🚀 go-pugleaf

On Sunday, May 27, 2018 at 4:12:02 PM UTC-6, Andreas Kohlbach wrote:
> I found this IMO interesting video about the birth of BASIC on
> YouTube. It also shows some old IBM and other machines used back in the
> 1960s.

I hope it shows a GE 225!

John Savard

On Sunday, May 27, 2018 at 7:43:21 PM UTC-6, Quadibloc wrote:
> On Sunday, May 27, 2018 at 4:12:02 PM UTC-6, Andreas Kohlbach wrote:
> > I found this IMO interesting video about the birth of BASIC on
> > YouTube. It also shows some old IBM and other machines used back in the
> > 1960s.

> I hope it shows a GE 225!
... or maybe a 235. Sources are inconsistent.

John Savard

On Sun, 27 May 2018 18:52:37 -0700 (PDT), Quadibloc
<jsa...@ecn.ab.ca> wrote:

>On Sunday, May 27, 2018 at 7:43:21 PM UTC-6, Quadibloc wrote:
>> On Sunday, May 27, 2018 at 4:12:02 PM UTC-6, Andreas Kohlbach wrote:
>> > I found this IMO interesting video about the birth of BASIC on
>> > YouTube. It also shows some old IBM and other machines used back in the
>> > 1960s.
>
>> I hope it shows a GE 225!
>... or maybe a 235. Sources are inconsistent.

Around 23:50 there's a diagram that clearly states GE 225.  They show
very little of the machines--the LGP-30, the 225 being unloaded and a
brief shot of the rack with what appears to be the front panel, and a
few seconds of the 635 in Kiewit.

I find it interesting that one of my employer's competitors, which
just sold us its entire US sales operation, owns DTSS.

Quadibloc <jsa...@ecn.ab.ca> Wrote in message:
> 
>> I hope it shows a GE 225!
> ... or maybe a 235. Sources are inconsistent.

The 235 wasn't ready in early 64, so GE shipped a 225 and replaced
 it with a 235 in the fall. They both appear in the
 film.

I don't. :(

sarr

-- 


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

Andreas Kohlbach wrote:

> I found this IMO interesting video about the birth of BASIC on
> YouTube. It also shows some old IBM and other machines used back in the
> 1960s.
> 
> https://www.youtube.com/watch?v=WYPNjSoDrqw
First picture is John von Neuman (horribly mispelled in subtitles) standing 
to the side of the IAS computer, with its rows of Williams tubes.  All 
vacuum tubes, built from 1946 - 1951 in Princeton, NJ.  Now at the 
Smithsonian.

Jon

On 29/05/2018 04:29, Jon Elson wrote:
> Andreas Kohlbach wrote:
> 
>> I found this IMO interesting video about the birth of BASIC on
>> YouTube. It also shows some old IBM and other machines used back in the
>> 1960s.
>>
>> https://www.youtube.com/watch?v=WYPNjSoDrqw
> First picture is John von Neuman (horribly mispelled in subtitles) standing
> to the side of the IAS computer, with its rows of Williams tubes.  All
> vacuum tubes, built from 1946 - 1951 in Princeton, NJ.  Now at the
> Smithsonian.

A moment's thought and suddenly there is nothing new under the sun ...

AIUI, the Williams Tubes relied upon the transient charge storage on the
phosphor of the screen, and therefore not dissimilar in principle to
dynamic semiconductor RAM.

On 2018-05-29, Gareth's Downstairs Computer
<headstone255.but.n...@yahoo.com> wrote:

> On 29/05/2018 04:29, Jon Elson wrote:
>
>> Andreas Kohlbach wrote:
>> 
>>> I found this IMO interesting video about the birth of BASIC on
>>> YouTube. It also shows some old IBM and other machines used back in the
>>> 1960s.
>>>
>>> https://www.youtube.com/watch?v=WYPNjSoDrqw

Nice program.  I particularly like the way they briefly mention
Bill Gates only to point out that they were there first.

>> First picture is John von Neuman (horribly mispelled in subtitles) standing
>> to the side of the IAS computer, with its rows of Williams tubes.  All
>> vacuum tubes, built from 1946 - 1951 in Princeton, NJ.  Now at the
>> Smithsonian.
>
> A moment's thought and suddenly there is nothing new under the sun ...
>
> AIUI, the Williams Tubes relied upon the transient charge storage on the
> phosphor of the screen, and therefore not dissimilar in principle to
> dynamic semiconductor RAM.

What an interesting thought...

-- 
/~\  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.
/ \  Fight low-contrast text in web pages!  http://contrastrebellion.com

On 30 May 2018 01:00:36 GMT, Charlie Gibbs wrote:
>
> On 2018-05-29, Gareth's Downstairs Computer
> <headstone255.but.n...@yahoo.com> wrote:
>
>> On 29/05/2018 04:29, Jon Elson wrote:
>>
>>> Andreas Kohlbach wrote:
>>> 
>>>> I found this IMO interesting video about the birth of BASIC on
>>>> YouTube. It also shows some old IBM and other machines used back in the
>>>> 1960s.
>>>>
>>>> https://www.youtube.com/watch?v=WYPNjSoDrqw
>
> Nice program.  I particularly like the way they briefly mention
> Bill Gates only to point out that they were there first.

The first to commercialize on BASIC. Without BASIC and the MITS 8800
there might not be a Microsoft company today. Would be interesting to
know if CP/M then would have lead the market well into the 1990s and
beyond with GEM becoming the GUI of choice and Gary if Kildall would
still be alive, not going to this bar and had this fight leading to his
death...
-- 
Andreas

https://www.youtube.com/watch?v=ptoGMcls9RM

On Thursday, May 31, 2018 at 3:11:01 PM UTC-6, Andreas Kohlbach wrote:
> On 30 May 2018 01:00:36 GMT, Charlie Gibbs wrote:

> > Nice program.  I particularly like the way they briefly mention
> > Bill Gates only to point out that they were there first.

> The first to commercialize on BASIC.

I presume it was "only to point out that (Kemeny and Kurtz) were first (instead 
of Bill Gates)" that he meant.

It's true the CINET Basic for the PDP-8 was a free program, although it was 
built from DEC's FOCAL, so you needed a FOCAL license to be permitted to order 
it from DECUS.

However, BASIC for the Honeywell 316 and 516 came with a price tag, IIRC. And 
there was CALL/360 BASIC from IBM, and I think Tymshare had BASIC... so I don't 
think Microsoft was the first to charge money for a BASIC interpreter.

John Savard

On Friday, June 1, 2018 at 3:11:14 AM UTC-4, Quadibloc wrote:

> However, BASIC for the Honeywell 316 and 516 came with a price tag, IIRC. And 
> there was CALL/360 BASIC from IBM, and I think Tymshare had BASIC... so I don't 
> think Microsoft was the first to charge money for a BASIC interpreter.

GE Timesharing offered BASIC.

HP offered its 2000 series with BASIC and was popular in time sharing.
But I don't know if users of the 2000 series had to pay for the
compiler or if it was bundled in.

hanc...@bbs.cpcn.com writes:
>On Friday, June 1, 2018 at 3:11:14 AM UTC-4, Quadibloc wrote:
>
>> However, BASIC for the Honeywell 316 and 516 came with a price tag, IIRC. And 
>> there was CALL/360 BASIC from IBM, and I think Tymshare had BASIC... so I don't 
>> think Microsoft was the first to charge money for a BASIC interpreter.
>
>GE Timesharing offered BASIC.
>
>HP offered its 2000 series with BASIC and was popular in time sharing.
>But I don't know if users of the 2000 series had to pay for the
>compiler or if it was bundled in.
>

HP-3000 had a nice BASIC interpreter as well.

On Fri, 1 Jun 2018 00:11:12 -0700 (PDT), Quadibloc wrote:
>
> On Thursday, May 31, 2018 at 3:11:01 PM UTC-6, Andreas Kohlbach wrote:
>> On 30 May 2018 01:00:36 GMT, Charlie Gibbs wrote:
>
>> > Nice program.  I particularly like the way they briefly mention
>> > Bill Gates only to point out that they were there first.
>
>> The first to commercialize on BASIC.
>
> I presume it was "only to point out that (Kemeny and Kurtz) were first (instead 
> of Bill Gates)" that he meant.
>
> It's true the CINET Basic for the PDP-8 was a free program, although it was 
> built from DEC's FOCAL, so you needed a FOCAL license to be permitted to order 
> it from DECUS.
>
> However, BASIC for the Honeywell 316 and 516 came with a price tag, IIRC. And 
> there was CALL/360 BASIC from IBM, and I think Tymshare had BASIC... so I don't 
> think Microsoft was the first to charge money for a BASIC interpreter.

Well (Microsoft) BASIC was part of most of the 8bit computers from the
late 70s to mid 80s. Any "home computer" (but the Lynx if I recall
correctly) booted directly into (MS) BASIC.

Other providers like the mentioned Honeywell might have provided BASIC,
but it were the commercial available computers for the novice in the late
70s and mid 80s kicking off people dealing with BASIC and writing
programs. Just because it came with their computer.
-- 
Andreas

My random toughts and comments
https://news-commentaries.blogspot.com/

hanc...@bbs.cpcn.com Wrote in message:
> 
> GE Timesharing offered BASIC.

GE sold the entire Dartmouth system. They didn't diverge until the
 635 based systems.

And, to kick one of my favorite dead horses, Dartmouth BASIC was
 NOT an interpreter. It compiled to real machine code. Other
 implementations varied.

Sarr


-- 


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

On Friday, June 1, 2018 at 3:09:44 PM UTC-6, Sarr Blumson wrote:

> And, to kick one of my favorite dead horses, Dartmouth BASIC was
>  NOT an interpreter. It compiled to real machine code. Other
>  implementations varied.

Yes, that's quite true, it was, I presume, one of those "load-and-go" compiler 
systems; quite a few of those had already been written for FORTRAN by that time.

But _after_ Dartmouth BASIC, nearly all the other implementations were 
interpreters. Quite a bit later, of course, it was considered worthwhile to make 
compilers for BASIC that let you save the object... CBASIC, Borland's Turbo 
BASIC, and so on. Microsoft's BASICA, GW-BASIC, and so on, were interpreters, 
while QBASIC compiled, although it didn't let you save the object - a lot like 
the original Dartmouth BASIC.

John Savard

In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
Quadibloc  <jsa...@ecn.ab.ca> wrote:
>On Friday, June 1, 2018 at 3:09:44 PM UTC-6, Sarr Blumson wrote:
>
>> And, to kick one of my favorite dead horses, Dartmouth BASIC was
>>  NOT an interpreter. It compiled to real machine code. Other
>>  implementations varied.
>
>Yes, that's quite true, it was, I presume, one of those "load-and-go" compiler 
>systems; quite a few of those had already been written for FORTRAN by that time.

Dartmouth's compiler was quite impressive, typically compiling an
entire program in one or two clock ticks.  It was so fast that nobody
bothered to save object code.

Later implementations tended to be tokenize and interpret, which has
plenty of its own issues.

-- 
Regards,
John Levine, jo...@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

On 6/1/2018 1:14 PM, Scott Lurndal wrote:
> hanc...@bbs.cpcn.com writes:
>> On Friday, June 1, 2018 at 3:11:14 AM UTC-4, Quadibloc wrote:
>>
>>> However, BASIC for the Honeywell 316 and 516 came with a price tag, IIRC. And
>>> there was CALL/360 BASIC from IBM, and I think Tymshare had BASIC... so I don't
>>> think Microsoft was the first to charge money for a BASIC interpreter.
>>
>> GE Timesharing offered BASIC.
>>
>> HP offered its 2000 series with BASIC and was popular in time sharing.
>> But I don't know if users of the 2000 series had to pay for the
>> compiler or if it was bundled in.
>>
> 
> HP-3000 had a nice BASIC interpreter as well.
> 

The HP 3000, and especially the HP 2000, minicomputers were popular to 
use in an educational setting.  Several high schools/junior highs might 
have an HP 3000 somewhere... a machine on which each of the schools 
would establish dial-up accounts for their students.

-- 
numerist at aquaporin4 dot com

On 6/1/2018 5:21 PM, Quadibloc wrote:
> On Friday, June 1, 2018 at 3:09:44 PM UTC-6, Sarr Blumson wrote:
> 
>> And, to kick one of my favorite dead horses, Dartmouth BASIC was
>>   NOT an interpreter. It compiled to real machine code. Other
>>   implementations varied.
> 
> Yes, that's quite true, it was, I presume, one of those "load-and-go" compiler
> systems; quite a few of those had already been written for FORTRAN by that time.
> 
> But _after_ Dartmouth BASIC, nearly all the other implementations were
> interpreters. Quite a bit later, of course, it was considered worthwhile to make
> compilers for BASIC that let you save the object... CBASIC, Borland's Turbo
> BASIC, and so on. Microsoft's BASICA, GW-BASIC, and so on, were interpreters,
> while QBASIC compiled, although it didn't let you save the object - a lot like
> the original Dartmouth BASIC.
> 

Sometimes BASIC would be compiled *not* to machine language, but to some 
mid-level code that was much easier to interpret than working from the 
original BASIC source code.  Although the mid-level language was *not* 
as efficient as the machine code would have been, this still would speed 
along the interpreter quite a bit.  For example, a FOR loop that 
executed its code 10,000 times would take *much* longer if you had to 
re-translate the source of each statement every time through the loop!

-- 
numerist at aquaporin4 dot com

On 02/06/2018 03:44, John Levine wrote:
> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>> On Friday, June 1, 2018 at 3:09:44 PM UTC-6, Sarr Blumson wrote:
>>
>>> And, to kick one of my favorite dead horses, Dartmouth BASIC was
>>>   NOT an interpreter. It compiled to real machine code. Other
>>>   implementations varied.
>>
>> Yes, that's quite true, it was, I presume, one of those "load-and-go" compiler
>> systems; quite a few of those had already been written for FORTRAN by that time.
> 
> Dartmouth's compiler was quite impressive, typically compiling an
> entire program in one or two clock ticks.  It was so fast that nobody
> bothered to save object code.

Interesting, does that also indicate interactive development where
the compilation speed would be largely unapparent?

On Friday, June 1, 2018 at 5:03:15 PM UTC-4, Andreas Kohlbach wrote:

> Other providers like the mentioned Honeywell might have provided BASIC,
> but it were the commercial available computers for the novice in the late
> 70s and mid 80s kicking off people dealing with BASIC and writing
> programs. Just because it came with their computer.

>From the late 1960s onward, many schools got Teletypes and timesharing
and offered BASIC classes.

In 1968, our school district had GE Timesharing.

(Some school districts had other kinds of computers.)

On Sat, 2 Jun 2018 12:23:56 +0100, Gareth's Downstairs Computer wrote:
>
> On 02/06/2018 03:44, John Levine wrote:
>> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>
>> Dartmouth's compiler was quite impressive, typically compiling an
>> entire program in one or two clock ticks.  It was so fast that nobody
>> bothered to save object code.
>
> Interesting, does that also indicate interactive development where
> the compilation speed would be largely unapparent?

I guess I am missing a link here. Why did none of the BASICs coming with
the 8bit micros need a compiler and rather interpret the code instead?
-- 
Andreas

My random toughts and comments
https://news-commentaries.blogspot.com/

On Sat, 02 Jun 2018 16:07:09 -0400, Andreas Kohlbach
<a...@spamfence.net> wrote:

>On Sat, 2 Jun 2018 12:23:56 +0100, Gareth's Downstairs Computer wrote:
>>
>> On 02/06/2018 03:44, John Levine wrote:
>>> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
>>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>>
>>> Dartmouth's compiler was quite impressive, typically compiling an
>>> entire program in one or two clock ticks.  It was so fast that nobody
>>> bothered to save object code.
>>
>> Interesting, does that also indicate interactive development where
>> the compilation speed would be largely unapparent?
>
>I guess I am missing a link here. Why did none of the BASICs coming with
>the 8bit micros need a compiler and rather interpret the code instead?

I think the real issue is the nature of a compiler and cost and
availability of hardware.

A compiler takes a source program, generates an object program, and
then to actually do anything you need to load and execute the object
program.

To make this work you need to have enough storage of some kind (paper
tape, mag tape, disk, drum, RAM, doesn't matter, you need to have
enough) to store all the pieces.  Early micros did not have much RAM
and typically if they had a storage device other than the operator's
paper notebook it was paper tape or an audio cassette recorder, so
turnaround time for compiled code would be slow.

An interpreter only needs storage for the source code, a scratchpad,
and the output, so it can run reasonably well on a machine with
limited RAM and a slow storage device, hence the popularity on early
hobbyist micros (and yes, to the nits who think that hobbyists are
only allowed to own toys, I mean the ALTAIR and its clones).

On Sat, 02 Jun 2018 16:07:09 -0400, Andreas Kohlbach wrote:

> On Sat, 2 Jun 2018 12:23:56 +0100, Gareth's Downstairs Computer wrote:
>>
>> On 02/06/2018 03:44, John Levine wrote:
>>> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
>>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>>
>>> Dartmouth's compiler was quite impressive, typically compiling an
>>> entire program in one or two clock ticks.  It was so fast that nobody
>>> bothered to save object code.
>>
>> Interesting, does that also indicate interactive development where the
>> compilation speed would be largely unapparent?
> 
> I guess I am missing a link here. Why did none of the BASICs coming with
> the 8bit micros need a compiler and rather interpret the code instead?

Several reasons, but space was one issue. If you compiled into a custom 
interpretive code, you could optimise the design of that code to be very 
compact. This saved precious memory space. One of my colleagues produced 
a BASIC system that didn't store the source code at all, but rather 
stored a bit more with the interpretive code - he could then regenerate 
the source code if the user wanted to list it or save it.

There was also the overhead of linkers, loaders etc. all of which were 
not needed if interpreted. For most purposes, speed wasn't an issue.



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

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

On Sat, 02 Jun 2018 17:20:59 -0400, J. Clarke
<jclarke...@gmail.com> wrote:

>On Sat, 02 Jun 2018 16:07:09 -0400, Andreas Kohlbach
><a...@spamfence.net> wrote:

[snip]

>>I guess I am missing a link here. Why did none of the BASICs coming with
>>the 8bit micros need a compiler and rather interpret the code instead?
>
>I think the real issue is the nature of a compiler and cost and
>availability of hardware.
>
>A compiler takes a source program, generates an object program, and
>then to actually do anything you need to load and execute the object
>program.
>
>To make this work you need to have enough storage of some kind (paper
>tape, mag tape, disk, drum, RAM, doesn't matter, you need to have
>enough) to store all the pieces.  Early micros did not have much RAM
>and typically if they had a storage device other than the operator's
>paper notebook it was paper tape or an audio cassette recorder, so
>turnaround time for compiled code would be slow.
>
>An interpreter only needs storage for the source code, a scratchpad,
>and the output, so it can run reasonably well on a machine with
>limited RAM and a slow storage device, hence the popularity on early
>hobbyist micros (and yes, to the nits who think that hobbyists are
>only allowed to own toys, I mean the ALTAIR and its clones).

     To get specific, a tiny BASIC (interpreted) could be implemented
in 2K.  A compiler needs a lot more.

Sincerely,

Gene Wirchenko

On 6/2/2018 3:07 PM, Andreas Kohlbach wrote:
> On Sat, 2 Jun 2018 12:23:56 +0100, Gareth's Downstairs Computer wrote:
>>
>> On 02/06/2018 03:44, John Levine wrote:
>>> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
>>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>>
>>> Dartmouth's compiler was quite impressive, typically compiling an
>>> entire program in one or two clock ticks.  It was so fast that nobody
>>> bothered to save object code.
>>
>> Interesting, does that also indicate interactive development where
>> the compilation speed would be largely unapparent?
> 
> I guess I am missing a link here. Why did none of the BASICs coming with
> the 8bit micros need a compiler and rather interpret the code instead?
> 
Osborne came with CBASIC as well as MBASIC. At least sometimes... the 
mix of bundled software seemed to change over time. I had a copy of 
CBASIC "borrowed" from an Osborne that ran on my Morrow MD-3. That 
might not work in an Altair, but worked fine in the generic 64K CP/M 
computers.

The immediacy of an interpreter was popular with new computer users, 
though.

On 02/06/2018 21:07, Andreas Kohlbach wrote:
> On Sat, 2 Jun 2018 12:23:56 +0100, Gareth's Downstairs Computer wrote:
>>
>> On 02/06/2018 03:44, John Levine wrote:
>>> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
>>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>>
>>> Dartmouth's compiler was quite impressive, typically compiling an
>>> entire program in one or two clock ticks.  It was so fast that nobody
>>> bothered to save object code.
>>
>> Interesting, does that also indicate interactive development where
>> the compilation speed would be largely unapparent?
> 
> I guess I am missing a link here. Why did none of the BASICs coming with
> the 8bit micros need a compiler and rather interpret the code instead?
> 
I guess in the early days there were not many folks with the cash for a 
floppy disk drive. Given you are working with tape loading and running 
from memory is attractive...

Dave

On Sat, 02 Jun 2018 16:07:09 -0400
Andreas Kohlbach <a...@spamfence.net> wrote:

> I guess I am missing a link here. Why did none of the BASICs coming with
> the 8bit micros need a compiler and rather interpret the code instead?

	Because for an interactive system where the language and the CLI
are one and the same (most 8 bit home micros) an interpreted language is
pretty much essential.

-- 
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/

On Sat, 2 Jun 2018 12:18:10 -0700 (PDT)
hanc...@bbs.cpcn.com wrote:

> On Friday, June 1, 2018 at 5:03:15 PM UTC-4, Andreas Kohlbach wrote:
> 
> > Other providers like the mentioned Honeywell might have provided BASIC,
> > but it were the commercial available computers for the novice in the
> > late 70s and mid 80s kicking off people dealing with BASIC and writing
> > programs. Just because it came with their computer.
> 
> From the late 1960s onward, many schools got Teletypes and timesharing
> and offered BASIC classes.

	In the early to mid 70s our school had a Teletype, and there was a
lunchtime BASIC class offered to a few of us. The Teletype did not connect
to a timesharing service, it connected to another Teletype to send and
receive paper tapes of programs to be run, and the results of yesterday's
runs. Getting a syntax error back was a real time waster.

	The Teletype was only available to us during break and lunch. After
school a sixth former would make the call and deal with the tapes whilst
trying to chat up the operator at the other end.

-- 
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/

On 03/06/2018 08:06, David Wade wrote:
> On 02/06/2018 21:07, Andreas Kohlbach wrote:
>> On Sat, 2 Jun 2018 12:23:56 +0100, Gareth's Downstairs Computer wrote:
>>>
>>> On 02/06/2018 03:44, John Levine wrote:
>>>> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
>>>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>>>
>>>> Dartmouth's compiler was quite impressive, typically compiling an
>>>> entire program in one or two clock ticks.  It was so fast that nobody
>>>> bothered to save object code.
>>>
>>> Interesting, does that also indicate interactive development where
>>> the compilation speed would be largely unapparent?
>>
>> I guess I am missing a link here. Why did none of the BASICs coming with
>> the 8bit micros need a compiler and rather interpret the code instead?
>>
> I guess in the early days there were not many folks with the cash for a 
> floppy disk drive. Given you are working with tape loading and running 
> from memory is attractive...
Well it certainly was a pain in the bum on an early PDP11/20 with a 10
chars per sec tape reader on the teletype!

3/4 hour to load the editor, then another 3/4 hour to load the assembler,
not to even mention the two passes of source through the assembler!

On 03/06/2018 14:18, Ahem A Rivet's Shot wrote:
> On Sat, 02 Jun 2018 16:07:09 -0400
> Andreas Kohlbach <a...@spamfence.net> wrote:
> 
>> I guess I am missing a link here. Why did none of the BASICs coming with
>> the 8bit micros need a compiler and rather interpret the code instead?
> 
> 	Because for an interactive system where the language and the CLI
> are one and the same (most 8 bit home micros) an interpreted language is
> pretty much essential.
> 

For compactness (and also write-only-non-readable code) FORTH takes the 
biscuit.

Unfortunately some big systems have been implemented in FORTH, but only
because the programmer with ten year's experience had only 1 week's 
experience repeated 519 times  :-(

J. Clarke <jclarke...@gmail.com> wrote:
> On Sat, 02 Jun 2018 16:07:09 -0400, Andreas Kohlbach
> <a...@spamfence.net> wrote:
> 
>> On Sat, 2 Jun 2018 12:23:56 +0100, Gareth's Downstairs Computer wrote:
>>> 
>>> On 02/06/2018 03:44, John Levine wrote:
>>>> In article <cb11f87e-44a9-4f0b...@googlegroups.com>,
>>>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>>> 
>>>> Dartmouth's compiler was quite impressive, typically compiling an
>>>> entire program in one or two clock ticks.  It was so fast that nobody
>>>> bothered to save object code.
>>> 
>>> Interesting, does that also indicate interactive development where
>>> the compilation speed would be largely unapparent?
>> 
>> I guess I am missing a link here. Why did none of the BASICs coming with
>> the 8bit micros need a compiler and rather interpret the code instead?
> 
> I think the real issue is the nature of a compiler and cost and
> availability of hardware.
> 
> A compiler takes a source program, generates an object program, and
> then to actually do anything you need to load and execute the object
> program.
> 
> To make this work you need to have enough storage of some kind (paper
> tape, mag tape, disk, drum, RAM, doesn't matter, you need to have
> enough) to store all the pieces.  Early micros did not have much RAM
> and typically if they had a storage device other than the operator's
> paper notebook it was paper tape or an audio cassette recorder, so
> turnaround time for compiled code would be slow.
> 
> An interpreter only needs storage for the source code, a scratchpad,
> and the output, so it can run reasonably well on a machine with
> limited RAM and a slow storage device, hence the popularity on early
> hobbyist micros (and yes, to the nits who think that hobbyists are
> only allowed to own toys, I mean the ALTAIR and its clones).
> 

Good point.

-- 
Pete

On 2018-06-01 6:21 PM, Quadibloc wrote:
> On Friday, June 1, 2018 at 3:09:44 PM UTC-6, Sarr Blumson wrote:
> 
>> And, to kick one of my favorite dead horses, Dartmouth BASIC was
>>   NOT an interpreter. It compiled to real machine code. Other
>>   implementations varied.
> 
> Yes, that's quite true, it was, I presume, one of those "load-and-go" compiler
> systems; quite a few of those had already been written for FORTRAN by that time.
> 
> But _after_ Dartmouth BASIC, nearly all the other implementations were
> interpreters.

BASIC under GECOS TSS on the 635 was a compiler.
Actually, today it STILL IS although the machines with
that architecture are called DPS-9000, and GECOS renamed GCOS8.
However, it is essentially the same BASIC as the Dartmouth
implementation under DTSS on the 635.

On 2018-06-02 7:23 AM, Gareth's Downstairs Computer wrote:
> On 02/06/2018 03:44, John Levine wrote:

>> Dartmouth's compiler was quite impressive,
>> typically compiling an entire program in one or
>> two clock ticks.  It was so fast that nobody
>> bothered to save object code.

Since the 635 and several generations of successors
were asynchronous machines I'm not clear what is
meant by "clock tick"
> 
> Interesting, does that also indicate interactive
> development where
> the compilation speed would be largely unapparent?

*date
06/04/18 10:08:23
*list
10 let x=0
20 for i= 1 to 10
30 let x = x + i
40 next i
50 print x


*run



           55


ready
*date
06/04/18 10:08:23

On Monday, June 4, 2018 at 11:12:41 AM UTC-6, Alan Bowler wrote:
> On 2018-06-02 7:23 AM, Gareth's Downstairs Computer wrote:
> > On 02/06/2018 03:44, John Levine wrote:
> 
> >> Dartmouth's compiler was quite impressive,
> >> typically compiling an entire program in one or
> >> two clock ticks.  It was so fast that nobody
> >> bothered to save object code.
> 
> Since the 635 and several generations of successors
> were asynchronous machines I'm not clear what is
> meant by "clock tick"

I don't think any compiler, no matter how impresive, could compile a program of 
any length in two processor clock cycles.

Compiling one in *two seconds* of wall-clock time, however, is perfectly 
possible, and it would have been impressive on a computer made from discrete 
transistors.

So I think the problem is that he meant a completely different kind of "clock 
tick" from the kind you were thinking of that the GE 635 didn't have.

John Savard

On Monday, June 4, 2018 at 11:28:18 AM UTC-6, Quadibloc wrote:
> On Monday, June 4, 2018 at 11:12:41 AM UTC-6, Alan Bowler wrote:
> > On 2018-06-02 7:23 AM, Gareth's Downstairs Computer wrote:
> > > On 02/06/2018 03:44, John Levine wrote:

> > >> Dartmouth's compiler was quite impressive,
> > >> typically compiling an entire program in one or
> > >> two clock ticks.  It was so fast that nobody
> > >> bothered to save object code.

> > Since the 635 and several generations of successors
> > were asynchronous machines I'm not clear what is
> > meant by "clock tick"

> I don't think any compiler, no matter how impresive, could compile a program of 
> any length in two processor clock cycles.

> Compiling one in *two seconds* of wall-clock time, however, is perfectly 
> possible, and it would have been impressive on a computer made from discrete 
> transistors.

> So I think the problem is that he meant a completely different kind of "clock 
> tick" from the kind you were thinking of that the GE 635 didn't have.

Or it could have even been able to compile a program in 1/30 of a second, not 
just two seconds - which would be two 1/60 second "ticks" of the computer's 
real-time clock... at least, many mainframes of that era synchronized their real 
time clock to the mains frequency.

Oops, the AC power frequency. If they called it _that_, it would be 1/50 second.

John Savard

On Saturday, June 2, 2018 at 4:07:10 PM UTC-4, Andreas Kohlbach wrote:
> I guess I am missing a link here. Why did none of the BASICs coming with
> the 8bit micros need a compiler and rather interpret the code instead?

FWIW, to provide some background:

When I bought my 286, it came with MS-DOS 3 which included GW-BASIC.

I went out and bought the M/S QuickBasic compiler, which cost about
$50 back then.  (This was the much cheaper home version of the 
"Professional BASIC"*).  Some compiled programs ran much faster
than the GW version, and the development environment was easier.
QB included what later became the DOS EDIT (introduced as a standalone
in DOS 5.0, which also included a smaller QB).

Aside from being compiled, QB offered features to make better
use of memory and better buffering which sped up I/O.  It also
allowed accessing interrupts, so one could program in handlers
for I/O device problems, e.g. paper not loaded or diskette not
inserted.

I purchased the book, QuickBasic Bible, which contained a
tremendous amount of useful information, including advanced
features.


* They later had a promotion for the Professional Version for $100
and I got it.  But QB did virtually everything the PB did and I
rarely used the PB version.  One feature was a currency data-type,
but I didn't need that.

On Sunday, June 3, 2018 at 9:30:04 AM UTC-4, Ahem A Rivet's Shot wrote:

> > From the late 1960s onward, many schools got Teletypes and timesharing
> > and offered BASIC classes.
> 
> 	In the early to mid 70s our school had a Teletype, and there was a
> lunchtime BASIC class offered to a few of us. The Teletype did not connect
> to a timesharing service, it connected to another Teletype to send and
> receive paper tapes of programs to be run, and the results of yesterday's
> runs. Getting a syntax error back was a real time waster.
> 
> 	The Teletype was only available to us during break and lunch. After
> school a sixth former would make the call and deal with the tapes whilst
> trying to chat up the operator at the other end.

OMG, that would be cumbersome--a 24 hour turnaround.  I would think
that would be very discouraging to students--having to wait an
entire day only to learn a comma was left out, and then needing to
wait another day for another submission.

However, there were numerous times in industry in the 1980s and 1990s
where the company computer would be so busy with production work that
programmer work had to wait until overnight, so there was the similar
long wait for results.

In those days, we were often encouraged to work evening, late
night, or weekend hours to use the computer.  People with families
didn't care for it since they missed family time.  However, single
people liked it* because it gave them a lot of flexibility.

I liked it because I could take care of day chores, such as car
repair or doctor appts without taking time off from work--I'd just
work late or on a weekend to make up the time.  I liked the flexibility.

However, I've noticed today that modern management is much less flexible
and expect staff to work on their prescribed schedule without any
deviation**.  That kind of sucks.


* There were a few people going through a rough patch at home,
and they often worked odd hours to get out of the house.

 
** My neighbor was directed to fly to a different city, take a training
class, fly home, and report to work for the night shift, all in ONE
day.  She'd be working basically 24 hours straight through.  She'd
be only paid for one day; travel time and class time was not covered.

Indeed, I've heard of a lot of companies these days, including
government, having their workers take mandated training on their
own time, and even their own expense.  To me, if a company requires
you to take certain training, they should cover all expenses of it.

On 04/06/2018 20:38, hanc...@bbs.cpcn.com wrote:
> 
> OMG, that would be cumbersome--a 24 hour turnaround.  I would think
> that would be very discouraging to students--having to wait an
> entire day only to learn a comma was left out, and then needing to
> wait another day for another submission.

That was what happened in 1969 for we engineering students on
the FORTRAN course at Essex; Put your punched card deck in an
in-tray, and collect the output, wrapped around your card deck, the
following day.

'twas an ICL 1900 in them thar days.

On Monday, June 4, 2018 at 4:31:09 PM UTC-4, Gareth's Downstairs Computer wrote:
> On 04/06/2018 20:38, hanc...@bbs.cpcn.com wrote:
> > 
> > OMG, that would be cumbersome--a 24 hour turnaround.  I would think
> > that would be very discouraging to students--having to wait an
> > entire day only to learn a comma was left out, and then needing to
> > wait another day for another submission.
> 
> That was what happened in 1969 for we engineering students on
> the FORTRAN course at Essex; Put your punched card deck in an
> in-tray, and collect the output, wrapped around your card deck, the
> following day.
> 
> 'twas an ICL 1900 in them thar days.

If I were in England, I'd have done anything to work with the
Leo/Lyons snake cake company and their early computer; both to get
computer work and get snack cakes.

https://en.wikipedia.org/wiki/LEO_(computer)


We had a tour of a commercial bakery back as kids.  It amazes me
today we were taken back among the giant mixers and fermentation
room--a bunch of little kids sneezing and such.

Back then, the process seemed quite automated.  The raw materials
were unloaded by air suction from freight trains and delivered
to the top floor.  There they loaded to mixers, and passed by
chute to the lower floors for successive steps. I presume today
it is even more automated.

I wonder if there is still a Q/A dept were workers sampled by
eating the various products.  They gave us plenty of free samples
to take home.  We enjoyed that!




today

In article <r1c6hd9alvtcdvjo0...@4ax.com>,
Gene Wirchenko  <ge...@telus.net> wrote:
>     To get specific, a tiny BASIC (interpreted) could be implemented
>in 2K.  A compiler needs a lot more.

I don't think the Dartmouth compiler was much bigger than that since mainframe memory
in the 1960s was very expensive, but they did have fast disks.

-- 
Regards,
John Levine, jo...@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

On Tue, 05 Jun 2018 08:27:55 GMT, Huge <Huge@nowhere.much.invalid> wrote:


[some lines snipped]
> 
> (Note for American readers; whenever I say "biscuit" in the following,
> please read "cookie" instead!)
> 

I refuse cookies as a matter of principle!
[]

-- 
Bah, and indeed, Humbug.

On 05/06/2018 09:27, Huge wrote:
> 
> Early on in my career, I spent a couple of years writing process control
> software for United Biscuits;

>  We
> also got an allowance of free product and you were allowed to help
> yourself off the production lines.
> 
> I put on a lot of weight.
> 
> 

Early on in my career, I spent a few summer holidays from school
working at Hale's Cakes in Clevedon.

They said eat what you want on the first day, but on the second
you never wanted to eat any of it again!

Working on the Battenburg production line, it was surprising how
much went to pig swill as a matter of course.

Quadibloc <jsa...@ecn.ab.ca> wrote:
> On Monday, June 4, 2018 at 11:28:18 AM UTC-6, Quadibloc wrote:
>> On Monday, June 4, 2018 at 11:12:41 AM UTC-6, Alan Bowler wrote:
>>> On 2018-06-02 7:23 AM, Gareth's Downstairs Computer wrote:
>>>> On 02/06/2018 03:44, John Levine wrote:
> 
>>>>> Dartmouth's compiler was quite impressive,
>>>>> typically compiling an entire program in one or
>>>>> two clock ticks.  It was so fast that nobody
>>>>> bothered to save object code.
> 
>>> Since the 635 and several generations of successors
>>> were asynchronous machines I'm not clear what is
>>> meant by "clock tick"
> 
>> I don't think any compiler, no matter how impresive, could compile a program of 
>> any length in two processor clock cycles.
> 
>> Compiling one in *two seconds* of wall-clock time, however, is perfectly 
>> possible, and it would have been impressive on a computer made from discrete 
>> transistors.
> 
>> So I think the problem is that he meant a completely different kind of "clock 
>> tick" from the kind you were thinking of that the GE 635 didn't have.
> 
> Or it could have even been able to compile a program in 1/30 of a second, not 
> just two seconds - which would be two 1/60 second "ticks" of the computer's 
> real-time clock... at least, many mainframes of that era synchronized their real 
> time clock to the mains frequency.
> 
> Oops, the AC power frequency. If they called it _that_, it would be 1/50 second.

<g>

-- 
Pete

Kerr-Mudd,John <nots...@invalid.org> wrote:
> On Tue, 05 Jun 2018 08:27:55 GMT, Huge <Huge@nowhere.much.invalid> wrote:
> 
> 
> [some lines snipped]
>> 
>> (Note for American readers; whenever I say "biscuit" in the following,
>> please read "cookie" instead!)
>> 
> 
> I refuse cookies as a matter of principle!
> []
> 

I agree to accept cookies.

-- 
Pete

<hanc...@bbs.cpcn.com> wrote:
> On Sunday, June 3, 2018 at 9:30:04 AM UTC-4, Ahem A Rivet's Shot wrote:
> 
>>> From the late 1960s onward, many schools got Teletypes and timesharing
>>> and offered BASIC classes.
>> 
>> In the early to mid 70s our school had a Teletype, and there was a
>> lunchtime BASIC class offered to a few of us. The Teletype did not connect
>> to a timesharing service, it connected to another Teletype to send and
>> receive paper tapes of programs to be run, and the results of yesterday's
>> runs. Getting a syntax error back was a real time waster.
>> 
>> The Teletype was only available to us during break and lunch. After
>> school a sixth former would make the call and deal with the tapes whilst
>> trying to chat up the operator at the other end.
> 
> OMG, that would be cumbersome--a 24 hour turnaround.  I would think
> that would be very discouraging to students--having to wait an
> entire day only to learn a comma was left out, and then needing to
> wait another day for another submission.
> 
> However, there were numerous times in industry in the 1980s and 1990s
> where the company computer would be so busy with production work that
> programmer work had to wait until overnight, so there was the similar
> long wait for results.

BTDT. I used to work on several programs at once, so one could be waiting
for a debug run, another in keypunching, and a third (or more) being coded.

> 
> In those days, we were often encouraged to work evening, late
> night, or weekend hours to use the computer.  People with families
> didn't care for it since they missed family time.  However, single
> people liked it* because it gave them a lot of flexibility.
> 
> I liked it because I could take care of day chores, such as car
> repair or doctor appts without taking time off from work--I'd just
> work late or on a weekend to make up the time.  I liked the flexibility.
> 
> However, I've noticed today that modern management is much less flexible
> and expect staff to work on their prescribed schedule without any
> deviation**.  That kind of sucks.

Nowadays (from what I've seen, I'm happily out of the rat race) they expect
you to work your regular hours plus whatever time you need at home to fix
some "urgent" problem that could easily wait on nights and weekends. (
without extra pay or comp time, of course).

-- 
Pete

On Tue, 5 Jun 2018 12:02:25 -0400, Peter Flass <peter...@yahoo.com>
wrote:

>Quadibloc <jsa...@ecn.ab.ca> wrote:
>> On Monday, June 4, 2018 at 11:28:18 AM UTC-6, Quadibloc wrote:
>>> On Monday, June 4, 2018 at 11:12:41 AM UTC-6, Alan Bowler wrote:
>>>> On 2018-06-02 7:23 AM, Gareth's Downstairs Computer wrote:
>>>>> On 02/06/2018 03:44, John Levine wrote:
>> 
>>>>>> Dartmouth's compiler was quite impressive,
>>>>>> typically compiling an entire program in one or
>>>>>> two clock ticks.  It was so fast that nobody
>>>>>> bothered to save object code.
>> 
>>>> Since the 635 and several generations of successors
>>>> were asynchronous machines I'm not clear what is
>>>> meant by "clock tick"
>> 
>>> I don't think any compiler, no matter how impresive, could compile a program of 
>>> any length in two processor clock cycles.
>> 
>>> Compiling one in *two seconds* of wall-clock time, however, is perfectly 
>>> possible, and it would have been impressive on a computer made from discrete 
>>> transistors.
>> 
>>> So I think the problem is that he meant a completely different kind of "clock 
>>> tick" from the kind you were thinking of that the GE 635 didn't have.
>> 
>> Or it could have even been able to compile a program in 1/30 of a second, not 
>> just two seconds - which would be two 1/60 second "ticks" of the computer's 
>> real-time clock... at least, many mainframes of that era synchronized their real 
>> time clock to the mains frequency.
>> 
>> Oops, the AC power frequency. If they called it _that_, it would be 1/50 second.
>
><g>

Or 1/60 or 1/400.

John Levine <jo...@taugh.com> Wrote in message:
> In article <r1c6hd9alvtcdvjo0...@4ax.com>,
> Gene Wirchenko  <ge...@telus.net> wrote:
>>     To get specific, a tiny BASIC (interpreted) could be implemented
>>in 2K.  A compiler needs a lot more.
> 
> I don't think the Dartmouth compiler was much bigger than that since mainframe memory
> in the 1960s was very expensive, but they did have fast disks.

The 225/235 had 16k 20 bit words. By convention the upper 8k was
 read only system code, roughly 2k for the OS and 6k for the
 "system program" eg the compiler and runtime library. The lower
 8k was similarly divided into 2k of data for the OS and 6k for
 the program. The user program was compiled into the 6k data
 space.

BASIC was small enough to fit the compiler and runtime into one 6k
 load but ALGOL, for example was not. The scheduler was smart
 enough to put user jobs that used the same "high segment"
 together to save having to reload it, because the disks weren't
 actually all that fast by modern standards. 100 msec seeks. And
 you could see the data transfer in the lights.

Far more than you wanted to know.

Sarr




-- 


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

In article <pf6p3h$g5u$1...@dont-email.me>,
Sarr Blumson  <sarr.blumson@alum.dartmouth.orgq> wrote:
>John Levine <jo...@taugh.com> Wrote in message:
>> In article <r1c6hd9alvtcdvjo0...@4ax.com>,
>> Gene Wirchenko  <ge...@telus.net> wrote:
>>>     To get specific, a tiny BASIC (interpreted) could be implemented
>>>in 2K.  A compiler needs a lot more.
>> 
>> I don't think the Dartmouth compiler was much bigger than that since
>mainframe memory
>> in the 1960s was very expensive, but they did have fast disks.
>
>The 225/235 had 16k 20 bit words. By convention the upper 8k was
> read only system code, roughly 2k for the OS and 6k for the
> "system program" eg the compiler and runtime library. The lower
> 8k was similarly divided into 2k of data for the OS and 6k for
> the program. The user program was compiled into the 6k data
> space.
>
>BASIC was small enough to fit the compiler and runtime into one 6k
> load but ALGOL, for example was not. The scheduler was smart
> enough to put user jobs that used the same "high segment"
> together to save having to reload it, because the disks weren't
> actually all that fast by modern standards. 100 msec seeks. And
> you could see the data transfer in the lights.
>
>Far more than you wanted to know.

[smw] But it brings back lots of good memories.

[smw] The first system I was paid to use was a GE-265 run
by Pillsbury's Call-A-Computer division.

[smw] In Fortran (II as I recall) there was a trick of
using a negative index into an array to get various pieces
of system information from the OS area.

[smw] In addition to BASIC, ALGOL, and Fortran, there was
also LISP 1.5 available.

[smw] What was more fun than watching the disk lights was
watching the arms move.  The disk drive on the system I used
was in a big box with windows so you could see this happen.

[smw] On a different branch of this topic, in high school for
my junior year, we handed in coding sheets which were punched
by the school department's key punchers and taken by our
teacher to first Harvard (7094) and later a defense contractor
(360/20 IIRC.)  We usually got the cards and printout a week later.
The next year's class got to use a TTY33 to access the GE-265.

Sam

wei...@TheWorld.com (Sam Weiner) Wrote in message:
> 
> 
> [smw] The first system I was paid to use was a GE-265 run
> by Pillsbury's Call-A-Computer division.

235 + (Datanet)30 = 265. Thank you GE marketing.

> [smw] In addition to BASIC, ALGOL, and Fortran, there was
> also LISP 1.5 available.

FORTRAN was the only piece not developed at Dartmouth. GE's
 refusal to let us use that was a source of much conversation.
 :)

Sarr


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

On 2018-06-05 6:43 AM, John Levine wrote:
> In article <r1c6hd9alvtcdvjo0...@4ax.com>,
> Gene Wirchenko  <ge...@telus.net> wrote:
>>      To get specific, a tiny BASIC (interpreted) could be implemented
>> in 2K.  A compiler needs a lot more.
> 
> I don't think the Dartmouth compiler was much bigger than that since mainframe memory
> in the 1960s was very expensive, but they did have fast disks.
> 

Well today the BASIC compiler on GCOS8
(essentially the Dartmouth compiler for the GE-635)
needs 13479 36-bit words (53916 9-bit bytes) to load.

On 2018-06-04 1:28 PM, Quadibloc wrote:
> On Monday, June 4, 2018 at 11:12:41 AM UTC-6, Alan Bowler wrote:
>> On 2018-06-02 7:23 AM, Gareth's Downstairs Computer wrote:
>>> On 02/06/2018 03:44, John Levine wrote:
>>
>>>> Dartmouth's compiler was quite impressive,
>>>> typically compiling an entire program in one or
>>>> two clock ticks.  It was so fast that nobody
>>>> bothered to save object code.
>>
>> Since the 635 and several generations of successors
>> were asynchronous machines I'm not clear what is
>> meant by "clock tick"
> 
> I don't think any compiler, no matter how impresive, could compile a program of
> any length in two processor clock cycles.
> 
> Compiling one in *two seconds* of wall-clock time, however, is perfectly
> possible, and it would have been impressive on a computer made from discrete
> transistors.
> 
> So I think the problem is that he meant a completely different kind of "clock
> tick" from the kind you were thinking of that the GE 635 didn't have.
> 

I should have expanded a bit more.  The 635 did have CPU timer
that was used to give programmed interrupts (used by the
CPU dispatcher).  It worked in units of 1/64000 second.