5 Jan
2019
5 Jan
'19
2:26 a.m.
I was given a copy of Walter Isaacson's "The Innovators: How a Group of
Hackers, Geniuses, and Geeks Created the Digital Revolution". It devotes
ten pages to Stallman and Gnu, Torvalds and Linux, even Tannebaum and
Minix, but never mentions Thompson and Ritchie. Unix is identified only
as a product from Bell Labs from which the others learned something--he
doesn't say what. I have heard also that Isaacson's "Idea Factory"
(about Bell Labs) barely mentions Unix. Is Isaacson blind, biased,
or merely brainwashed?
In the case of Steve Jobs, Isaacson tells not just that the Alto system
from Xerox inspired him, but also who its star creators were: Lampson,
Thacker and Kay. But then he stomps on them: "Once again, the greatest
innovation would come not from the people who created the breakthroughs,
but from the people who applied them usefully." While he very describes
innovation as a continuum from invention through engineering to marketing,
he seems to be more impressed by the later stages.
Or maybe he just likes to tell stories, and didn't pick up all the
good ones about Ken. Isaacson describes spacewar, arguably the first
stage of computer-game innovation, at great length. At the same time,
all he has to say about early-stage operating systems is a single
sentence that credits John McCarthy with leading a time-sharing effort
at MIT. (In my recollection, McCarthy proseletized; Corbato led.) He
tells how ARPANET, which he says was mainly developed by BB&N, connected
time-shared computers, but breathes not a word about Berkeley's work,
without which ARPANET would have been an open circuit.
"Innovators" won general critical praise. A couple of reviews predicted
it would become the standard of the field. However, an evidently
knowledgeable review in IEEE Annals of the History of Computing faulted
it for peddling familiar potted legends without really digging for
deeper insight. Regarding Thompson and Ritchie, it looks more like
overt suppression.
Doug
5 Jan
5 Jan
2:35 a.m.
Yep, it’s pretty superficial when it comes to looking at where we are today.
Further, the puppy love over RMS is entirely unjustified. He was in the right place with
a rant about the industry but he’s oft unduly credited by a lot of the early GNU hangers
on like Len Tower who made the project a success in spite
of RMS.
If anybody truly knew RMS they’d not tolerate any of this. He’s the most odiferous,
objectionable, sexist, pedophile I have ever met (though I’ve not met the President yet).
3:31 p.m.
+1. RMS always talked big but the real work was done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
On Fri, Jan 04, 2019 at 09:35:53PM -0500, Ronald Natalie wrote:
Yep, it???s pretty superficial when it comes to
looking at where we are today.
Further, the puppy love over RMS is entirely unjustified. He was in the right place
with a rant about the industry but he???s oft unduly credited by a lot of the early GNU
hangers on like Len Tower who made the project a success in spite
of RMS.
If anybody truly knew RMS they???d not tolerate any of this. He???s the most odiferous,
objectionable, sexist, pedophile I have ever met (though I???ve not met the President
yet).
--
---
Larry McVoy lm at mcvoy.com http://www.mcvoy.com/lm
5:01 p.m.
On Sat, Jan 5, 2019, 10:39 AM Larry McVoy <lm(a)mcvoy.com wrote:
+1. RMS always talked big but the real work was done
by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
I'm going to refrain from either praising or disparaging the man. I think
the book Hackers by Steven Levy does a good job of describing him and how
the idea for the GNU project came about. I'm going to paraphrase what I
remember from it, but it's been a long time since I read it.
If I recall correctly, stallman graduated summa cum laude with a physics
degree from Harvard. He spent many of his undergrad days and nights working
at the MIT AI lab. Among his character defects, I've never heard anyone
accuse him of being a dumb guy. Awkward, yes, strange, perhaps, but not
dumb.
At the AI lab he found a place where he fit in. The systems ran ITS, an MIT
homegrown operating system. It was an open environment where people debated
technical issues over Chinese take-out, an intellectual society in which
Stallman felt at home as a rightful citizen.
The camaraderie he knew there dissolved as its members struck out to become
entrepreneurs. My memory is fuzzy here, but I believe his main nemesis was
Symbolics, marketers of a proprietary version of MIT's CADR Lisp machine
and operating system. As they released system updates, Stallman would would
reverse engineer the changes and add the new features to the MIT system.
Around the same time, ITS was being replaced on the computers by
proprietary operating systems. Stallman began running into roadblocks, bugs
in the OS where the code was not available to fix. To access the code he
would have to sign an NDA, which he refused to do.
In short, his little utopia collapsed. The lab as he knew it was gone. He
wondered to himself whether he could rebuild it somehow, and this was the
inception of the GNU project. He chose to re-implement Unix, not because he
considered it an ideal operating system but because he considered it
adequate. (I am among those on this list who would beg to differ.) He has
said many times that he does not agree with the Unix philosophy, but I
don't know specifically what he means by this.
Building an operating system in and of itself was not so much his goal as
building the friendships and community surrounding it.
On Fri, Jan 04, 2019 at 09:35:53PM -0500, Ronald
Natalie wrote:
Yep, it???s pretty superficial when it comes to
looking at where we are
today.
Further, the puppy love over RMS is entirely
unjustified. He was in
the right place with a rant about the industry but he???s
oft unduly
credited by a lot of the early GNU hangers on like Len Tower who made the
project a success in spite
of RMS.
If anybody truly knew RMS they???d not tolerate any of this. He???s
the most
odiferous, objectionable, sexist, pedophile I have ever met
(though I???ve not met the President yet).
--
---
Larry McVoy lm at mcvoy.com
http://www.mcvoy.com/lm
8:27 p.m.
On 1/5/19, A. P. Garcia <a.phillip.garcia(a)gmail.com> wrote:
[concerning Richard Stallman]
Building an operating system in and of itself was not so much his goal as
building the friendships and community surrounding it.
The GNU Hurd kernel certainly seems to have gotten nowhere, and with
the success of Linux IMO the free software community doesn't need it
anymore. But FSF certainly has made a big impact and contribution
with the gcc toolchain and the free versions of the Unix shell and
utilities.
-Paul W.
9:38 p.m.
On Sat, Jan 5, 2019, 3:27 PM Paul Winalski <paul.winalski(a)gmail.com wrote:
On 1/5/19, A. P. Garcia
<a.phillip.garcia(a)gmail.com> wrote:
[concerning Richard Stallman]
But RMS sort of invented that community, just like Al Gore sort of invented
the internet (as we know it today). He was certainly an important catalyst,
and his views remain influential to many people.
He was wrong about Unix, though. It is not merely an adequate OS. It is
ideal. ;-)
Building an operating system in and of itself was not so much his goal as
building the friendships and community surrounding it.
The GNU Hurd kernel certainly seems to have gotten nowhere, and with
the success of Linux IMO the free software community doesn't need it
anymore. But FSF certainly has made a big impact and contribution
with the gcc toolchain and the free versions of the Unix shell and
utilities.
9:51 p.m.
I heard when "The Idea Factory" came out that Gertner left out the
computing stuff because he was planning a second book focusing on that
topic. Of course that might not be true.
-rob
9:53 p.m.
By the way, I felt that "The Idea Factory" did a great job of
communicating the feeling of working at Bell Labs Research and have
recommended it widely. Doug, you might enjoy it.
-rob
On Sun, Jan 6, 2019 at 8:51 AM Rob Pike <robpike(a)gmail.com> wrote:
I heard when "The Idea Factory" came out that Gertner left out the
computing stuff because he was planning a second book focusing on that
topic. Of course that might not be true.
-rob
9:52 p.m.
On Sat, Jan 05, 2019 at 04:38:40PM -0500, A. P. Garcia wrote:
On Sat, Jan 5, 2019, 3:27 PM Paul Winalski
<paul.winalski(a)gmail.com wrote:
I'll remind you all of a story I'm sure I shared here in the past.
I was friends with the 3 Cygnus founders and I was having a dinner
with them at Gumby's house. (This is an aside but it is important:
Gumby was, at that time, married to a very nice German woman named
Silka; she still had a pretty strong accent, English was not her
first language).
Cygnus was pretty much a 100% GPL / LGPL shop. So the discussions
were all "RMS this", "RMS that" and went on for a long time (hours).
Silka was clearing the table and she pipes up with "Are you saying
'RMS this'?" We say yes, and she responds with "Huh, all this time I
heard 'Our mess this', 'Our mess that'". We all looked at each
other
in silence, sort of replaying everying through that view. And started
laughing (and freaking out a little) that every sentence made sense as
"Our mess <whatever>".
I'm clearly not a fan of RMS, yeah, he did some good but in the process
he took credit for an enormous body of work in which he didn't write a
line of code. As a programmer, that's lieing and I can't stand liars.
And that's all I have to say on the RMS topic, it's probably too much
as it is.
On 1/5/19, A. P. Garcia
<a.phillip.garcia(a)gmail.com> wrote:
[concerning Richard Stallman]
But RMS sort of invented that community, just like Al Gore sort of invented
the internet (as we know it today). He was certainly an important catalyst,
and his views remain influential to many people.
Building an operating system in and of itself was not so much his goal as
building the friendships and community surrounding it.
The GNU Hurd kernel certainly seems to have gotten nowhere, and with
the success of Linux IMO the free software community doesn't need it
anymore. But FSF certainly has made a big impact and contribution
with the gcc toolchain and the free versions of the Unix shell and
utilities.
6 Jan
6 Jan
1:50 a.m.
On Jan 5, 2019, at 9:01 AM, A. P. Garcia <a.phillip.garcia(a)gmail.com> wrote:
On Sat, Jan 5, 2019, 10:39 AM Larry McVoy <lm(a)mcvoy.com wrote:
Dan Weinreb, who was in charge of Symbolics at the time, strongly disputed the RMS (and
“Hackers”) story of GNU’s inspiration from what Symbolics “did to” the AI Lab.
By Weinreb’s account, Symbolics hired relatively few people away from the Lab, and RMS
wasn’t simply rewriting Symbolics’ enhancements (which were shared with the Lab, and
Symbolics’ customers of course) for the MIT and LMI environments, he was actually caught
copying their code directly.
— Chris
+1. RMS always talked big but the real work was
done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
I'm going to refrain from either praising or disparaging the man. I think the book
Hackers by Steven Levy does a good job of describing him and how the idea for the GNU
project came about.
2:18 a.m.
On Sat, Jan 5, 2019, 8:50 PM Chris Hanson <cmhanson(a)eschatologist.net wrote:
On Jan 5, 2019, at 9:01 AM, A. P. Garcia
<a.phillip.garcia(a)gmail.com>
wrote:
On Sat, Jan 5, 2019, 10:39 AM Larry McVoy <lm(a)mcvoy.com wrote:
A google search turned up a speech by rms that I hadnˋt previously seen, as
well as Weinrebˋs rebuttal:
https://www.gnu.org/gnu/rms-lisp.en.html
http://ergoemacs.org/misc/Daniel_Weinreb_rebuttal_to_rms.html
+1. RMS always talked big but the real work was
done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
I'm going to refrain from either praising or disparaging the man. I think
the book Hackers by Steven Levy does a good job of describing him and how
the idea for the GNU project came about.
Dan Weinreb, who was in charge of Symbolics at the time, strongly disputed
the RMS (and “Hackers”) story of GNU’s inspiration from what Symbolics “did
to” the AI Lab.
By Weinreb’s account, Symbolics hired relatively few people away from the
Lab, and RMS wasn’t simply rewriting Symbolics’ enhancements (which were
shared with the Lab, and Symbolics’ customers of course) for the MIT and
LMI environments, he was actually caught copying their code directly.
5 Jan
5 Jan
5:07 p.m.
At 5 Jan 2019 15:40:13 +0000 (+00:00) from Larry McVoy <lm(a)mcvoy.com>om>:
+1. RMS always talked big but the real work was done
by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
RMS hired a programmer for
emacs already in the 80ths. He also became the maintainer then. It was the dude lucid
hired for xemacs. After he skilled jwz at. al. they fired him. Tragical case.
9:20 p.m.
Stig?
Joe McGuckin
ViaNet Communications
joe(a)via.net
650-207-0372 cell
650-213-1302 office
650-969-2124 fax
On Jan 5, 2019, at 9:07 AM, Donald ODona
<mutiny.mutiny(a)india.com> wrote:
At 5 Jan 2019 15:40:13 +0000 (+00:00) from Larry McVoy <lm(a)mcvoy.com>om>:
+1. RMS always talked big but the real work was
done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
RMS hired a programmer for
emacs already in the 80ths. He also became the maintainer then. It was the dude lucid
hired for xemacs. After he skilled jwz at. al. they fired him. Tragical case.
6 Jan
6 Jan
1:43 a.m.
On Jan 5, 2019, at 7:31 AM, Larry McVoy <lm(a)mcvoy.com> wrote:
+1. RMS always talked big but the real work was done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
Which I thought was originally derived from Unipress emacs (Gosmacs), and was why old
source code used to be hard to find.
— Chris
2:40 a.m.
New subject: Emacs (was: Isaacson v Unix)
On Saturday, 5 January 2019 at 17:43:43 -0800, Chris Hanson wrote:
On Jan 5, 2019, at 7:31 AM, Larry McVoy
<lm(a)mcvoy.com> wrote:
I don't think there's any serious doubt that rms wrote the original
Emacs, in TECO. If I understand it correctly, though, he took
significant improvements, including screen redisplay, from Gosling
Emacs.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
+1. RMS always talked big but the real work was done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
Which I thought was originally derived from Unipress emacs
(Gosmacs), and was why old source code used to be hard to find.
6:01 a.m.
New subject: Emacs (was: Isaacson v Unix)
teco, a very early (even earlier than qed/ed) character based editor, rather a stream
editor in unix terms, was written by the student dan murphy for the pdp-1 in around 1964.
In the 70ths two teco macro collections were popular in MIT's AI lab:
tmacs&temacs, written by gus steele et al., leaving both packages unmaintained. rms
took over maintenance, consolidating and improving them. That's all. He neither
written teco nor the teco macro packages.
At 6 Jan 2019 02:51:32 +0000 (+00:00) from Greg 'groggy' Lehey
<grog(a)lemis.com>om>:
On Saturday, 5 January 2019 at 17:43:43 -0800, Chris
Hanson wrote:
On Jan 5, 2019, at 7:31 AM, Larry McVoy
<lm(a)mcvoy.com> wrote:
I don't think there's any serious doubt that rms wrote the original
Emacs, in TECO. If I understand it correctly, though, he took
significant improvements, including screen redisplay, from Gosling
Emacs.
Greg
--
Sent from my desktop computer.
Finger grog(a)lemis.com for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft mail program
reports problems, please read <a target="_blank"
href="http://lemis.com/broken-MUA">http://lemis.com/broken-MUA…
+1. RMS always talked big but the real work was done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
Which I thought was originally derived from Unipress emacs
(Gosmacs), and was why old source code used to be hard to find.
6:03 a.m.
New subject: Emacs (was: Isaacson v Unix)
This is a distraction, but didn't the BDS C guy take some of those
interfaces into his editor?
On Sun, Jan 06, 2019 at 06:01:44AM +0000, Donald ODona wrote:
teco, a very early (even earlier than qed/ed)
character based editor, rather a stream editor in unix terms, was written by the student
dan murphy for the pdp-1 in around 1964. In the 70ths two teco macro collections were
popular in MIT's AI lab: tmacs&temacs, written by gus steele et al., leaving both
packages unmaintained. rms took over maintenance, consolidating and improving them.
That's all. He neither written teco nor the teco macro packages.
At 6 Jan 2019 02:51:32 +0000 (+00:00) from Greg 'groggy' Lehey
<grog(a)lemis.com>om>:
> On Saturday, 5 January 2019 at 17:43:43 -0800, Chris Hanson wrote:
> > On Jan 5, 2019, at 7:31 AM, Larry McVoy <lm(a)mcvoy.com> wrote:
> >>
> >> +1. RMS always talked big but the real work was done by other people.
> >> GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
> >> etc. I think RMS hacked on emacs but not much else.
> >
> > Which I thought was originally derived from Unipress emacs
> > (Gosmacs), and was why old source code used to be hard to find.
>
> I don't think there's any serious doubt that rms wrote the original
> Emacs, in TECO. If I understand it correctly, though, he took
> significant improvements, including screen redisplay, from Gosling
> Emacs.
>
> Greg
> --
> Sent from my desktop computer.
> Finger grog(a)lemis.com for PGP public key.
> See complete headers for address and phone numbers.
> This message is digitally signed. If your Microsoft mail program
> reports problems, please read <a target="_blank"
href="http://lemis.com/broken-MUA">http://lemis.com/broken-MUA…
--
---
Larry McVoy lm at mcvoy.com http://www.mcvoy.com/lm
6:26 a.m.
New subject: Emacs (was: Isaacson v Unix)
On Jan 5, 2019, at 6:40 PM, Greg 'groggy' Lehey <grog(a)lemis.com> wrote:
On Saturday, 5 January 2019 at 17:43:43 -0800, Chris Hanson wrote:
That’s not what I’m referring to, I’m referring to GNU emacs.
(Also, he didn’t write the original emacs. He took it over.)
On Jan 5, 2019, at 7:31 AM, Larry McVoy
<lm(a)mcvoy.com> wrote:
I don't think there's any serious doubt that rms wrote the original
Emacs, in TECO.
+1. RMS always talked big but the real work was done by other people.
GCC was Tiemann at Sun and then at Cygnus, groff was James Clark,
etc. I think RMS hacked on emacs but not much else.
Which I thought was originally derived from Unipress emacs
(Gosmacs), and was why old source code used to be hard to find. If I understand it correctly, though, he took
significant improvements, including screen redisplay, from Gosling
Emacs.
GNU emacs is not a descendant of the TECO package. What I’m referring to is that GNU emacs
started as a set of hacks on Gosmacs, and the Gosmacs code had to be excised because it
wasn’t actually something FSF could redistribute as their own.
-- Chris
7:50 a.m.
New subject: Emacs
Greg 'groggy' Lehey wrote:
Also correct. I found a copy of GNU Emacs 13.8 which I believe is the
first version to be distributed. It has the Gosling display code.
I don't think there's any serious doubt that
rms wrote the original
Emacs, in TECO.
That's an oversimplification. RMS may have done most of the work, but
he was not the first. Here's some interesting reading from 1978:
https://github.com/PDP-10/its/blob/master/doc/eak/emacs.lore
Chris Hanson wrote:
(Also, he didn’t write the original emacs. He took it
over.)
That's right.
RMS is often credited with the ^R real-time display feature, but that
too originated earlier by Mikkelsen.
If I
understand it correctly, though, he took significant
improvements, including screen redisplay, from Gosling Emacs.
GNU emacs is not a descendant of the TECO package. What I’m referring
to is that GNU emacs started as a set of hacks on Gosmacs, and the
Gosmacs code had to be excised because it wasn’t actually something
FSF could redistribute as their own. 
2:23 p.m.
New subject: Emacs
On 06/01/2019 07:50, Lars Brinkhoff wrote:
I found a copy of GNU Emacs 13.8 which I believe is
the
first version to be distributed. It has the Gosling display code.
Wow! Where did you find this? Online?
What OS does it run on - ITS? Is it available for something like Unix?
Andrew
--
OpenPGP key: EB28 0338 28B7 19DA DAB0 B193 D21D 996E 883B E5B9
2:30 p.m.
New subject: Emacs
Andrew Luke Nesbit wrote:
Yes, on a DECUS tape with VMS software. I'm not sure it runs on VMS,
though.
This link sometimes works, sometimes doesn't. But I have it mirrored.
http://decuslib.com/decus/vax85b/gnuemax/emacs/
I found a copy
of GNU Emacs 13.8 which I believe is the first version
to be distributed. It has the Gosling display code.
Wow! Where did you find
this? Online? What OS does it run on - ITS? Is it available for
something like Unix?
I haven't tried to compile it, but my guess would be 4.2 ish BSD.
Whatever BSD was contemporary around 1985 I suppose.
I did try 16.56 on 4.1BSD, but it wasn't an immediate success. Some
things seemed to be missing from BSD.
4.3BSD came with a copy of GNU Emacs 17.61.
5 Jan
5 Jan
9:08 a.m.
Okay, I will say it: Fake News.
The irony is that UNIX through programs like troff/nroff maintained extensive and
elaborate mechanisms to support citations. Indeed, there was a time when research
required meticulous care of citations.
In fact, I would argue that the importance of the great breakthrough UNIX papers by
Thompson et al. in combination with the many excellent cites within are as significant of
a work product as the UNIX proper.
I used to think that the greatest attribute of a digital document was its inability to
suffer decay, degrade or otherwise fade away like its analog analog.
Yet, digital document decay occurs indirectly as we can see here with Isaacson’s work
product: Revisionism by omission.
It’s as simple as it is dangerous. Since newer is ALWAYS better, the classic texts will
eventually disappear replaced by garbage paradoxically “filled” with omissions.
Bill Corcoran
“One small step for main; one giant leap for mainkind.”
On Jan 4, 2019, at 9:27 PM, Doug McIlroy
<doug(a)cs.dartmouth.edu> wrote:
I was given a copy of Walter Isaacson's "The Innovators: How a Group of
Hackers, Geniuses, and Geeks Created the Digital Revolution". It devotes
ten pages to Stallman and Gnu, Torvalds and Linux, even Tannebaum and
Minix, but never mentions Thompson and Ritchie. Unix is identified only
as a product from Bell Labs from which the others learned something--he
doesn't say what. I have heard also that Isaacson's "Idea Factory"
(about Bell Labs) barely mentions Unix. Is Isaacson blind, biased,
or merely brainwashed?
In the case of Steve Jobs, Isaacson tells not just that the Alto system
from Xerox inspired him, but also who its star creators were: Lampson,
Thacker and Kay. But then he stomps on them: "Once again, the greatest
innovation would come not from the people who created the breakthroughs,
but from the people who applied them usefully." While he very describes
innovation as a continuum from invention through engineering to marketing,
he seems to be more impressed by the later stages.
Or maybe he just likes to tell stories, and didn't pick up all the
good ones about Ken. Isaacson describes spacewar, arguably the first
stage of computer-game innovation, at great length. At the same time,
all he has to say about early-stage operating systems is a single
sentence that credits John McCarthy with leading a time-sharing effort
at MIT. (In my recollection, McCarthy proseletized; Corbato led.) He
tells how ARPANET, which he says was mainly developed by BB&N, connected
time-shared computers, but breathes not a word about Berkeley's work,
without which ARPANET would have been an open circuit.
"Innovators" won general critical praise. A couple of reviews predicted
it would become the standard of the field. However, an evidently
knowledgeable review in IEEE Annals of the History of Computing faulted
it for peddling familiar potted legends without really digging for
deeper insight. Regarding Thompson and Ritchie, it looks more like
overt suppression.
Doug
12:09 p.m.
Nonsense from someone who evidently had some sort of axe to grind.
Kerninghan, Ritchie, Pine, Thonpson, Joy, and the rest being excluded
is like crediting everything that Sun has done to Scott McNealy.
Sounds like the marketers being credited with innovation.
On 1/5/19, William Corcoran <wlc(a)jctaylor.com> wrote:
Okay, I will say it: Fake News.
The irony is that UNIX through programs like troff/nroff maintained
extensive and elaborate mechanisms to support citations. Indeed, there was
a time when research required meticulous care of citations.
In fact, I would argue that the importance of the great breakthrough UNIX
papers by Thompson et al. in combination with the many excellent cites
within are as significant of a work product as the UNIX proper.
I used to think that the greatest attribute of a digital document was its
inability to suffer decay, degrade or otherwise fade away like its analog
analog.
Yet, digital document decay occurs indirectly as we can see here with
Isaacson’s work product: Revisionism by omission.
It’s as simple as it is dangerous. Since newer is ALWAYS better, the
classic texts will eventually disappear replaced by garbage paradoxically
“filled” with omissions.
Bill Corcoran
“One small step for main; one giant leap for mainkind.”
On Jan 4, 2019, at 9:27 PM, Doug McIlroy
<doug(a)cs.dartmouth.edu> wrote:
I was given a copy of Walter Isaacson's "The Innovators: How a Group of
Hackers, Geniuses, and Geeks Created the Digital Revolution". It devotes
ten pages to Stallman and Gnu, Torvalds and Linux, even Tannebaum and
Minix, but never mentions Thompson and Ritchie. Unix is identified only
as a product from Bell Labs from which the others learned something--he
doesn't say what. I have heard also that Isaacson's "Idea Factory"
(about Bell Labs) barely mentions Unix. Is Isaacson blind, biased,
or merely brainwashed?
In the case of Steve Jobs, Isaacson tells not just that the Alto system
from Xerox inspired him, but also who its star creators were: Lampson,
Thacker and Kay. But then he stomps on them: "Once again, the greatest
innovation would come not from the people who created the breakthroughs,
but from the people who applied them usefully." While he very describes
innovation as a continuum from invention through engineering to
marketing,
he seems to be more impressed by the later stages.
Or maybe he just likes to tell stories, and didn't pick up all the
good ones about Ken. Isaacson describes spacewar, arguably the first
stage of computer-game innovation, at great length. At the same time,
all he has to say about early-stage operating systems is a single
sentence that credits John McCarthy with leading a time-sharing effort
at MIT. (In my recollection, McCarthy proseletized; Corbato led.) He
tells how ARPANET, which he says was mainly developed by BB&N, connected
time-shared computers, but breathes not a word about Berkeley's work,
without which ARPANET would have been an open circuit.
"Innovators" won general critical praise. A couple of reviews predicted
it would become the standard of the field. However, an evidently
knowledgeable review in IEEE Annals of the History of Computing faulted
it for peddling familiar potted legends without really digging for
deeper insight. Regarding Thompson and Ritchie, it looks more like
overt suppression.
Doug
2:15 p.m.
On 1/4/19, Doug McIlroy <doug(a)cs.dartmouth.edu> wrote:
In the case of Steve Jobs, Isaacson tells not just that the Alto system
from Xerox inspired him, but also who its star creators were: Lampson,
Thacker and Kay. But then he stomps on them: "Once again, the greatest
innovation would come not from the people who created the breakthroughs,
but from the people who applied them usefully." While he very describes
innovation as a continuum from invention through engineering to marketing,
he seems to be more impressed by the later stages.
I would argue that Isaacson does have a point here. After Lampson
left Xerox PARC he set up a similar outfit at Digital'--the Western
Research Lab (WRL). They did a lot of interesting work in the area of
software development tools. I was working in the software tools
engineering group at the time, and we would have loved to take WRL's
work and to incorporate it in our products. But we couldn't. Why?
Because they wrote everything in Modula 3, and we were using BLISS.
It was too expensive and time-consuming to do the translation. If
they had worked in BLISS, we could have just taken their code and run
with it. From my perspective it looked as though they were
deliberately setting up barriers to prevent us from sullying their
research by actually turning it into useful products.
In one memo to DEC's engineering staff, Gordon Bell proposed a "Xerox
PARC" award to the R&D project that advanced the state-of-the-art the
most while simultaneously advancing DEC's bottom line the least.
Yes, PARC invented the modern windows-based GUI, but, as with so many
PARC innovations, Xerox did nothing with it. Based on how the PARC
alumni at WRL behaved at DEC,I would argue that this was the fault of
PARC as much as of Xerox management.
All that being said, I don't think this argument applies in any way to
Bell Labs and Unix. Unix was "applied usefully" long before Stallman
and Torvalds came along. Not crediting its inventors is inexcusable.
-Paul W.
3:01 p.m.
All that being said, I don't think this argument
applies in any way to
Bell Labs and Unix. Unix was "applied usefully" long before Stallman
and Torvalds came along. Not crediting its inventors is inexcusable.
Completely agree!
On 1/5/19, Paul Winalski <paul.winalski(a)gmail.com> wrote:
> On 1/4/19, Doug McIlroy <doug(a)cs.dartmouth.edu> wrote:
>>
>> In the case of Steve Jobs, Isaacson tells not just that the Alto system
>> from Xerox inspired him, but also who its star creators were: Lampson,
>> Thacker and Kay. But then he stomps on them: "Once again, the greatest
>> innovation would come not from the people who created the breakthroughs,
>> but from the people who applied them usefully." While he very describes
>> innovation as a continuum from invention through engineering to
>> marketing,
>> he seems to be more impressed by the later stages.
>
> I would argue that Isaacson does have a point here. After Lampson
> left Xerox PARC he set up a similar outfit at Digital'--the Western
> Research Lab (WRL). They did a lot of interesting work in the area of
> software development tools. I was working in the software tools
> engineering group at the time, and we would have loved to take WRL's
> work and to incorporate it in our products. But we couldn't. Why?
> Because they wrote everything in Modula 3, and we were using BLISS.
> It was too expensive and time-consuming to do the translation. If
> they had worked in BLISS, we could have just taken their code and run
> with it. From my perspective it looked as though they were
> deliberately setting up barriers to prevent us from sullying their
> research by actually turning it into useful products.
>
> In one memo to DEC's engineering staff, Gordon Bell proposed a "Xerox
> PARC" award to the R&D project that advanced the state-of-the-art the
> most while simultaneously advancing DEC's bottom line the least.
>
> Yes, PARC invented the modern windows-based GUI, but, as with so many
> PARC innovations, Xerox did nothing with it. Based on how the PARC
> alumni at WRL behaved at DEC,I would argue that this was the fault of
> PARC as much as of Xerox management.
>
All that being said, I don't think this argument
applies in any way to
Bell Labs and Unix. Unix was "applied usefully" long before Stallman
and Torvalds came along. Not crediting its inventors is inexcusable.
>
> -Paul W.
>
6 Jan
6 Jan
4:35 a.m.
"Innovators" won general critical praise. A
couple of reviews predicted
it would become the standard of the field. However, an evidently
knowledgeable review in IEEE Annals of the History of Computing faulted
it for peddling familiar potted legends without really digging for
deeper insight. Regarding Thompson and Ritchie, it looks more like
overt suppression.
Why not ask @WalterIsaacson on twitter. He was chairman & CEO
of CNN and managing editor of Time among other things so
presumably he is/was interested in facts.
4:52 a.m.
On Sat, 05 Jan 2019 20:35:49 -0800 Bakul Shah <bakul(a)bitblocks.com> wrote:
Bakul Shah writes:
And a professor of history at Tulane. I am sure he is interested
in getting the history right, right?!
"Innovators" won general critical praise. A couple of reviews predicted
it would become the standard of the field. However, an evidently
knowledgeable review in IEEE Annals of the History of Computing faulted
it for peddling familiar potted legends without really digging for
deeper insight. Regarding Thompson and Ritchie, it looks more like
overt suppression.
Why not ask @WalterIsaacson on twitter. He was chairman & CEO
of CNN and managing editor of Time among other things so
presumably he is/was interested in facts. 
5 a.m.
On 2019-01-05 11:35 PM, Bakul Shah wrote:
Dry wit indeed.
"Innovators" won general critical praise. A couple of reviews predicted
it would become the standard of the field. However, an evidently
knowledgeable review in IEEE Annals of the History of Computing faulted
it for peddling familiar potted legends without really digging for
deeper insight. Regarding Thompson and Ritchie, it looks more like
overt suppression.
Why not ask @WalterIsaacson on twitter. He was chairman & CEO
of CNN and managing editor of Time among other things so
presumably he is/was interested in facts.
1943
days inactive
1944
days old
27 comments
16 participants
participants (16)
-
A. P. Garcia -
Andrew Luke Nesbit -
Bakul Shah -
Chris Hanson -
Donald ODona -
Doug McIlroy -
Ed Carp -
Greg 'groggy' Lehey -
joe mcguckin -
Larry McVoy -
Lars Brinkhoff -
Paul Winalski -
Rob Pike -
Ronald Natalie -
Toby Thain -
William Corcoran