Soul of a New Machine movie
Andrew McVeigh
I am in the middle of reading the excellent book by Tracy
Kidder, titled "The Soul of a New Machine". On the back
of the book it says something to the effect that a major
motion picture about this was going to be made... Was
this movie ever made?
Also, can anyone give me any info about the Eclipse line
of computers manufactured by Data General, especially the
32-bit model called the Eagle in the book. How well did
it sell compared to Dec's VAX machine?
Cheers,
Andrew McVeigh
--
Hugh JE Davies
>
>I am in the middle of reading the excellent book by Tracy
>Kidder, titled "The Soul of a New Machine". On the back
>of the book it says something to the effect that a major
>motion picture about this was going to be made... Was
>this movie ever made?
No idea.
>
>Also, can anyone give me any info about the Eclipse line
>of computers manufactured by Data General, especially the
>32-bit model called the Eagle in the book. How well did
>it sell compared to Dec's VAX machine?
If I swivel to the right from my Sun, I find myself looking at a Data
General Dasher D460 terminal connected to our "Eagle" - a DG MV30000
Model 4, quad-processor turkey. It runs a hideous distant relative of Unix
as an operating system - AOS/VS (there's a newsgroup for this, BTW,
comp.sys.aos (nothing to do with IBM's AOS). It's a competent enough
minicomputer, although the physical construction is pretty dire, with
the exception of the terminal interface cards, the so-called "Intelligent
A?????? Controller", or IAC for short. These are a real hoot, they make
DECs various stabs in the dark at RS-232 controllers look rather good
by comparison.
No, they didn't sell all that well, at least not compared to the VAX.
Regards,
Hugh.
------------------------------------------------------------------
Huge...@rx.xerox.com Rank Xerox Technical Centre, WGC, UK.
I don't speak for Xerox, nor they for me.
The road to Paradise is through Intercourse.
Anthony A. Datri
>>it sell compared to Dec's VAX machine?
I have no numbers, but the fact that you had to ask probably says something.
>Model 4, quad-processor turkey. It runs a hideous distant relative of Unix
>as an operating system - AOS/VS
That uses ( for command-line switches, right?
>A?????? Controller", or IAC for short. These are a real hoot, they make
>DECs various stabs in the dark at RS-232 controllers look rather good
>by comparison.
Something that's *worse* than a DZ11?
> The road to Paradise is through Intercourse.
Does central Pennsylvania have a lot of fans there?
--
======================================================================8--<
Rodney Radford
>If I swivel to the right from my Sun, I find myself looking at a Data
>General Dasher D460 terminal connected to our "Eagle" - a DG MV30000
>Model 4, quad-processor turkey. It runs a hideous distant relative of Unix
>as an operating system - AOS/VS (there's a newsgroup for this, BTW,
I wouldn't really consider AOS/VS a relative of Unix any more than any
OS is a relative of any other (ie: they all have some items in common).
Actually both Unix and AOS/VS were descendents from Multics, but they
both went their own seperate ways - not even trying to resemble one another.
>comp.sys.aos (nothing to do with IBM's AOS). It's a competent enough
>minicomputer, although the physical construction is pretty dire, with
>the exception of the terminal interface cards, the so-called "Intelligent
>A?????? Controller", or IAC for short. These are a real hoot, they make
>DECs various stabs in the dark at RS-232 controllers look rather good
>by comparison.
IAC - Intelligent Asyncronous Controller
>No, they didn't sell all that well, at least not compared to the VAX.
Yes, they didn't sell very well, but that was a marketing problem, not a
technical problem. As is true with may of the winners/losers in the computer
market - it depends on how good your marketing group is, not whether your
product is worth selling..... ;-}
--
---
Rodney Radford || Computer Graphics/Imaging
sas...@unx.sas.com || SAS Institute, Inc.
(919) 677-8000 x7703 || Cary, NC 27513
Terry Kennedy, Operations Mgr.
> Yes, they didn't sell very well, but that was a marketing problem, not a
> technical problem. As is true with may of the winners/losers in the computer
> market - it depends on how good your marketing group is, not whether your
> product is worth selling..... ;-}
Definitely. We wanted to continue using our Eclipse S/200 for a few years
before switching over to the MV/8000 (the machine the book is about), but DG
said they were discontinuing all 16-bit software support and if we didn't like
it, tough. So, we went to another vendor (DEC) who said they had no plans to
discontinue 16-bit software support. That was in 1981, and we've had 12 good
years of 16-bit software support from DEC. Only recently did DEC close that
end of the company down.
Terry Kennedy Operations Manager, Academic Computing
te...@spcvxa.bitnet St. Peter's College, Jersey City, NJ USA
te...@spcvxa.spc.edu +1 201 915 9381
Hugh JE Davies
>>>32-bit model called the Eagle in the book. How well did
>>>it sell compared to Dec's VAX machine?
>
>I have no numbers, but the fact that you had to ask probably says something.
I didn't ask. I answered. Your attributions are screwed up.
>
>>Model 4, quad-processor turkey. It runs a hideous distant relative of Unix
>>as an operating system - AOS/VS
>
>That uses ( for command-line switches, right?
Actually, it uses '/'.
>
>>A?????? Controller", or IAC for short. These are a real hoot, they make
>>DECs various stabs in the dark at RS-232 controllers look rather good
>>by comparison.
>
>Something that's *worse* than a DZ11?
Oh yes. Makes a DZ11 look like heaven upon earth.
>
>> The road to Paradise is through Intercourse.
>
>Does central Pennsylvania have a lot of fans there?
>
What do you mean by "there"? I am a fan of central Pennsylvania, but then again,
my parents live there.
Regards,
Hugh.
------------------------------------------------------------------
Huge...@rx.xerox.com Rank Xerox Technical Centre, WGC, UK.
I don't speak for Xerox, nor they for me.
Jim Frost
>General Dasher D460 terminal connected to our "Eagle" - a DG MV30000
>Model 4, quad-processor turkey. It runs a hideous distant relative of Unix
>as an operating system - AOS/VS (there's a newsgroup for this, BTW,
>comp.sys.aos (nothing to do with IBM's AOS).
I would be interested in hearing exactly what makes you think that
AOS/VS is in any way related to UNIX.
ObFolklore: I used to work with an MV hacker ("whacker" might be more
appropriate, actually) who decided that the UNIX fixed-count inode
structure was too limiting and reimplemented the AOS filesystem inside
a single UNIX file. I gotta give him credit, we never did figure out
how to port the damn thing. I eventually tore it all out and put in a
replacement based on ndbm for name lookup and got a monstrous speed
boost and wonderful portability in one swell foop.
(I have a lot of stories about that guy that most of you wouldn't
believe. I was never quite sure whether he was a genious or totally
demented....)
jim frost
ji...@centerline.com
Rodney Wines
> Yes, they didn't sell very well, but that was a marketing problem, not a
> technical problem. As is true with may of the winners/losers in the computer
> market - it depends on how good your marketing group is, not whether your
> product is worth selling..... ;-}
I agree.
As best I can tell, Burroughs really pioneered many ideas in modern
hardware and software design such as virtual memory. They were also
implementing their systems using a system implementation language while
everybody else was still getting by in assembly language. The problem
was, IMHO, they had the world's worst marketing people. They didn't
tell anybody what wonders they'd developed. They also would only sell
their customers FORTRAN, COBOL, and RPG.
As for the Eagle, I still think it was a turkey, even for its day,
although I've got a couple of friends that swear that AOS is far better
than VMS.
I had several friends and acquaintences who worked on the Fountainhead
Project (FHP) in North Carolina during the time of "Soul Of A New
Machine", and my impression from the outside was like what you'd get if
you handed a bunch of academics a potfull of money and said, "Design and
build me a new machine." They were quite a bunch of free spirits. I
interviewed in the Tools group once, and my interview was scheduled for
11:00. The group manager didn't show up for work in time for my
interview.
There were some really top-notch people on the project. They seemed to
want to be state of the art with every single thing about FHP, and they
designed all their hardware and software completely from scratch, and
tried to put every bell and whistle in it that any CSC thesis had ever
discussed. A more reasonable thing to have done would've been to decide
upon a product strategy, implement a marketable subset, and plan a
growth path. This'll guarantee you customer upgrades in the future.
DEC seemed to take that approach with the VAX and VMS.
Please take my opinions on FHP as just that, MHO. I was on the outside,
and my memories of the project are 20 years old or more. I'd enjoy
hearing what any of the participants have to say.
===================================================================
| Rodney Wines, Alcatel STR AG, | Phone: +41 1 465-2205 |
| Friesenbergstr. 75, CH-8055 Zurich | FAX: +41 1 465-2411 |
-------------------------------------------------------------------
| Internet: rodney...@alcatel.ch | "I always wanted roots, |
| X.400: c=CH a=arCom p=Alcatel | but if I can't have roots |
| s=Wines g=Rodney | I'll have wings." |
===================================================================
Mark Stevens
-> manufactured by Data General, especially the 32-bit model called the
-> Eagle in the book. How well did it sell compared to Dec's VAX
-> machine?
The machine was marketed as the MV8000. In years following there were
larger machines (MV10000) and smaller machines (MV4000 and MV6000's). I
was at a company that received the first few machines (we were in
desparate need for more horsepower). We had about 2 years of nightmares
getting the hardware and software to work together. The machines
weren't as successfull as DG had hoped.
* RM 1.2 00602 * RoboMail -- The next generation QWK compatible reader!
Danny R. Faught
Mark Stevens <mark.s...@factory.com> wrote:
>-> Also, can anyone give me any info about the Eclipse line of computers
>-> manufactured by Data General, especially the 32-bit model called the
>-> Eagle in the book. How well did it sell compared to Dec's VAX
>-> machine?
>
>The machine was marketed as the MV8000. In years following there were
>larger machines (MV10000) and smaller machines (MV4000 and MV6000's).
If they had named it, oh, maybe "Eagle", we wouldn't have to go
through the mapping from useful name -> meaningless jumble of letters
and numbers. A while back the engineering group here got mail from
someone in marketing, saying it was about time to think of names for
the machine we're building. It was a nice gesture, but a big mistake!
In my reply, I mentioned the fact that we already *had* a name, and a
good one, because it applies to the underlying paradigm that the
machine is built upon. Moreover, he wanted suggestions for renaming a
component of the machine from the simple descriptive name that it
already had, to an *acronym* of a more snazzy sounding name. I made
it clear that the whole idea made me nauseous.
I know that engineers can come up with some weird names. I wouldn't
mind at all letting marketing having a say from the beginning of the
design phase what the name should be, assuming that they would
continue to use the name when the product ships. But not if they
insist on some acronym and number mumbo-jumbo!
Sorry for the ranting and raving, but this really bugs me. If I'm
missing some important point on the other side of the argument, please
let me know.
--
Danny Faught -- Convex rookie -- MPP OS Test Development
"Everything is deeply intertwingled." (Ted Nelson, _Computer Lib_)
Joe Presley
+machine is built upon. Moreover, he wanted suggestions for renaming a
+component of the machine from the simple descriptive name that it
+already had, to an *acronym* of a more snazzy sounding name. I made
+it clear that the whole idea made me nauseous.
This certainly is APPALLING (Acronym Production, Particularly At Lavish
Levels, Is No Good). It's amazing the contortions that some people go
to create acronyms. Imagine every car name having to be an acronym!
--
======================
Joe Presley
j.h.p...@att.com
======================