mchinni@PICA.ARMY.MIL ("Michael J. Chinni, SMCAR-CCS-E") (11/30/89)
[encore!pinocchio!cook@cs.bu.edu] (Dale C. Cook) recently posted: > Again, I want to start by stating that I have about as much say so here > as Joe the janitor, but I think this posting does the company an injustice > and I'm going to speak up! > > Yes, NP2, the follow on to NP1, was cancelled. > > When Encore bought Gould/SEL it was obvious that some drastic action had > to be taken to stauch the financial bleeding that had brought down Gould. > Further, it had to be taken quickly or there wasn't going to be any more > Gould or Encore! > > The NP line, though technically successful, was loosing money. A lot of > money. The decision to kill off that line was a BUSINESS one. We are > a business. We have stockholders. We can't do a lot of research "for > the good of all mankind" without funding. It just doesn't work that > way in capitalism. What makes you think that Gould was in trouble and was "brought down", and that the NP line was loosing a lot of money. The story I heard (and read in a UNIX trade publication) was the Gould, Inc. was bought by a Japanese Mining Corp.. Since Gould Inc., Computer Systems Division (CSD) had several contracts in place at the time with the DoD, the sale required DoD approval. DoD said, the sale would be approved -IF- CSD was sold off to a U.S.-owned company. Encore then bought CSD. { My opinions start here } Encore then found itself with two competeing UNIX lines - their's and CSD's. Therefore, since their's was their's and CSD's was the former competitor, CSD's had to go. I get this opinion from: - talks with CSD service reps. who mentioned that no further upgrades (other than bug fixes) are to be done for the CSD UNIX line (PowerNodes and NP/1s) - talk in this list with mention made of further upgrades for the Encore UNIX line. Assuming that the basis for my opinions is true (not neccessarily a vaild assumption). My conclusion is that the decision to close down the CSD UNIX line may have been made in part for business reasons, but that the other part was the opportunity to kill off a competitor. This may sound harsh, but has anyone ever heard of a company buying a competitor, and then replacing it's own established product line with the line of its new acquisition (and former competitor). - Standard Disclamer about me not speaking for my employers - /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ Michael J. Chinni Chief Scientist, Simulation Techniques and Workplace Automation Team US Army Armament Research, Development, and Engineering Center User to skeleton sitting at cobweb () Picatinny Arsenal, New Jersey and dust covered workstation () ARPA: mchinni@pica.army.mil "System been down long?" () UUCP: ...!uunet!pica.army.mil!mchinni /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
art@bu-cs.BU.EDU (Al Thompson) (11/30/89)
In article <8911291654.aa28195@CC2.PICA.ARMY.MIL> mchinni@PICA.ARMY.MIL ("Michael J. Chinni, SMCAR-CCS-E") writes: >[...] >What makes you think that Gould was in trouble and was "brought down", and that >the NP line was loosing a lot of money. The story I heard (and read in a UNIX >trade publication) was the Gould, Inc. was bought by a Japanese Mining Corp.. >Since Gould Inc., Computer Systems Division (CSD) had several contracts in >place at the time with the DoD, the sale required DoD approval. DoD said, the >sale would be approved -IF- CSD was sold off to a U.S.-owned company. Encore >then bought CSD. > >{ My opinions start here } >Encore then found itself with two competeing UNIX lines - their's and CSD's. >Therefore, since their's was their's and CSD's was the former competitor, CSD's >had to go. I get this opinion from: > - talks with CSD service reps. who mentioned that no further upgrades > (other than bug fixes) are to be done for the CSD UNIX line > (PowerNodes and NP/1s) > - talk in this list with mention made of further upgrades for the Encore > UNIX line. > >Assuming that the basis for my opinions is true (not neccessarily a vaild >assumption). My conclusion is that the decision to close down the CSD UNIX line >may have been made in part for business reasons, but that the other part was >the opportunity to kill off a competitor. > >This may sound harsh, but has anyone ever heard of a company buying a >competitor, and then replacing it's own established product line with the line >of its new acquisition (and former competitor). I attended the Encore annual meeting and Encore's CEO, Ken Fischer addressed this very point. He said that the Gould product line was indeed losing money and "something" has to be done. Fischer did not indicate that he considered CSD to be a competitor, rather he felt it opened a potential new line of business for Encore.
clp@beartrk.beartrack.com (Charlie Pilzer) (12/01/89)
In article <8911291654.aa28195@CC2.PICA.ARMY.MIL> mchinni@PICA.ARMY.MIL ("Michael J. Chinni, SMCAR-CCS-E") writes: >>[...] >>What makes you think that Gould was in trouble and was "brought down", and that >>the NP line was loosing a lot of money. ... I think the issue is more complicated than just killing off a machine. CPL machines 6xxx and 9xxx machines will continue to exist. The only trick is that the emphasis for these machines are real time as in "MPX". MPX is Gould's (Encore's) real time O/S. Most of the programming used to be done in FORTRAN, now C and its a real memory system, no virtual stuff for MPX. It runs in simulators, high speed data channels (telemetry) and other areas where fast, consistent response is required. As I understand it, the goal is to migrate all existing Unix customers from PowerNodes to Multimaxes. I can't say that I blame Encore. Currently, they have to support: UMAX 4.2 - BSD style UMAX UMAX V - System V style UMAX UMAX Mach or whatever - UMAX based on MACH Gould/CSD UTX 2.1 - Gould's BSD/ATT unix on PowerNodes UTC 3.0 - Gould's BSD/ATT Unix on NP/1 On top of this are various other packages (Database, Languages, Comm, O/A) that Encore and Gould both supported. Its too much! To port Sys V 4.0, for example, requires (3) separate ports (Multimax, PowerNode & NP/1). Considering that NP/2 and NP/1 were not compatible (directly anyway), its no wonder it got cancelled. As for existing PowerNode (and I guess NP/1) installations, getting rev locked isn't all that terrible if you are a commercial/production client. In the case of research or education institutions, it could be a problem. The machines themselves will be supported for a number of years. There are existing contracts that must be fulfilled (some of them are government). There may be no hardware upgrades but current equipment will be maintained. I think the same goes for software. No new releases but there will be bug fixes. I support a number of PowerNode installations and I've worked with UTX since Version 1.1. There were a lot of really good things about UTX. The PowerNodes had really good price/performance in their time (1984-1987). There were some problems in getting the new hardware (NPL) out the door but from what I've heard they are solid, dependable beasts. THe saddest part of the Gould story wasn't the buyout. That's pretty positive. No, it was watching a company fall apart. Many of the best people in Gould operations were laid off or got nervous and left. Morale became terrible as the survivors just waited for another axe to fall. Development teams and projects were dismantled. And this happened before Encore bought the remnants of Gould. I don't know what's going to happen now. Encore has been loosing money since acquiring Gould/CSD. It must turn around or there will be no support for any of the machines. Hard business decisions were made and the costs were not negligable. People were laid off and years of work were terminated. There are more costs ahead, both for Encore and for Encore's customers. At some point, the Gould Unix customer base will be moved off of the PowerNodes. Whether the customers move to a Multimax remains to be seen. Once you are moving to a different hardware platform, the reasons to stay with a given vendor decrease. Encore may wind up loosing customers as the customers decommission their PowerNodes. I would hope that this group takes up the cause of the PowerNodes and NP/1 (Ever notice that NP is PN backwards :-) as well as continued discussion of Multimax. There is no longer any "Gould/CSD", its all Encore. Please note that I do not speak for Encore or Gould. I work with their equipment on behalf of my clients. Part of the above is observation, part is speculation. Charlie Pilzer clp@beartrk.beartrack.com
cook@pinocchio.Encore.COM (Dale C. Cook) (12/01/89)
[mchinni@PICA.ARMY.MIL ("Michael J. Chinni, SMCAR-CCS-E") recently posted that:
|
|What makes you think that Gould was in trouble
It was common knowledge both within Gould and "on the streets" that it was
"on the blocks" (for sale) for a period of about a year before we bought
it. Many potential suitors looked it over.
Also, most of the people I have talked with who lived through the Gould ->
Nippon Mining days volunteered the view that "we were in trouble". Where
do you get information to the contrary?
| and was "brought down", and that the NP line was loosing a lot of money.
That was the explanation given by Ken Fisher when the merger was explained
internally. I have no reason to not believe that was the case. I'd assume
that Mr Fisher studied the bucks pretty closely before doing the deal and
can't think of any reason why he would misrepresent them to us internally.
Again, where's your data from?
| The story I heard (and read in a UNIX
|trade publication) was the Gould, Inc. was bought by a Japanese Mining Corp..
That's accurate. It's also true that Nippon Mining was having trouble
selling to DoD who don't like doing (much) business with foreign controlled
companys. Their management currently holds a minority position on our
board of directors.
|Since Gould Inc., Computer Systems Division (CSD) had several contracts in
|place at the time with the DoD, the sale required DoD approval. DoD said, the
|sale would be approved -IF- CSD was sold off to a U.S.-owned company. Encore
|then bought CSD.
|
The deal certainly required federal approval (I think all big mergers do,
if nothing else to make sure there is no monopolistic advantage to the new
company.) I doubt very seriously if Dod or anyone else made any stipulations
on how Encore was to manage the new company. That's a pretty bizarre
claim. Can you back it up with data?
|{ My opinions start here }
Some of us think they may have already begun!
|Encore then found itself with two competeing UNIX lines - their's and CSD's.
|Therefore, since their's was their's and CSD's was the former competitor,
| CSD's had to go. I get this opinion from:
| - talks with CSD service reps. who mentioned that no further upgrades
| (other than bug fixes) are to be done for the CSD UNIX line
| (PowerNodes and NP/1s)
| - talk in this list with mention made of further upgrades for the
| Encore UNIX line.
|
Well opinions are like noses: (and other parts of the body) everyone has
one! Mine is that if there was some good business reason to have multiple
UNIX products, we would surely do that. Why wouldn't we? DEC now offers
ULTRIX in direct competition with VMS. Why? Because there are bucks to
be made there. I think the decision came down to the fact that there was
not room (financially) in the new company for both lines and Multimax was
chosen because of its leading edge multiprocessing technology (a decision
recently validated by OSF's selection of our Mach operting system upon
which to base OSF1.) It sure seems straightforward to me.
|Assuming that the basis for my opinions is true (not neccessarily a vaild
|assumption). My conclusion is that the decision to close down the CSD UNIX
| line may have been made in part for business reasons, but that the other
| part was the opportunity to kill off a competitor.
|
I think you have a pretty active imagination if you think large corporations
think like that! Look at Chrysler/AMC. They kept the lines that looked
promising and killed the rest. There's not a lot of room for cops and
robbers in modern American business. You have heard of "the bottom line"?
|This may sound harsh, but has anyone ever heard of a company buying a
|competitor, and then replacing it's own established product line with the line
|of its new acquisition (and former competitor).
|
Hogwash! Many mergers are done for the specific purpose of ACQUIRING a
product to market. Sometimes the new company phases out both sides old
products to produce a completely new line. This deal was done so that
a small company (Old Encore) could aquire an old and fading company's
world-wide sales, service and marketing organizations to sell (eventually)
new products to the benefit of both. Come back and talk to us in a
year if you still think this is some kind of plot to kill NP!
|- Standard Disclamer about me not speaking for my employers -
|
God I hope you realize that I am not either! Again, I'm just a tiny cog
in the vast machine called Encore; I just happen to be a very loud-mouthed,
partisan and opinionated one!
| Michael J. Chinni
- Dale (N1US) Encore Computer Corporation, Marlborough, Mass.
INTERNET: cook@encore.com "In the carriages of the past you can't
UUCP: buita \ go anywhere." -- Maxim Gorkey
talcott } !encore!cook
bellcore /cook@pinocchio.Encore.COM (Dale C. Cook) (12/01/89)
[clp@beartrk.beartrack.com (Charlie Pilzer) recently posted that: | [Largely accurate discussion of the many machines, operating systems and systems now supported by Encore.] | |As for existing PowerNode (and I guess NP/1) installations, getting rev locked |isn't all that terrible if you are a commercial/production client. In the |case of research or education institutions, it could be a problem. The |machines themselves will be supported for a number of years. There are |existing contracts that must be fulfilled (some of them are government). |There may be no hardware upgrades but current equipment will be maintained. |I think the same goes for software. No new releases but there will be bug fixes. | Gould has an excellent reputation for supporting every product ever made and from everything I've heard the new management has no plans to change that. Also note that Ken Fisher, CEO, recently preannounced (what the hey is that?) a major new machine, dubbed Seahawk, in the real-time, Concept, MPX arena. | |THe saddest part of the Gould story wasn't the buyout. | That's pretty positive. |No, it was watching a company fall apart. Many of the best people in Gould |operations were laid off or got nervous and left. Morale became terrible as |the survivors just waited for another axe to fall. Development teams and |projects were dismantled. And this happened before Encore bought the remnants |of Gould. | And a major positive thing is watching the new teams from Massachusetts and Florida get together and make plans to go forward together building on the strengths each brings to the table. Yes, a lot of good people were lost. But, yes, there remain a lot of good people dedicated to turning the new Encore around and making excellent product. |I don't know what's going to happen now. Encore has been loosing money since |acquiring Gould/CSD. It must turn around or there will be no support for any |of the machines. Hard business decisions were made and the costs were not |negligable. There's a very cautious optimism bubbling up about this quarter and down right bullishness for next year. Ken's a pretty seasoned businessman and quite a lot of us believe he will get the financials back on track and lead us to bigger and better things. |a given vendor decrease. Encore may wind up loosing customers as the |customers decommission their PowerNodes. | (Obviously) we will do everything in our power to keep ALL of our customers! Sometimes those powers are limited by resources and require your patience. | |Charlie Pilzer |clp@beartrk.beartrack.com - Dale (N1US) Encore Computer Corporation, Marlborough, Mass. INTERNET: cook@encore.com "In the carriages of the past you can't UUCP: buita \ go anywhere." -- Maxim Gorkey talcott } !encore!cook bellcore /
zawada@EN.ECN.PURDUE.EDU (Paul J Zawada) (12/02/89)
clp@beartrk.beartrack.com (Charlie Pilzer): > > Considering that NP/2 and NP/1 were not compatible (directly anyway), its > no wonder it got cancelled. The NP/1 and NP/2 were binary compatible.. Anything compiled on an NP/1 was supposed to be able to run on an NP/2. > > I think the same goes for software. No new releases but there will be bug fixes. Some of the fixes from Purdue aren't being widely distributed though. (A little trivia.. Did you know that all NP/1s without a patch written here last June will crash 24 days, 20 hours and 31 minutes after being booted up? The UIOM hardware clock goes negative!) New bugs really aren't being actively pursued either.. at least here... > > I support a number of PowerNode installations and I've worked with UTX since > Version 1.1. There were a lot of really good things about UTX. The > PowerNodes had really good price/performance in their time (1984-1987). There > were some problems in getting the new hardware (NPL) out the door but from > what I've heard they are solid, dependable beasts. Our NP/1s run faster than our Computing Center's white elephant, the ETA-10P. Just the other night we had 180+ users on one of the NP/1s, and performance was quite good. (Of course, it's still not like the performance with only 10 people on...) > > THe saddest part of the Gould story wasn't the buyout. That's pretty positive. > No, it was watching a company fall apart. Many of the best people in Gould > How very true. Sure Encore might not have been the best company to buy out Gould/CSD, but then again who was? Nobody wanted the company for over a year. It is sad that a lot of good engineering had to go right down the tubes along with a good company. I have to apologize everyone at Encore for "lashing" out at them in my last article. I realize business decisions had to be made and someting had to be done, and I don't know the whole story. IMHO though, I think killing the NP/2 was a bad decision though. Here they had a high performance machine ready for beta testing, and they killed it. Granted, it was Gould's research $$ that they dropped down the drain, but wouldn't have been a good idea to take advantage of that research? I was so anxious to get my hands on the NP/2... I heard how great a machine it was going to be for months... and then I find out it's not worth the millions of dollars and years of research that was put into it. pjz... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Paul J Zawada | zawada@ee.ecn.purdue.edu "E-site" Student Consultant | ...!pur-ee!zawada Purdue University | Engineering Computer Network | GO BOILERS!!!