From: Bill Birch
Subject: In mourning after LISP chuck-out session!
Date: 
Message-ID: <34602EB8.710C26C2@ctp.com>
Hi, 

Due to space restrictions at home, I recently went through my collection
of LISPS (and papers on LISP) and chucked most of them out. The list
included: GCLISP, XLISP 1.7, KERN, LE-LISP in French, ISO LISP, DYLAN. I
am now in a state of mourning. I'm sure that I'll need to refer to one
of these one day. But then Common LISP does it all.

I kept my copy of the LISP 1.5 Programmers Manual though :-) Does
anyone  have source for the original LISP, I wonder if it still runs... 

Bill

From: William Paul Vrotney
Subject: Re: In mourning after LISP chuck-out session!
Date: 
Message-ID: <vrotneyEJ74wz.709@netcom.com>
In article <·················@ctp.com> Bill Birch <······@ctp.com> writes:

> 
> Due to space restrictions at home, I recently went through my collection
> of LISPS (and papers on LISP) and chucked most of them out. The list
> included: GCLISP, XLISP 1.7, KERN, LE-LISP in French, ISO LISP, DYLAN. I
> am now in a state of mourning. I'm sure that I'll need to refer to one
> of these one day. But then Common LISP does it all.
> 
> I kept my copy of the LISP 1.5 Programmers Manual though :-) Does
> anyone  have source for the original LISP, I wonder if it still runs... 
> 

I assume that the original Lisp was written in assembly language in which
case you would have to find a running IBM 704 computer.

Just an aside.  The first Lisp was supposed to have commas between arguments
and the programmer had a bug that failed to print the commas.  When they saw
the output without commas they decided they liked it and to leave it that
way.  This has been documented as "a case of a good bug".

-- 

William P. Vrotney - ·······@netcom.com
From: Donald Fisk
Subject: Re: In mourning after LISP chuck-out session!
Date: 
Message-ID: <34619C01.3535@bt-sys.bt.spamblock.co.uk>
William Paul Vrotney wrote:
> I assume that the original Lisp was written in assembly language in which
> case you would have to find a running IBM 704 computer.

Or an emulator for the 704, which could be written as long as a 
specification of the instruction set is still available (which
I presume it is).

> Just an aside.  The first Lisp was supposed to have commas between arguments
> and the programmer had a bug that failed to print the commas.  When they saw
> the output without commas they decided they liked it and to leave it that
> way.  This has been documented as "a case of a good bug".

IIRC John McCarthy's original paper, Recursive Functions of Symbolic 
Expressions, has an example of a symbol with spaces.   I don't
have the paper to hand, but the example given was something like
APPLE PIE NUMBER 3.   This would not have been handled correctly
unless commas were treated as delimiters.

> William P. Vrotney - ·······@netcom.com

-- 
Le Hibou (mo bheachd fhe/in: my own opinion)
"What the ... This is Lambic!   Where's my culture of amoebic 
dysentery?"
			-- Gary Larson
From: William Paul Vrotney
Subject: Re: In mourning after LISP chuck-out session!
Date: 
Message-ID: <vrotneyEJ83r8.MIx@netcom.com>
In article <·············@bt-sys.bt.spamblock.co.uk> Donald Fisk <···········@bt-sys.bt.spamblock.co.uk> writes:
> 
> William Paul Vrotney wrote:
> > Just an aside.  The first Lisp was supposed to have commas between arguments
> > and the programmer had a bug that failed to print the commas.  When they saw
> > the output without commas they decided they liked it and to leave it that
> > way.  This has been documented as "a case of a good bug".
> 
> IIRC John McCarthy's original paper, Recursive Functions of Symbolic 
> Expressions, has an example of a symbol with spaces.   I don't
> have the paper to hand, but the example given was something like
> APPLE PIE NUMBER 3.   This would not have been handled correctly
> unless commas were treated as delimiters.
> 

Of course today when JM uses Common Lisp he can type

        (setq |APPLE PIE NUMBER 3| '|APPLE PIE NUMBER 3|)

and we're back to not needing commas as delimiters even for that.  So the
"case of a good bug" still stands. :-)




-- 

William P. Vrotney - ·······@netcom.com
From: Geoffrey Cuttle
Subject: Re: In mourning after LISP chuck-out session!
Date: 
Message-ID: <memo.19971107084248.45873B@gcuttle.compulink.co.uk>
In article <·············@bt-sys.bt.spamblock.co.uk>, ···········@bt-sys.bt.spamblock.co.uk
(Donald Fisk) wrote:

> William Paul Vrotney wrote:
> > I assume that the original Lisp was written in
>  assembly language in which
> > case you would have to find a running IBM 704
> computer.
>
> Or an emulator for the 704, which could be writt
> en as long as a
> specification of the instruction set is still av
> ailable (which
> I presume it is).
>
> > Just an aside.  The first Lisp was supposed to
>  have commas between arguments
> > and the programmer had a bug that failed to pr
> int the commas.  When they saw
> > the output without commas they decided they li
> ked it and to leave it that
> > way.  This has been documented as "a case of a
>  good bug".
>
> IIRC John McCarthy's original paper, Recursive F
> unctions of Symbolic
> Expressions, has an example of a symbol with spa
> ces.   I don't
> have the paper to hand, but the example given wa
> s something like
> APPLE PIE NUMBER 3.   This would not have been h
> andled correctly
> unless commas were treated as delimiters.
>
> > William P. Vrotney - ·······@netcom.com
>
> --
> Le Hibou (mo bheachd fhe/in: my own opinion)
> "What the ... This is Lambic!   Where's my cultu
> re of amoebic
> dysentery?"
> 			-- Gary Larson
>
I have unearthed my original copy of Recursive Functions etc,  and it confirms Donald Fisk's memory.
The example he quotes reads more fully:

"For atomic symbols, we shall use strings of capital Latin letters and digits with single embedded blanks.
Examples of atomic symbols are
	A
	ABA
	APPLE PIE NUMBER 3
There .is a twofold reason for departing from the usual mathematical practice of using single letters for
atomic symbols.  First, computer programs frequently require hundreds of distinguishable symbols that
must be formed from the 47 characters that are printable by the IBM 704 computer. Second, it is
convenient to allow English words and phrases to stand for atomic entities for mnemonic reasons,  The
symbols are atomic in the sense that any substructure they may have as sequences of characters is
ignored.  We assume only that different symbols can be distinguished."

McCarthy is also quite explicit about the use of commas at that time to separate elements in a list
(S-expression).  After having introduced the dot notation in its familiar form,  with examples like:

	((AB.C).D)

he goes on to state:

"An S-expression is then simply an ordered pair,  the terms of which may be atomic symbols or simpler
S-expressions.  We can represent a list of arbitrary length in terms of S-expressions as follows.  The list
	(m1,m2, ..., mn)
is represented by the S-expression
	(m1.(m2.(...(mn.NIL)...)))
Her NIL is an atomic symbol used to terminate lists."

And there are many eexamples later of list structures,  all with comma delimiters.

At first glance however (not having looked at it for many years) the paper is confusing because much of
it is written in the metalanguage (I guess borrowed from Newell,  Shaw & Simon,  and maybe also
Perlis) used to describe functions,  with square brackets and semicolons as delimiters and functions in
lower case as distinct to atomic symbols in upper case.  For example:

"3. pair[x;y].  This function gives the list of pairs of corresponding elements of the lists x and y. We have

pair[x;y] = [null[x]^null[y]->NIL;~atom[x]^~atom[y]->cons[list[car[x];car[y]]; pair[cdr[x];cdr[y]]]]

An example is

pair[(A,B,C);(X,(Y,Z),U)] = ((A,X),(B,(Y,Z)),(C,U))"

Sadly I cannot find any good intermediate papers to show when the comma vanished,  nor to support
the benevolent bug theory though it sounds highly plausible. That's how printers were in those days.
Nor did I get much chance to use LISP then as the only machine in the UK powerful enough to run it
sensibly was in GCHQ at Cheltenham (Government Communications) and we were not meant to know it
existed!

Geoff Cuttle
From: Donald Fisk
Subject: Re: In mourning after LISP chuck-out session!
Date: 
Message-ID: <3462F4BB.7034@bt-sys.bt.spamblock.co.uk>
Geoffrey Cuttle wrote:

> Sadly I cannot find any good intermediate papers to show when the comma vanished,  nor to support
> the benevolent bug theory though it sounds highly plausible. That's how printers were in those days.
> Nor did I get much chance to use LISP then as the only machine in the UK powerful enough to run it
> sensibly was in GCHQ at Cheltenham (Government Communications) and we were not meant to know it
> existed!

JMcC wrote a paper on the history of Lisp + a minimal Lisp
interpreter, which I have a copy of somewhere at home.   He
explains about the commas, though he's a bit vague about dates.
Your recollection about the commas is correct -- there was a bug
which prevented them from being displayed, and they thought that
it looked neater that way.

The paper has an amusing appendix about the garbage collector.

> Geoff Cuttle

-- 
Le Hibou (mo bheachd fhe/in: my own opinion)
"What the ... This is Lambic!   Where's my culture of amoebic 
dysentery?"
			-- Gary Larson
From: Bill Birch
Subject: IBM 704 (was: In mourning after LISP chuck-out session!)
Date: 
Message-ID: <34667872.5AE0D10A@ctp.com>
Donald Fisk wrote:
> 
> William Paul Vrotney wrote:
> > I assume that the original Lisp was written in assembly language in which
> > case you would have to find a running IBM 704 computer.
> 
> Or an emulator for the 704, which could be written as long as a
> specification of the instruction set is still available (which
> I presume it is).
> 

Wouldn't it be great if someone wrote an emulator and ported LISP 1.5,
we could all run the original LISP? I think there should be a virtual
software museum which provides downloadable versions of these absoulute
classics. I visited the Computer museum in Boston once, which is a great
place to see valve machines, and so on. A virtual computer and software
museum would be very powerful I feel. There are already many 'retro'
emulators out there for classics like the ZX81. ENIAC emulator
anyone????

Bill