Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Two countries separated by a common language?

49 views
Skip to first unread message

Gareth's Downstairs Computer

unread,
May 7, 2017, 5:41:47 AM5/7/17
to
Brit designs had order codes, the Yanks, instruction sets.

Brit computers hesitated whilst the Yank ones did DMA.

Any others? ...

In 1971, the TAD instruction of the PDP8 was a Yank joke
unheard of in Brit, and now 46 years later the term, a tad,
is widely used in UK (and was used in a recent Brit TV program
set in the 1930s!)

ma...@mail.com

unread,
May 7, 2017, 6:07:42 AM5/7/17
to
The sad thing about British computer business was the the Lyons
tearooms control systems was way ahead of any such system anywhere, but
was not persued to logical conclusions.

And yes, there are many terms are used in historical dramas which
only came into use much later.. and americanisms used in dramas set in
the UK.


--
greymaus.ireland.ie
Just_Another_Grumpy_Old_Man

Ahem A Rivet's Shot

unread,
May 7, 2017, 6:29:03 AM5/7/17
to
On Sun, 7 May 2017 10:41:46 +0100
Gareth's Downstairs Computer
<headstone255.but.n...@yahoo.com> wrote:

> In 1971, the TAD instruction of the PDP8 was a Yank joke
> unheard of in Brit, and now 46 years later the term, a tad,
> is widely used in UK (and was used in a recent Brit TV program
> set in the 1930s!)

Well it certainly predates 1971 - according to the OED it
originates from the late 19th century as a term for a small child with the
modern usage dating from the 1940s. The OED speculates that it derives from
Tadpole while Miriam Webster opts for it deriving from toad - I'm inclined
to think the OED one more likely. I didn't hear it much until the late
1970s and I had the impression it was a mostly northern expression like
graidly, nesh and nowt.

--
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/

Bob Eager

unread,
May 7, 2017, 6:51:49 AM5/7/17
to
On Sun, 07 May 2017 10:07:39 +0000, mausg wrote:

> On 2017-05-07, Gareth's Downstairs Computer
> <headstone255.but.n...@yahoo.com> wrote:
>> Brit designs had order codes, the Yanks, instruction sets.
>>
>> Brit computers hesitated whilst the Yank ones did DMA.
>>
>> Any others? ...
>>
>> In 1971, the TAD instruction of the PDP8 was a Yank joke unheard of in
>> Brit, and now 46 years later the term, a tad, is widely used in UK (and
>> was used in a recent Brit TV program set in the 1930s!)
>>
>>
> The sad thing about British computer business was the the Lyons tearooms
> control systems was way ahead of any such system anywhere, but was not
> persued to logical conclusions.

I have one of the books on LEO, and it's very good.

I hope to attend this:

http://www.computerconservationsociety.org/lectures/2016-17/20170518.htm



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

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

Gareth's Downstairs Computer

unread,
May 7, 2017, 6:56:12 AM5/7/17
to
On 07/05/2017 11:07, ma...@mail.com wrote:
>
> The sad thing about British computer business was the the Lyons
> tearooms control systems was way ahead of any such system anywhere, but
> was not persued to logical conclusions.

For my first home computer project, 1973, I had two blocks of
core store (96 4K planes) from one of those LEOs. In the end, I only
kept 8, to give me 4K bytes, and I've now only 7 left in my mini
museum as an acquaintance wanted to experiment with little
ferrite rings.

In the end, I didn't use it because along came static RAM, and
with 4-off sample 256 by 1 bit chips and a bit of digital
jiggery pokery, I had 128 bytes of static RAM, and felt myself
well off even with that!



Gareth's Downstairs Computer

unread,
May 7, 2017, 7:07:37 AM5/7/17
to
On 07/05/2017 11:21, Ahem A Rivet's Shot wrote:
> On Sun, 7 May 2017 10:41:46 +0100
> Gareth's Downstairs Computer
> <headstone255.but.n...@yahoo.com> wrote:
>
>> In 1971, the TAD instruction of the PDP8 was a Yank joke
>> unheard of in Brit, and now 46 years later the term, a tad,
>> is widely used in UK (and was used in a recent Brit TV program
>> set in the 1930s!)
>
> Well it certainly predates 1971 - according to the OED it
> originates from the late 19th century as a term for a small child with the
> modern usage dating from the 1940s. The OED speculates that it derives from
> Tadpole while Miriam Webster opts for it deriving from toad - I'm inclined
> to think the OED one more likely. I didn't hear it much until the late
> 1970s and I had the impression it was a mostly northern expression like
> graidly, nesh and nowt.
>

Then probably like gotten to; taken to Yankland with the pilgrim
fathers while falling into disuse here until brought back from
over the pond some time later.


Stephen Wolstenholme

unread,
May 7, 2017, 8:10:38 AM5/7/17
to
On 7 May 2017 10:51:47 GMT, Bob Eager <news...@eager.cx> wrote:

>On Sun, 07 May 2017 10:07:39 +0000, mausg wrote:
>
>> On 2017-05-07, Gareth's Downstairs Computer
>> <headstone255.but.n...@yahoo.com> wrote:
>>> Brit designs had order codes, the Yanks, instruction sets.
>>>
>>> Brit computers hesitated whilst the Yank ones did DMA.
>>>
>>> Any others? ...
>>>
>>> In 1971, the TAD instruction of the PDP8 was a Yank joke unheard of in
>>> Brit, and now 46 years later the term, a tad, is widely used in UK (and
>>> was used in a recent Brit TV program set in the 1930s!)
>>>
>>>
>> The sad thing about British computer business was the the Lyons tearooms
>> control systems was way ahead of any such system anywhere, but was not
>> persued to logical conclusions.
>
>I have one of the books on LEO, and it's very good.

My first computer job was a LEO engineer in the late 1960s but I never
worked for Lyons. I worked on LEO III. I knew a chap who worked on the
first LEO in 1951.

Steve

--
Neural Network Software for Windows http://www.npsnn.com

Dennis Boone

unread,
May 7, 2017, 10:47:32 AM5/7/17
to
> Brit designs had order codes, the Yanks, instruction sets.

Some of the earlier colonial machines called them "order codes". The
"IAS" family of machines (Illiac, MISTIC, etc.) did so, for example.

My impression is that said usage mostly died out in the 50s, but I
can't prove it.

De

Whiskers

unread,
May 7, 2017, 11:06:42 AM5/7/17
to
On 2017-05-07, Ahem A Rivet's Shot <ste...@eircom.net> wrote:
> On Sun, 7 May 2017 10:41:46 +0100
> Gareth's Downstairs Computer
> <headstone255.but.n...@yahoo.com> wrote:
>
>> In 1971, the TAD instruction of the PDP8 was a Yank joke
>> unheard of in Brit, and now 46 years later the term, a tad,
>> is widely used in UK (and was used in a recent Brit TV program
>> set in the 1930s!)
>
> Well it certainly predates 1971 - according to the OED it
> originates from the late 19th century as a term for a small child with
> the modern usage dating from the 1940s. The OED speculates that it
> derives from Tadpole while Miriam Webster opts for it deriving from
> toad - I'm inclined to think the OED one more likely. I didn't hear it
> much until the late 1970s and I had the impression it was a mostly
> northern expression like graidly, nesh and nowt.

I'm sure it was in common usage in the south of England in 'middle
class' circles in the 1950s, in the sense of 'a small amount'. Along
with 'smidgin'. OED says the latter is also "orig and chiefly U.S." so
I suspect both words arrived over here along with US servicemen posted
here during WWII.

--
-- ^^^^^^^^^^
-- Whiskers
-- ~~~~~~~~~~

John Levine

unread,
May 7, 2017, 3:38:35 PM5/7/17
to
In article <oempum$9is$1...@dont-email.me>,
>In 1971, the TAD instruction of the PDP8 was a Yank joke
>unheard of in Brit, ...

When I was programming a PDP-8 in 1968, TAD didn't seem like a joke,
just a mnemonic. The contemporary PDP-7 had both one's complement
and two's complement add, and TAD came from that, to remind you that
the '8 was two's complement.

What I don't understand is why anyone ever thought one's complement
was a good idea. Negating is simpler, just flip the bits, but
addition and subtraction requires an end-around carry or borrow, and
two representations of zero caused hardware and programming pain.

I suppose if you're a mathematician two's complement is unforgivably
ugly because there are more negative numbers than positive ones, but
really now.

R's,
John

Peter Flass

unread,
May 7, 2017, 6:35:53 PM5/7/17
to
John Levine <jo...@iecc.com> wrote:
> In article <oempum$9is$1...@dont-email.me>,
>> In 1971, the TAD instruction of the PDP8 was a Yank joke
>> unheard of in Brit, ...
>
> When I was programming a PDP-8 in 1968, TAD didn't seem like a joke,
> just a mnemonic. The contemporary PDP-7 had both one's complement
> and two's complement add, and TAD came from that, to remind you that
> the '8 was two's complement.
>
> What I don't understand is why anyone ever thought one's complement
> was a good idea. Negating is simpler, just flip the bits, but
> addition and subtraction requires an end-around carry or borrow, and
> two representations of zero caused hardware and programming pain.

I think Cray used it because it required less circuitry, and hence was
faster. I think many machines forced one representation of zero in
arithmetic results, leaving the other available to tag uninitialized data.
etc.

>
> I suppose if you're a mathematician two's complement is unforgivably
> ugly because there are more negative numbers than positive ones, but
> really now.
>
> R's,
> John
>



--
Pete

jmfbahciv

unread,
May 8, 2017, 8:35:13 AM5/8/17
to
Also remember that, in those days, all manufacturers were experimenting
with the best way to do something.

Isn't there something in math which uses a negative and positive zero?
ISTR something about temperatures in physics but that could be a
random retrieval.

/BAH

Quadibloc

unread,
May 8, 2017, 9:52:29 AM5/8/17
to
On Monday, May 8, 2017 at 6:35:13 AM UTC-6, jmfbahciv wrote:

> Isn't there something in math which uses a negative and positive zero?
> ISTR something about temperatures in physics but that could be a
> random retrieval.

Floating-point representations tend to use sign-magnitude notation,
which allows for a negative zero.

Yes, negative absolute temperature is a physical concept associated with
lasers, and involves a negative zero - in this case, "infinite"
temperature is physically realizable, and involves equal numbers of
particles in the ground state and the excited state.

All particles in the ground state is (positive) absolute zero, more
particles in the excited state than the ground state is a negative
temperature - hotter than infinity, not colder than zero - and so
negative zero represents all particles in the excited state.

It's a quirky result of applying a classical definition of temperature
to a quantum system with an upper bound on energy.

John Savard

Charlie Gibbs

unread,
May 8, 2017, 12:14:27 PM5/8/17
to
On 2017-05-08, jmfbahciv <See....@aol.com> wrote:

> John Levine wrote:
>
>> In article <oempum$9is$1...@dont-email.me>,
>>
>>> In 1971, the TAD instruction of the PDP8 was a Yank joke
>>> unheard of in Brit, ...
>>
>> When I was programming a PDP-8 in 1968, TAD didn't seem like a joke,
>> just a mnemonic. The contemporary PDP-7 had both one's complement
>> and two's complement add, and TAD came from that, to remind you that
>> the '8 was two's complement.
>>
>> What I don't understand is why anyone ever thought one's complement
>> was a good idea. Negating is simpler, just flip the bits, but
>> addition and subtraction requires an end-around carry or borrow, and
>> two representations of zero caused hardware and programming pain.

Yup. I got some of that when doing assembly language programming
for IBM 360 clones. Packed decimal fields had three acceptable zeros -
0c, 0d, and 0f - although the hardware would jam the sign to C after
any calculation (including ZAP).

>> I suppose if you're a mathematician two's complement is unforgivably
>> ugly because there are more negative numbers than positive ones, but
>> really now.

"Two's complement, three's a crowd." -- Stan Kelly-Bootle

> Also remember that, in those days, all manufacturers were experimenting
> with the best way to do something.
>
> Isn't there something in math which uses a negative and positive zero?
> ISTR something about temperatures in physics but that could be a
> random retrieval.

Dunno about that, but on cold days I've occasionally seen those digital
time-and-temperature signs reading -0.

--
/~\ cgi...@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!

Scott Lurndal

unread,
May 8, 2017, 12:22:07 PM5/8/17
to
Charlie Gibbs <cgi...@kltpzyxm.invalid> writes:
>On 2017-05-08, jmfbahciv <See....@aol.com> wrote:
>
>> John Levine wrote:
>>
>>> In article <oempum$9is$1...@dont-email.me>,
>>>
>>>> In 1971, the TAD instruction of the PDP8 was a Yank joke
>>>> unheard of in Brit, ...
>>>
>>> When I was programming a PDP-8 in 1968, TAD didn't seem like a joke,
>>> just a mnemonic. The contemporary PDP-7 had both one's complement
>>> and two's complement add, and TAD came from that, to remind you that
>>> the '8 was two's complement.
>>>
>>> What I don't understand is why anyone ever thought one's complement
>>> was a good idea. Negating is simpler, just flip the bits, but
>>> addition and subtraction requires an end-around carry or borrow, and
>>> two representations of zero caused hardware and programming pain.
>
>Yup. I got some of that when doing assembly language programming
>for IBM 360 clones. Packed decimal fields had three acceptable zeros -
>0c, 0d, and 0f - although the hardware would jam the sign to C after
>any calculation (including ZAP).

Burroughs medium systems used 'C' for positive and 'D' for negative, but
the hardware would accept anything other than 'D' as positive (and
update it to C in any results).

John Levine

unread,
May 8, 2017, 1:37:32 PM5/8/17
to
In article <NW0QA.75310$yk6....@fx06.iad>,
Scott Lurndal <sl...@pacbell.net> wrote:
>>Yup. I got some of that when doing assembly language programming
>>for IBM 360 clones. Packed decimal fields had three acceptable zeros -
>>0c, 0d, and 0f - although the hardware would jam the sign to C after
>>any calculation (including ZAP).
>
>Burroughs medium systems used 'C' for positive and 'D' for negative, but
>the hardware would accept anything other than 'D' as positive (and
>update it to C in any results).

That was an artifact of EBCDIC and punch cards. Each card column had
two (conceptual) parts, a digit from 0-9 and two zone punches known
as 11 and 12. By convention, a plus sign was a zone 12 (top row)
overpunch and minus was zone 11. For text, the 12+1 to 12+9 were
the letters A-I and 11+1 to 11+9 were J-R. So if, say, your number
were positive 3, on a 360 and its successors thats 3C, which unpacks
as C3 which is the letter C which punches as 12+3. Minus 4 would
be 4B which unpacks to B4 which is the letter M which is 11+4. To
complete the kludge, EBCDIC for digits 0-9 is F0-F9, and an F sign
is positive.

It's a kludge, but one that greatly simplified coding in memory
constrained programs that used punch cards.

All the decimal arithemetic is alive and well on IBM's latest zSeries,
along with some johnny-come-lately instructions like Test Packed, to
see whether a decimal value is valid, and ASCII pack and unpack.

R's,
John



Quadibloc

unread,
May 8, 2017, 1:53:34 PM5/8/17
to
On Sunday, May 7, 2017 at 1:38:35 PM UTC-6, John Levine wrote:

> What I don't understand is why anyone ever thought one's complement
> was a good idea. Negating is simpler, just flip the bits, but
> addition and subtraction requires an end-around carry or borrow, and
> two representations of zero caused hardware and programming pain.

I remember that one manufacturer of 24-bit computers used as a selling
point the fact that their computer (like the IBM 7090) used sign-magnitude
representation of integers.

That was apparently the deluxe way of doing arithmetic! One's complement
was an inferior imitation of it... and *two's complement* was only used on
the really low-priced machines that were cutting costs by eliminating
every possible transistor!

Then IBM replaced the 7090 with the System/360 and this nonsense came to
an abrupt end. Too late to save the PDP-15, due to the need for
compatibility with its ancestors, the PDP-4, 7, and 9, from having two
versions of every arithmetic instruction - a one's complement version and
a two's complement version.

John Savard

Rich Alderson

unread,
May 8, 2017, 4:39:38 PM5/8/17
to
John Levine <jo...@iecc.com> writes:

> That was an artifact of EBCDIC and punch cards. Each card column had
> two (conceptual) parts, a digit from 0-9 and two zone punches known
> as 11 and 12. By convention, a plus sign was a zone 12 (top row)
> overpunch and minus was zone 11. For text, the 12+1 to 12+9 were
> the letters A-I and 11+1 to 11+9 were J-R. So if, say, your number
> were positive 3, on a 360 and its successors thats 3C, which unpacks
> as C3 which is the letter C which punches as 12+3. Minus 4 would
> be 4B which unpacks to B4 which is the letter M which is 11+4. To

4D, unpacking as D4 = 'M'. B4 = superscript-4 (12-11-0-4 punch).
I checked my White Card (GX20-1850-2) to verify.

The hex digits B and D were negative sign markers, but B was used in
USASCII mode, D in EBCDIC mode (PSW bit 12 to select) on the 360. I
checked that on GX20-1703-7 (the Green Card).

The other hex digits were all positive, and forced to A or C depending
on PWS bit 12.

> complete the kludge, EBCDIC for digits 0-9 is F0-F9, and an F sign
> is positive.

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

John Levine

unread,
May 8, 2017, 6:40:30 PM5/8/17
to
In article <mddtw4v...@panix5.panix.com>,
Rich Alderson <ne...@alderson.users.panix.com> wrote:
>John Levine <jo...@iecc.com> writes:
>
>> That was an artifact of EBCDIC and punch cards. Each card column had
>> two (conceptual) parts, a digit from 0-9 and two zone punches known
>> as 11 and 12. By convention, a plus sign was a zone 12 (top row)
>> overpunch and minus was zone 11. For text, the 12+1 to 12+9 were
>> the letters A-I and 11+1 to 11+9 were J-R. So if, say, your number
>> were positive 3, on a 360 and its successors thats 3C, which unpacks
>> as C3 which is the letter C which punches as 12+3. Minus 4 would
>> be 4B which unpacks to B4 which is the letter M which is 11+4. To
>
>4D, unpacking as D4 = 'M'. B4 = superscript-4 (12-11-0-4 punch).
>I checked my White Card (GX20-1850-2) to verify.

You're right, I can't type.

R's,
John

J. Clarke

unread,
May 8, 2017, 8:02:44 PM5/8/17
to
In article <oeqacr$t6k$1...@gal.iecc.com>, jo...@iecc.com says...
They've also added decimal floating point, which I have not yet had an
opportunity to play with.
>
> R's,
> John


JimP.

unread,
May 8, 2017, 8:56:00 PM5/8/17
to
On 8 May 2017 16:13:47 GMT, Charlie Gibbs <cgi...@kltpzyxm.invalid>
wrote:
Thats because the poor thing gave up and broke.
--
Jim

Charlie Gibbs

unread,
May 8, 2017, 11:14:48 PM5/8/17
to
:-)

(That's more fun than going on about Gray codes and when you should
be using them...)
0 new messages