non-privileged user can do nothing but destroy his own
>>> personal environment. The worst that can happen is that you need a
>>> qualified privileged user to restore it.
>>>
>>> That many people mistakenly believe themselves to be the Qualified
>>> Privileged User is their own problem ;)
>>>
>>
>> No but Unix beginners having to be a Qualified Privileged User is a
>> major design problem.
>
> I never said they had to *BE* one ... I said they had to *HAVE* one.
>
> Like learning pretty much *anything* in life, it's helpful to have
> someone who already knows how to do it to guide you.
>
> In that respect, Unix is entirely unlike every other operating system in
> the world. If you don't know what you're doing, and you carelessly
> scrag critical system infrastructure, you're hosed. Unix. VMS. CP/M.
> Anything.
Dammit -- I was trying to channel Douglas Adamas and say "not unlike"
but I flubbed it.
|
|
|
|
| Re: The ICL 2900 [message #338207 is a reply to message #338200] |
Thu, 23 February 2017 12:41 |
Quadibloc
Messages: 4399
Registered: June 2012
Karma: 0
|
Senior Member |
|
|
On Thursday, February 23, 2017 at 10:16:05 AM UTC-7, Lawrence Statton NK1G wrote:
> In that respect, Unix is entirely unlike every other operating system in
> the world. If you don't know what you're doing, and you carelessly
> scrag critical system infrastructure, you're hosed. Unix. VMS. CP/M.
> Anything.
I presume when you said "unlike", you were being facetious.
However, while it's certainly true you can delete critical system files in
MS-DOS or whatever, it _is_ a bit easier to cause a disaster in Unix. The short
command names, the fact that the training wheels are off by default, and a few
other things contribute to that.
John Savard
|
|
|
|
| Re: The ICL 2900 Buying a computer in the 1960s [message #338208 is a reply to message #338177] |
Thu, 23 February 2017 13:08 |
Charlie Gibbs
Messages: 5313
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On 2017-02-23, jmfbahciv <See.above@aol.com> wrote:
> Charlie Gibbs wrote:
>
>> At a PPOE, when we got an early batch of personal computers,
>> it was arranged that the managers would get the most powerful
>> machines with the biggest screens - even though they almost
>> never used them. Meanwhile, our poor data entry clerk was
>> stuck squinting all day at a fuzzy 14-inch screen.
>
> So why didn't someone go toe to toe with the manager and retrain
> him? That's what would have happened at DEC.
This all happened just after a management invasion. The new management
team preached the values of teamwork while systematically destroying
the existing teams that had kept the place going for 10 years.
Due to the way she was treated, said data entry clerk several times
stormed out of the office in tears. One day she never came back.
Many of us - not just in the computer department but throughout the
organization - followed.
--
/~\ cgibbs@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
|
|
|
|
| Re: The ICL 2900 Buying a computer in the 1960s [message #338209 is a reply to message #338152] |
Thu, 23 February 2017 13:08 |
Charlie Gibbs
Messages: 5313
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On 2017-02-23, Mike Spencer <mds@bogus.nodomain.nowhere> wrote:
> hancock4@bbs.cpcn.com writes:
>
>> Back in the 1960s, suburbanites all rushed out to buy expensive
>> "hi fi" audio systems, including paying someone to properly install
>> the system and speakers for proper effect. They'd buy demo
>> records and drag in the neighbors to demonstrate such sounds as
>> a 707 taking off in their living room. Alan King wrote about this
>> in his spoof about the suburbs, but I remember people doing this.
>>
>> Probably 90% of those people couldn't distinguish true hi-fi sound
>> from a cheap table radio, but as you said, they wanted the bragging
>> rights of a fancy system. Indeed, when their kids went off to
>> college, the kids also like to brag about their audio systems
>> with huge speakers and high output amps and pre-amps. I remember
>> someone buying the cartridge for their phonograph tone-arm and
>> the cartridge alone was extremely expensive.
>
> In the early 60s, while still in school, I shared an appartment with a
> guy who was inordinately proud of his stereo hi-fi system --
> ca. $1,500 in 60s dollars -- with big fancy speaker cabinets. But it
> didn't bother him that he had the speakers wired wrong. It was fine
> for the 707 takeoff demos but the violins were on the wrong side in
> orchestral music, albeit very precisely reproduced.
My pet peeve are the many commercial establishments who want to play
music in the store. So they put a speaker at one end, another speaker
at the other end, look at the back of the amp, and say, "Hey, there
are two speaker terminals here - let's hook one up to each." So at
one end of the store you hear one half of the music, and at the other
end you hear the other half. Grrr...
--
/~\ cgibbs@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
|
|
|
|
| Re: The ICL 2900 [message #338210 is a reply to message #338139] |
Thu, 23 February 2017 13:08 |
Charlie Gibbs
Messages: 5313
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On 2017-02-23, JimP <solosam90@gmail.com> wrote:
> On Wed, 22 Feb 2017 22:40:33 +0000, Ahem A Rivet's Shot
> <steveo@eircom.net> wrote:
>
>> On Wed, 22 Feb 2017 21:28:25 +0000
>> Andrew Swallow <am.swallow@btinternet.com> wrote:
>>
>>> No but Unix beginners having to be a Qualified Privileged User is a
>>
>> Windows beginners have to be a QPU, so do Mac beginners and so did
>> TRS-80 beginners and Apple ][ beginners and every other type of personal
>> computer beginner since the things first appeared.
Macs and Windows boxes have the OS pre-installed. Few of the people who
go on about how Windows is so easy to use have ever witnessed an installation
from scratch, and don't know about the gotchas.
>>> major design problem.
>>
>> So solve it! Blue sky anything goes (except ever requiring a
>> competent sysadmin) - how do you create an OS so that a complete beginner
>> with average or lower intelligence can use it safely and do whatever
>> administrative tasks may be required without being able to damage it.
>>
>> I think it is impossible unless you start with a read only OS that
>> requires no maintenance, has every application you need pre-installed and
>> has an infallible automatic update system that never causes any problems.
Like your typical modern Mac or Windows box? They work (sort of) well -
as long as you stay in the sandbox.
>> Now can I tell you about this wonderful investment opportunity ...
>
> Dilbert cartoon where he and Wally are told to go look at the new
> computer as sales is having some sort of problem with it.
>
> The monitor shows a large button. The user presses the button on the
> screen and the sales people aren't sure what happens after that, but
> it is out of their ability to understand.
>
> Dilbert and Wally are not happy with the new computer.
Let me guess. The button is labeled DWIM (Do What I Mean).
For those who suggest training, remember the typical user's attitude:
<whine>
"I don't want to know how it works. I want to know which button to push!"
</whine>
--
/~\ cgibbs@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
|
|
|
|
| Re: The ICL 2900 [message #338211 is a reply to message #338206] |
Thu, 23 February 2017 13:11 |
Charlie Gibbs
Messages: 5313
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On 2017-02-23, Lawrence Statton NK1G <lawrence@senguio.mx> wrote:
> Lawrence Statton NK1G <lawrence@senguio.mx> writes:
>
>> In that respect, Unix is entirely unlike every other operating system in
>> the world. If you don't know what you're doing, and you carelessly
>> scrag critical system infrastructure, you're hosed. Unix. VMS. CP/M.
>> Anything.
>
> Dammit -- I was trying to channel Douglas Adamas and say "not unlike"
> but I flubbed it.
Yes, I suspected you were thinking about the machine that produced a
beverage that was almost but not exactly unlike tea. I mention that
one myself from time to time (and pray that if I trip over the syntax
I catch it in proffredding).
--
/~\ cgibbs@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
|
|
|
|
| Re: The ICL 2900 [message #338212 is a reply to message #338210] |
Thu, 23 February 2017 13:13 |
Quadibloc
Messages: 4399
Registered: June 2012
Karma: 0
|
Senior Member |
|
|
On Thursday, February 23, 2017 at 11:05:18 AM UTC-7, Charlie Gibbs wrote:
> For those who suggest training, remember the typical user's attitude:
> <whine>
> "I don't want to know how it works. I want to know which button to push!"
> </whine>
That attitude makes sense if you have one button to push, once, right now.
But if you intend to keep using the computer for any length of time, how would
you *remember* which button to push for what thing without understanding how it
works?
Of course, not *everyone* got good marks in math and science, but was hopeless
at learning French or Spanish in school, so maybe some people's minds work
differently...
John Savard
|
|
|
|
| Re: The ICL 2900 [message #338214 is a reply to message #338210] |
Thu, 23 February 2017 14:53 |
Mike Spencer
Messages: 997
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
Charlie Gibbs <cgibbs@kltpzyxm.invalid> writes:
> On 2017-02-23, JimP <solosam90@gmail.com> wrote:
>
>> Dilbert cartoon where he and Wally are told to go look at the new
>> computer as sales is having some sort of problem with it.
>>
>> The monitor shows a large button. The user presses the button on the
>> screen and the sales people aren't sure what happens after that, but
>> it is out of their ability to understand.
>>
>> Dilbert and Wally are not happy with the new computer.
Circa '93, I was doing a litle contract to make graphics for some
eduware and was hacking round with the unfamiliar API. I had made a
large, screen-filling red button that would pop up and down when
clicked. Fiddling with it, I realized there were some anomalous
sounds. Discoverd that a small crowd of onlookers had accumulated
just outside the door, trying not to giggle audibly.
The button was unlabeled in the UP position but displayed PANIC! in
large letters when pressed.
> Let me guess. The button is labeled DWIM (Do What I Mean).
Mark II, if I ever get a round tuit, will have that on the un-pressed
button but will, of course, still result in PANIC when pressed.
--
Mike Spencer Nova Scotia, Canada
|
|
|
|
| Re: The ICL 2900 [message #338215 is a reply to message #338201] |
Thu, 23 February 2017 14:32 |
Ahem A Rivet's Shot
Messages: 4843
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On Thu, 23 Feb 2017 11:21:30 -0600
Lawrence Statton NK1G <lawrence@senguio.mx> wrote:
> Ahem A Rivet's Shot <steveo@eircom.net> writes:
>
>> On Thu, 23 Feb 2017 14:23:20 -0000 (UTC)
>> pechter@pechter.dyndns.org (William Pechter) wrote:
>>
>>> I have always wanted Pyramid conditional symbolic links so I could rig
>>> up FreeBSD with (at least) two universes -- a Linux one and a FreeBSD
>>> one.
>>
>> DragonFlyBSD has conditional symlinks, it's the only modern OS I
>> know of that does.
>
> Conditional on what? I'm curious.
They embed references to variables that can be set with the varsym
utility.
https://leaf.dragonflybsd.org/cgi/web-man?command=varsym& ;section=ANY
These and process checkpoints are two of the neatest features in
DragonFlyBSD.
--
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/
|
|
|
|
| Re: The ICL 2900 [message #338216 is a reply to message #338207] |
Thu, 23 February 2017 14:39 |
Ahem A Rivet's Shot
Messages: 4843
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
On Thu, 23 Feb 2017 09:41:52 -0800 (PST)
Quadibloc <jsavard@ecn.ab.ca> wrote:
> On Thursday, February 23, 2017 at 10:16:05 AM UTC-7, Lawrence Statton
> NK1G wrote:
>
>> In that respect, Unix is entirely unlike every other operating system in
>> the world. If you don't know what you're doing, and you carelessly
>> scrag critical system infrastructure, you're hosed. Unix. VMS. CP/M.
>> Anything.
>
> I presume when you said "unlike", you were being facetious.
>
> However, while it's certainly true you can delete critical system files
> in MS-DOS or whatever, it _is_ a bit easier to cause a disaster in Unix.
> The short command names, the fact that the training wheels are off by
> default, and a few other things contribute to that.
I would be disinclined to trust a sysadmin without at least one war
story of accidentally induced disaster - or one with too many of them. An
old engineer once told me "experience gained varies directly with cost of
equipment ruined", he wouldn't trust an engineer who hadn't managed to
break something yet.
--
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/
|
|
|
|
| Re: The ICL 2900 [message #338217 is a reply to message #338187] |
Thu, 23 February 2017 15:10 |
Anne & Lynn Wheel
Messages: 3156
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
pechter@pechter.dyndns.org (William Pechter) writes:
> (Youtube just made me bite my tongue. Damn BSDNow podcast just
> mentioned porting VMS from ia64 to amd64. They were kind of joking
> about it. Too bad some folks never had the exposure to the slick
> things that exist under VMS including real clusters.)
part of doing HA/CMP ... recent posts
http://www.garlic.com/~lynn/2017.html#27 History of Mainframe Cloud
http://www.garlic.com/~lynn/2017.html#58 The ICL 2900
http://www.garlic.com/~lynn/2017.html#85 The ICL 2900
http://www.garlic.com/~lynn/2017b.html#73 The ICL 2900
http://www.garlic.com/~lynn/2017b.html#75 The ICL 2900
was cluster commercial scaleup ... working with RDBMS vendors that had
both vax cluster & unix support in the same sort base. To make things
simpler, I did cluster distributed lock manager that supported the
vax/cluster DLM API. However, the RDBMS vendors also had a list of dozen
or so things in vax/cluster that could have been done better
.... included fall-over recovery elapsed time. I used that knowledge as
well as having worked on mainframe clusters (aka "loosely-coupled
multiprocessing") back to the mid-70s and having been involved in the
orginal relational/sql implementation
http://www.garlic.com/~lynn/submain.html#systemr
ha/cmp posts
http://www.garlic.com/~lynn/subtopic.html#hacmp
other trivia: my wife (must have been in kindergarten) had been in the
gburg JES group and POK executive con'ed her into coming to POK to be in
charge of "loosely-coupled" architecture. While there she did
peer-coupled shared-data architecture
http://www.garlic.com/~lynn/submain.html#sharedata
she didn't remain long, in part because of little real uptake (except
for IMS hot-standby until SYSPLEX and Parallel SYSPLEX) and because the
communication group was constantly trying to force her into using
SNA/VTAM for loosely-coupled operation.
past posts mentioning distributed lock manager
http://www.garlic.com/~lynn/2000.html#64 distributed locking patents
http://www.garlic.com/~lynn/2001.html#40 Disk drive behavior
http://www.garlic.com/~lynn/2001c.html#66 KI-10 vs. IBM at Rutgers
http://www.garlic.com/~lynn/2001e.html#2 Block oriented I/O over IP
http://www.garlic.com/~lynn/2001j.html#47 OT - Internet Explorer V6.0
http://www.garlic.com/~lynn/2001k.html#5 OT - Internet Explorer V6.0
http://www.garlic.com/~lynn/2002e.html#67 Blade architectures
http://www.garlic.com/~lynn/2002f.html#1 Blade architectures
http://www.garlic.com/~lynn/2002k.html#8 Avoiding JCL Space Abends
http://www.garlic.com/~lynn/2003i.html#70 A few Z990 Gee-Wiz stats
http://www.garlic.com/~lynn/2004i.html#1 Hard disk architecture: are outer cylinders still faster than inner cylinders?
http://www.garlic.com/~lynn/2004i.html#2 New Method for Authenticated Public Key Exchange without Digital Certificates
http://www.garlic.com/~lynn/2004m.html#0 Specifying all biz rules in relational data
http://www.garlic.com/~lynn/2004m.html#5 Tera
http://www.garlic.com/~lynn/2004q.html#10 [Lit.] Buffer overruns
http://www.garlic.com/~lynn/2004q.html#70 CAS and LL/SC
http://www.garlic.com/~lynn/2005.html#40 clusters vs shared-memory (was: Re: CAS and LL/SC (was Re: High Level Assembler for MVS & VM & VSE))
http://www.garlic.com/~lynn/2005.html#55 Foreign key in Oracle Sql
http://www.garlic.com/~lynn/2005h.html#26 Crash detection by OS
http://www.garlic.com/~lynn/2005i.html#42 Development as Configuration
http://www.garlic.com/~lynn/2006c.html#8 IBM 610 workstation computer
http://www.garlic.com/~lynn/2006c.html#41 IBM 610 workstation computer
http://www.garlic.com/~lynn/2006j.html#20 virtual memory
http://www.garlic.com/~lynn/2006o.html#32 When Does Folklore Begin???
http://www.garlic.com/~lynn/2006o.html#62 Greatest Software, System R
http://www.garlic.com/~lynn/2006x.html#3 Why so little parallelism?
http://www.garlic.com/~lynn/2007c.html#42 Keep VM 24X7 365 days
http://www.garlic.com/~lynn/2007i.html#50 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007i.html#61 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007l.html#19 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007l.html#24 Is Parallel Programming Just Too Hard?
http://www.garlic.com/~lynn/2007m.html#55 Capacity and Relational Database
http://www.garlic.com/~lynn/2007n.html#49 VLIW pre-history
http://www.garlic.com/~lynn/2007q.html#33 Google And IBM Take Aim At Shortage Of Distributed Computing Skills
http://www.garlic.com/~lynn/2007s.html#46 "Server" processors for numbercrunching?
http://www.garlic.com/~lynn/2007v.html#42 Newbie question about db normalization theory: redundant keys OK?
http://www.garlic.com/~lynn/2007v.html#43 distributed lock manager
http://www.garlic.com/~lynn/2008b.html#69 How does ATTACH pass address of ECB to child?
http://www.garlic.com/~lynn/2008c.html#81 Random thoughts
http://www.garlic.com/~lynn/2008d.html#70 Time to rewrite DBMS, says Ingres founder
http://www.garlic.com/~lynn/2008g.html#56 performance of hardware dynamic scheduling
http://www.garlic.com/~lynn/2008h.html#91 Microsoft versus Digital Equipment Corporation
http://www.garlic.com/~lynn/2008i.html#18 Microsoft versus Digital Equipment Corporation
http://www.garlic.com/~lynn/2008k.html#63 Intel: an expensive many-core future is ahead of us
http://www.garlic.com/~lynn/2008r.html#71 Curiousity: largest parallel sysplex around?
http://www.garlic.com/~lynn/2009.html#3 Is SUN going to become x86'ed ??
http://www.garlic.com/~lynn/2009b.html#40 "Larrabee" GPU design question
http://www.garlic.com/~lynn/2009h.html#26 Natural keys vs Aritficial Keys
http://www.garlic.com/~lynn/2009k.html#36 Ingres claims massive database performance boost
http://www.garlic.com/~lynn/2009k.html#67 Disksize history question
http://www.garlic.com/~lynn/2009m.html#39 ACP, One of the Oldest Open Source Apps
http://www.garlic.com/~lynn/2009m.html#84 A Faster Way to the Cloud
http://www.garlic.com/~lynn/2009o.html#57 U.S. begins inquiry of IBM in mainframe market
http://www.garlic.com/~lynn/2010b.html#32 Happy DEC-10 Day
http://www.garlic.com/~lynn/2010b.html#66 SYSENTER/SYSEXIT_vs._SYSCALL/SYSRET
http://www.garlic.com/~lynn/2010k.html#54 Unix systems and Serialization mechanism
http://www.garlic.com/~lynn/2010l.html#14 Age
http://www.garlic.com/~lynn/2010n.html#82 Hashing for DISTINCT or GROUP BY in SQL
http://www.garlic.com/~lynn/2011.html#23 zLinux OR Linux on zEnterprise Blade Extension???
http://www.garlic.com/~lynn/2011f.html#8 New job for mainframes: Cloud platform
http://www.garlic.com/~lynn/2012d.html#28 NASA unplugs their last mainframe
http://www.garlic.com/~lynn/2012n.html#53 history of Programming language and CPU in relation to each other
http://www.garlic.com/~lynn/2013m.html#86 'Free Unix!': The world-changing proclamation made 30 yearsagotoday
http://www.garlic.com/~lynn/2013m.html#87 'Free Unix!': The world-changing proclamation made 30 yearsagotoday
http://www.garlic.com/~lynn/2013n.html#19 z/OS is antique WAS: Aging Sysprogs = Aging Farmers
http://www.garlic.com/~lynn/2013o.html#44 the suckage of MS-DOS, was Re: 'Free Unix!
http://www.garlic.com/~lynn/2014.html#73 the suckage of MS-DOS, was Re: 'Free Unix!
http://www.garlic.com/~lynn/2014k.html#40 How Larry Ellison Became The Fifth Richest Man In The World By Using IBM's Idea
http://www.garlic.com/~lynn/2014m.html#142 IBM Continues To Crumble
--
virtualization experience starting Jan1968, online at home since Mar1970
|
|
|
|
| Re: The ICL 2900 [message #338218 is a reply to message #338182] |
Thu, 23 February 2017 15:18 |
Jon Elson
Messages: 646
Registered: April 2013
Karma: 0
|
Senior Member |
|
|
William Pechter wrote:
>
> As long as the toilets don't back up. I can take the clothes out
> to a laundromat if needed.
Oh, boy! When they hook the IoT to the toilets, then the hackers will
REALLY start to have fun (= create chaos). Oh, I just CAN'T wait for that!
Jon
|
|
|
|
| Re: The ICL 2900 [message #338219 is a reply to message #338108] |
Thu, 23 February 2017 15:26 |
hancock4
Messages: 6746
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
On Wednesday, February 22, 2017 at 6:59:15 PM UTC-5, Jon Elson wrote:
> Yes, we had it here. We had a pair of 370/145s, and each machine could
> support FOUR PROFS users! More than 4, and then the system got REALLY slow.
> Well, 145's were not speed demons, but that seems awfully mediocre.
I don't know what the machine was that supported our PROFS, but
PROFS was not resource intensive. All it handled was text. It
ran as part of CMS/VM.
Our S/360-40 handled four terminals under a "baby" CICS (MTCS?),
so I would guess a 145 could handle more.
|
|
|
|
| Re: The ICL 2900 [message #338221 is a reply to message #338114] |
Thu, 23 February 2017 15:30 |
hancock4
Messages: 6746
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
On Wednesday, February 22, 2017 at 8:14:59 PM UTC-5, Peter Flass wrote:
> What the heck were they doing? IME most PROFS users left the screen at the
> calendar unless they were doing email. You could support lots of users
> doing nothing on almost any system. We had PROFS/OV, but I think I stuck
> to simpler systems for mail, although the memory is all gone now.
We had PROFS first, then later got more modern email. Personally,
I was very happy with PROFS, despite it being text-only. For one
thing, it was very reliable, almost never crashing*. Secondly,
it was immune to viruses and other crap that more modern email
users suffered with.
I really liked PROFS' reminder feature, very easy to use, very handy.
I never found an adequate substitute after they took PROFS away.
* If someone sent you a very large email, say with an attachment,
it could overload your in-basket and lock up access. If that
happened, we had to telephone the administrator to go in and
clean up the mess.
PROFS ran under VM/CMS, and some maintenance tasks were easily
performed by those who knew VM. Those of us on MVS ran into
some challenges with that.
|
|
|
|
| Re: The ICL 2900 [message #338222 is a reply to message #338129] |
Thu, 23 February 2017 15:51 |
Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
William Pechter <pechter@pechter.dyndns.org> wrote:
>
> One recommendation... don't worry. Trashing a system and fixing it
> is the best way to learn. Did it a hundred times and that's why I'm
> much quicker fixing problems.
>
That's wonderful, unless you're trying to get some work done.
> Just like driving in the snow becomes easier with practice, system recovery
> gets easier. I do about 5-10 linux installs a week and have done them in my
> sleep these days.
>
Installs are simple, and assuming user data is backed up it is easily
restored. I have found that identifying Ll the apps I need and reinstalling
any that aren't installed by default, plus customizing desktops, etc. take
quite a bit of additional time.
--
Pete
|
|
|
|
| Re: The ICL 2900 [message #338223 is a reply to message #338134] |
Thu, 23 February 2017 15:51 |
Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
William Pechter <pechter@pechter.dyndns.org> wrote:
> In article <PM0005491EF13B7A70@aca480e6.ipt.aol.com>,
> jmfbahciv <See.above@aol.com> wrote:
>> Huge wrote:
>>> On 2017-02-21, Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>>>> On 21 Feb 2017 14:38:59 GMT
>>>> Huge <Huge@nowhere.much.invalid> wrote:
>>>>
>>>> > On 2017-02-21, jmfbahciv <See.above@aol.com> wrote:
>>>> >
>>>> >> Unix's user interface is dangerous
>>>> >> because I don't know what I'm doing
>>>> >
>>>> > And that's *Unix*'s problem?
>>>>
>>>> "If you decide to shoot yourself in the foot it is the job of the
>>>> operating system to ensure reliable delivery of the bullet." - I wish I
>> knew
>>>> who to attribute that gem to.
>>>>
>>>> Barb is right though, the standard unix user interface is not
>>>> designed to provide a safe learning environment for the user.
>>>
>>> I have seen Unix described as being "by computer scientists, for
>>> computer scientists." That suits me fine.
>>
>> <ahem> That's not complimentary.
>>
>> /BAH
>
> I used to think of it "By Unix bigots ... for Unix bigots" back in my
> DEC VMS days... but it grows on you... Kind of like a fungus. 8-)
You can get along OK with any OS that doesn't get in your way. If you use
it enough you get used to it. That doesn't make it optimal.
--
Pete
|
|
|
|
| Re: The ICL 2900 [message #338224 is a reply to message #338216] |
Thu, 23 February 2017 16:23 |
|
|
Originally posted by: Lawrence Statton NK1G
Ahem A Rivet's Shot <steveo@eircom.net> writes:
> I would be disinclined to trust a sysadmin without at least one war
> story of accidentally induced disaster - or one with too many of them. An
> old engineer once told me "experience gained varies directly with cost of
> equipment ruined", he wouldn't trust an engineer who hadn't managed to
> break something yet.
nearhost% ssh machine.in.other.country.net
# I need to change some of the network stuff ... I'll just ...
machine% sudo ifconfig ie0 down
[SHIT!]
[dial, dial, dial .... ]
Hi -- ummm ... this is Lawrence in DevOps in Mexico City ... can you go
to the console of the machine in Rack 14, slot 12 and bring the network
back up? I seem to have forgotten where I am. KTHXBYE
Fortunately, decades later, remote connections to virtual consoles are
commonplace, and those kinds of mistakes aren't as embarassing as they
once were.
Another of my "oh, shit" moments was very carefully loading the
configuration for the Mtn. View router of an ISP into the Oakland
router. That involved train, BART, and a taxi to un-snarl. The result
of that was a nice ad-hoc tool to manage router configurations directly
out of revision control that checked ... something. Serial number? I
don't remember now, but it was "clever" and prevented anything but
advanced malfeasance.
|
|
|
|
| Re: The ICL 2900 [message #338225 is a reply to message #338155] |
Thu, 23 February 2017 16:24 |
Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
Ahem A Rivet's Shot <steveo@eircom.net> wrote:
> On Thu, 23 Feb 2017 02:50:08 -0000 (UTC)
> pechter@pechter.dyndns.org (William Pechter) wrote:
>
>> Used to open the class with the question of "What are the three most
>> important things that a sysadmin must know."
>>
>> 1. Backup
>> 2. Backup
>> 3. Backup
>
> You missed out test the backup.
>
Much easier now than in the old days. Now you can define to test the
restore, and disk is cheap enough that you can have plenty. It's a bit
scary trying to test the restore of a production system .
--
Pete
|
|
|
|
| Re: The ICL 2900 [message #338226 is a reply to message #338171] |
Thu, 23 February 2017 16:24 |
Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
jmfbahciv <See.above@aol.com> wrote:
> Ahem A Rivet's Shot wrote:
>> On Thu, 23 Feb 2017 03:00:04 -0000 (UTC)
>> pechter@pechter.dyndns.org (William Pechter) wrote:
>>
>>> In article <20170222224033.fb3f9ceba0e8431876d1c2df@eircom.net>,
>>> Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>>>> On Wed, 22 Feb 2017 21:28:25 +0000
>>>> Andrew Swallow <am.swallow@btinternet.com> wrote:
>>>>
>>>> > No but Unix beginners having to be a Qualified Privileged User is a
>>>>
>>>> Windows beginners have to be a QPU, so do Mac beginners and so
>>>> did
>>>> TRS-80 beginners and Apple ][ beginners and every other type of personal
>>>> computer beginner since the things first appeared.
>>>>
>>>> > major design problem.
>>>>
>>>> So solve it! Blue sky anything goes (except ever requiring a
>>>> competent sysadmin) - how do you create an OS so that a complete beginner
>>>> with average or lower intelligence can use it safely and do whatever
>>>> administrative tasks may be required without being able to damage it.
>>>
>>> Old school answer from a former training sysadmin guy.
>>> Train them.
>>
>> This is a good answer - but the old saw about leading horses to
>> water applies.
>>
> I'm quite proficient learning new OSes. There are no books which
> train new system owners. Part of the problem is that there is no
> common user interface because of the shell concept. I'd like to
> see the command language which is underneath those shells.
That is the command language. The "commands" are the program names. The
shells handle things like redirection and command history and retrieval,
and include rudimentary scripting facilities.
--
Pete
|
|
|
|
| Re: The ICL 2900 [message #338227 is a reply to message #338210] |
Thu, 23 February 2017 16:24 |
Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
> On 2017-02-23, JimP <solosam90@gmail.com> wrote:
>
>> On Wed, 22 Feb 2017 22:40:33 +0000, Ahem A Rivet's Shot
>> <steveo@eircom.net> wrote:
>>
>>> On Wed, 22 Feb 2017 21:28:25 +0000
>>> Andrew Swallow <am.swallow@btinternet.com> wrote:
>>>
>>>> No but Unix beginners having to be a Qualified Privileged User is a
>>>
>>> Windows beginners have to be a QPU, so do Mac beginners and so did
>>> TRS-80 beginners and Apple ][ beginners and every other type of personal
>>> computer beginner since the things first appeared.
>
> Macs and Windows boxes have the OS pre-installed. Few of the people who
> go on about how Windows is so easy to use have ever witnessed an installation
> from scratch, and don't know about the gotchas.
>
Modern OSs have a big problem that VMS, VM, etc. didn't have - hardware.
They were written to run on the vendor's hardware, and anything else was
expected to be compatible with it. If there was a problem it was the
hardware's fault (even when it wasn't). Nowadays OSs are expected to run on
a lot of miscellaneous hardware. If windoze or Linux has a problem with my
display it's the OSs fault. Hardware needs to become more standardized. IBM
and Apple had the right idea with CHRP - define the standard and expect
that everyone will conform. I don't see why there have to be so many
flavors of network or graphics cards.
--
Pete
|
|
|
|
| Re: The ICL 2900 [message #338229 is a reply to message #338166] |
Thu, 23 February 2017 11:28 |
Morten Reistad
Messages: 2108
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
In article <PM00054932A0B00D6A@aca41e2b.ipt.aol.com>,
jmfbahciv <See.above@aol.com> wrote:
> capek@ieee.org <Peter Capek> wrote:
>> On Saturday, February 18, 2017 at 8:52:58 AM UTC-5, jmfbahciv wrote:
>>> Questor wrote:
>>>> On 15 Feb 2017 16:48:56 GMT, Huge <Huge@nowhere.much.invalid> wrote:
>>>> >On 2017-02-15, Scott Lurndal <scott@slp53.sl.home> wrote:
>>>> >> jmfbahciv <See.above@aol.com> writes:
>>>> >
>>>> >[20 lines snipped]
>>>> >>>
>>>> >>>This kind of computer usage was not the norm.
>>>> >>
>>>> >> Huh? It was quite the norm from mainframes to departmental servers.
>>>> >
>>>> >Babs has a remarkable parochial view of how the computer industry works.
>>>>
>>>> Ms. Huizenga has not been entirely candid about her work at DEC. She has
>>> used
>>>> circumlocution and misdirection to obscure her true role. Most of the
> time
>>> when
>>>> she talks about what "we" did, it should be taken and read as what "they"
>>> did.
>>>> There's a reason for the obvious gaps in her technical knowledge and her
>>> focus
>>>> on process.
>>>>
>>>> With regard to "parochial," literally her entire career was spent working
> in
>>> the
>>>> TOPS-10 group. Just about everything she knows about other computers and
>>>> operating systems is through the lens of how they relate to PDP-10s and
>>> TOPS-10.
>>>> Her lack of formal training and breadth of experience shows up in not
> having
>>> the
>>>> abstractions and generalizations that would ease learning a new
> environment,
>>>> such as UNIX, Instead, she tries to put everything into a TOPS-10 box.
>>>>
>>> You don't know anything about what I did at DEC.
>>>
>>> /BAH
>>
>> I don't either. Why don't you tell us? I'm interested to know.
>
> Morten covered the work I was paid to do in another post. I also
> did a lot of work I wasn't paid to do.
>
> /BAH
I only have rudimentary information about what you did at DEC (and when you
did what) . I would welcome a brief autobiography.
-- mrr
|
|
|
|
| Re: The ICL 2900 [message #338230 is a reply to message #338171] |
Thu, 23 February 2017 11:30 |
Morten Reistad
Messages: 2108
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
In article <PM0005493284D694EC@aca41e2b.ipt.aol.com>,
jmfbahciv <See.above@aol.com> wrote:
> Ahem A Rivet's Shot wrote:
>> On Thu, 23 Feb 2017 03:00:04 -0000 (UTC)
>> pechter@pechter.dyndns.org (William Pechter) wrote:
>>
>>> In article <20170222224033.fb3f9ceba0e8431876d1c2df@eircom.net>,
>>> Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>>>> On Wed, 22 Feb 2017 21:28:25 +0000
>>>> Andrew Swallow <am.swallow@btinternet.com> wrote:
>>>>
>>>> > No but Unix beginners having to be a Qualified Privileged User is a
>>>>
>>>> Windows beginners have to be a QPU, so do Mac beginners and so
>>>> did
>>>> TRS-80 beginners and Apple ][ beginners and every other type of personal
>>>> computer beginner since the things first appeared.
>>>>
>>>> > major design problem.
>>>>
>>>> So solve it! Blue sky anything goes (except ever requiring a
>>>> competent sysadmin) - how do you create an OS so that a complete beginner
>>>> with average or lower intelligence can use it safely and do whatever
>>>> administrative tasks may be required without being able to damage it.
>>>
>>> Old school answer from a former training sysadmin guy.
>>> Train them.
>>
>> This is a good answer - but the old saw about leading horses to
>> water applies.
>>
> I'm quite proficient learning new OSes. There are no books which
> train new system owners. Part of the problem is that there is no
> common user interface because of the shell concept. I'd like to
> see the command language which is underneath those shells.
>
> /BAH
You mean the instruction set and the syscalls manual ?
Or do you want to look in "appendix C" ? (the C source code).
I would recommend starting with the bash source code if you do
such an endeavour. It is somewhat more structured and easy to
understand.
-- mrr
|
|
|
|
| Re: The ICL 2900 [message #338231 is a reply to message #338174] |
Thu, 23 February 2017 11:33 |
Morten Reistad
Messages: 2108
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
In article <PM0005493306541D4F@aca41e2b.ipt.aol.com>,
jmfbahciv <See.above@aol.com> wrote:
> William Pechter wrote:
>> In article <PM0005491E9D353FCC@aca480e6.ipt.aol.com>,
>> jmfbahciv <See.above@aol.com> wrote:
>>
>>> Aren't today's controllers a part of the disk drive? So there's no longer
>>> one controller for many drives.
>>>
>>> /BAH
>>
>> Nope... except for the older IDE drive where the controller was a pretty
>> small bus interface.
>>
>> There's still the need for the thing that connects the drive to the I/O bus.
>> Much of the logic now is kind of like the stuff DEC did for DSA (Digital
>> Storage Architecture).
>>
>> The disks are smart devices talking some kind of data protocol to the system
>> via a controller.
>>
>> The disks are addressed by logical blocks and no longer Cylinder, Head and
>> Sector. The physical geometry is no longer tied to the disk software
>> geometry.
>
> Somebody has to do the physcial-logical mapping. I figured it was part of
> the [what I would think of as] the disk drive.
This (and the badspot, block moving and renaming etc) is all done in the
firmware of the "disk". Actually, the IBMism "DASD" becomes more fitting as
time goes on.
>> (This has had effects on some of the low level format and partition layout
>> software...like the BSD Disklabels).
>>
>> This change, as far as I know, started back in the 80's with DEC RA drives
>> and DSA as well as the SASI/SCSI stuff which was also becoming a standard
>> at the same time.
>
> Yes to the DEC RAs. TW worte that code for the PDP-10s; it used a
> comm spec similar to DECnet's approach. We did not like them
> because the disks were no availabe for booting the system. Another
> design flaw. another design flaw was the number of ports; it was
> still 2.
Was the not-bootable thing a front end/firmware thing or was it
a feature in tops10/20?
-- mrr
|
|
|
|
| Re: The ICL 2900 [message #338232 is a reply to message #338225] |
Thu, 23 February 2017 17:15 |
Peter Flass
Messages: 8375
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
Peter Flass <peter_flass@yahoo.com> wrote:
> Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>> On Thu, 23 Feb 2017 02:50:08 -0000 (UTC)
>> pechter@pechter.dyndns.org (William Pechter) wrote:
>>
>>> Used to open the class with the question of "What are the three most
>>> important things that a sysadmin must know."
>>>
>>> 1. Backup
>>> 2. Backup
>>> 3. Backup
>>
>> You missed out test the backup.
>>
>
> Much easier now than in the old days. Now you can define to test the
Words missing -- "define a VM"
> restore, and disk is cheap enough that you can have plenty. It's a bit
> scary trying to test the restore of a production system .
>
--
Pete
|
|
|
|
| Re: The ICL 2900 [message #338233 is a reply to message #338224] |
Thu, 23 February 2017 17:23 |
Morten Reistad
Messages: 2108
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
In article <87wpcgk3ev.fsf@senguio.mx>,
Lawrence Statton NK1G <lawrence@senguio.mx> wrote:
> Ahem A Rivet's Shot <steveo@eircom.net> writes:
>> I would be disinclined to trust a sysadmin without at least one war
>> story of accidentally induced disaster - or one with too many of them. An
>> old engineer once told me "experience gained varies directly with cost of
>> equipment ruined", he wouldn't trust an engineer who hadn't managed to
>> break something yet.
>
> nearhost% ssh machine.in.other.country.net
> # I need to change some of the network stuff ... I'll just ...
> machine% sudo ifconfig ie0 down
>
> [SHIT!]
>
> [dial, dial, dial .... ]
>
> Hi -- ummm ... this is Lawrence in DevOps in Mexico City ... can you go
> to the console of the machine in Rack 14, slot 12 and bring the network
> back up? I seem to have forgotten where I am. KTHXBYE
>
> Fortunately, decades later, remote connections to virtual consoles are
> commonplace, and those kinds of mistakes aren't as embarassing as they
> once were.
That, and auditors.
At a PPOE we went through 4 mergers in pretty rapid order, and ended
up having to report all expenses and inventory to some gnomes in den Haag.
In late november one year one of them was at our doors, and insisted
on seeing a random sample of our hardware. Even if three pieces of them
were in _very_ remote locations, serving dialup customers from some
rural locations.
One of these were Lofoten. This was before Leikes airport, so the way
to get there was 2 hours by plane to Bodo, ferry across the fjord (3 hours)
and then a 2 hour drive. Then someone local had to show up and lock him
into the bunker where the equipment was located.
This particular day there was a hurricane arriving. There was no bus
service on the other side, and all taxies were booked for the next two days.
So he rented a car in Bodo, and signed the total disclaimer the ferry company
insisted on before being let on. The car was more or less totalled from salt overwash
(and would have gone into the waves if it was nit secured).
We managed to get the local maintainer to go to the ferry to pick him
up, drive him to the bunker, let him in; look around for about 30 seconds
while noting the serial number of the box, drive him back and let him
wait for 6 hours for the ferry.
The car was a total loss, and his waivers also released the insurance
companies from any damages. So he was out around $20k. He never talked
to us ever again.
>
> Another of my "oh, shit" moments was very carefully loading the
> configuration for the Mtn. View router of an ISP into the Oakland
> router. That involved train, BART, and a taxi to un-snarl. The result
> of that was a nice ad-hoc tool to manage router configurations directly
> out of revision control that checked ... something. Serial number? I
> don't remember now, but it was "clever" and prevented anything but
> advanced malfeasance.
>
-- mrr
|
|
|
|
| Re: The ICL 2900 [message #338234 is a reply to message #338218] |
Thu, 23 February 2017 17:02 |
Morten Reistad
Messages: 2108
Registered: December 2011
Karma: 0
|
Senior Member |
|
|
In article <lc-dnbUbwM8f3jLFnZ2dnUU7-Q-dnZ2d@giganews.com>,
Jon Elson <jmelson@wustl.edu> wrote:
> William Pechter wrote:
>
>
>>
>> As long as the toilets don't back up. I can take the clothes out
>> to a laundromat if needed.
> Oh, boy! When they hook the IoT to the toilets, then the hackers will
> REALLY start to have fun (= create chaos). Oh, I just CAN'T wait for that!
And just how do you think the sewer pumps in flatlands are managed?
-- mrr
|
|
|
|
| Re: The ICL 2900 [message #338241 is a reply to message #338183] |
Thu, 23 February 2017 19:27 |
|
|
Originally posted by: luserdroog
On Thursday, February 23, 2017 at 8:33:08 AM UTC-6, William Pechter wrote:
> In article <PM0005493284D694EC@aca41e2b.ipt.aol.com>,
> jmfbahciv <See.above@aol.com> wrote:
>
>> I'm quite proficient learning new OSes. There are no books which
>> train new system owners. Part of the problem is that there is no
>> common user interface because of the shell concept. I'd like to
>> see the command language which is underneath those shells.
>>
>> /BAH
>
>
> Actually, there are books. Written by professionals with PHd's and experience.
> One of whom set up the corporate training at a prior employer.
>
> PDF's were on the laptop I sent. They're often slightly out of date
> since Linux changes so quickly. But they do work well as a guide.
> I used them (earlier edition) back in '84 or so...
>
Without knowing which pdfs you mean, here's a short list of books
that should get a technical person up to speed with the basics of
Unix.
Kernighan & Pike, The UNIX Programming Environment.
Perhaps overlooked as /too obvious/, read just the first half.
The second half is an extended tutorial on yacc.
O'Reilly. Unix in a Nutshell.
Doesn't much matter which version/edition. I actually recommend
an old edition since the new Linux ones are just too crammed full
of stuff. The sheer size becomes overly intimidating.
For TMI:
Maurice Bach, Design of the Unix Operating System.
inodes and init and fork[1].
John Lions, Commentary on Unix Version 6.
Most people find the pdp11 assembler part harder than the C part.
YMMV.
--
[1] (o my!)
|
|
|
|
| Re: The ICL 2900 [message #338249 is a reply to message #338227] |
Thu, 23 February 2017 20:38 |
Richard Thiebaud
Messages: 222
Registered: May 2013
Karma: 0
|
Senior Member |
|
|
On 02/23/2017 04:24 PM, Peter Flass wrote:
> Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>> On 2017-02-23, JimP <solosam90@gmail.com> wrote:
>>
>>> On Wed, 22 Feb 2017 22:40:33 +0000, Ahem A Rivet's Shot
>>> <steveo@eircom.net> wrote:
>>>
>>>> On Wed, 22 Feb 2017 21:28:25 +0000
>>>> Andrew Swallow <am.swallow@btinternet.com> wrote:
>>>>
>>>> > No but Unix beginners having to be a Qualified Privileged User is a
>>>>
>>>> Windows beginners have to be a QPU, so do Mac beginners and so did
>>>> TRS-80 beginners and Apple ][ beginners and every other type of personal
>>>> computer beginner since the things first appeared.
>>
>> Macs and Windows boxes have the OS pre-installed. Few of the people who
>> go on about how Windows is so easy to use have ever witnessed an installation
>> from scratch, and don't know about the gotchas.
>>
>
> Modern OSs have a big problem that VMS, VM, etc. didn't have - hardware.
> They were written to run on the vendor's hardware, and anything else was
> expected to be compatible with it. If there was a problem it was the
> hardware's fault (even when it wasn't). Nowadays OSs are expected to run on
> a lot of miscellaneous hardware. If windoze or Linux has a problem with my
> display it's the OSs fault. Hardware needs to become more standardized. IBM
> and Apple had the right idea with CHRP - define the standard and expect
> that everyone will conform. I don't see why there have to be so many
> flavors of network or graphics cards.
>
Would you like only one flavor of food?
|
|
|
|
| Re: The ICL 2900 [message #338250 is a reply to message #338241] |
Thu, 23 February 2017 21:36 |
Michael Black
Messages: 2799
Registered: February 2012
Karma: 0
|
Senior Member |
|
|
On Thu, 23 Feb 2017, luserdroog wrote:
> On Thursday, February 23, 2017 at 8:33:08 AM UTC-6, William Pechter wrote:
>> In article <PM0005493284D694EC@aca41e2b.ipt.aol.com>,
>> jmfbahciv <See.above@aol.com> wrote:
>>
>>> I'm quite proficient learning new OSes. There are no books which
>>> train new system owners. Part of the problem is that there is no
>>> common user interface because of the shell concept. I'd like to
>>> see the command language which is underneath those shells.
>>>
>>> /BAH
>>
>>
>> Actually, there are books. Written by professionals with PHd's and experience.
>> One of whom set up the corporate training at a prior employer.
>>
>> PDF's were on the laptop I sent. They're often slightly out of date
>> since Linux changes so quickly. But they do work well as a guide.
>> I used them (earlier edition) back in '84 or so...
>>
>
> Without knowing which pdfs you mean, here's a short list of books
> that should get a technical person up to speed with the basics of
> Unix.
>
> Kernighan & Pike, The UNIX Programming Environment.
> Perhaps overlooked as /too obvious/, read just the first half.
> The second half is an extended tutorial on yacc.
>
> O'Reilly. Unix in a Nutshell.
> Doesn't much matter which version/edition. I actually recommend
> an old edition since the new Linux ones are just too crammed full
> of stuff. The sheer size becomes overly intimidating.
>
> For TMI:
>
> Maurice Bach, Design of the Unix Operating System.
> inodes and init and fork[1].
>
> John Lions, Commentary on Unix Version 6.
> Most people find the pdp11 assembler part harder than the C part.
> YMMV.
>
There was a period when there were books available for the Unix user.
Since they didn't have to spend time talking about installing Unix or
managing it, they could spend more time on helping the user to use it.
The period was of course before Linux, and when there were lots of
students with access to Unix machines, and of course, they had no access
to root.
I've found some at used book sales over the years, and while I have
foundsome neat new things, their benefit is about users.
When I installed Slackware, it was because I came across a copy of
"SLackware Linux for Dummies", with a torn cover and a clearance price.
It still had the two CDROMs in there, so I grabbed it, an easy way to get
it when I didn't have high speed internet. But most of the book wsa about
installing, and then a bit about being "system adminstrator" leaving
little space for actual users. In Unix days, the user didn't have to
fuss, since there was someone managing it. But with Linux, generally you
have to do both.
And buying endless used and clearance Linux books never revealed too much
more, since they generally were getting the reader to the point where they
had Linux installed.
Even now, the books about specific editors or Gimp or something else have
faded.
Michael
|
|
|
|
| Re: The ICL 2900 [message #338251 is a reply to message #338156] |
Thu, 23 February 2017 21:42 |
Bill Findlay
Messages: 286
Registered: January 2012
Karma: 0
|
Senior Member |
|
|
Ahem A Rivet's Shot <steveo@eircom.net> wrote:
> On 23 Feb 2017 00:53:08 GMT
> Bill Findlay <findlaybill@blueyonder.co.uk > wrote:
>
>> Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>
>>> But for home use there is one user and they have to do it all.
>>
>> Wrong, simply wrong, for macOS at least.
>
> <sigh> One human being has to do it all. One user physical not
> logical. No matter how inexperienced or how stupid that one person has to
> do the admin work. There is no trained sysadmin.
There is no need for a trained sysadmin.
</sigh>
--
Bill Findlay
|
|
|
|
Members
Search
Help
Register
Login
Home
