Do Not Ask For Whom The Bell Tolls
Ron
The decision by Sun Microsystems to "kill" Solaris x86 means their days
as a viable business unit are numbered. Don't get me wrong, I'm not
being alarmist and saying it's going to happen this month or this year,
but mark my words: Sun is dead if they continue to pursue the business
strategy exemplified by the decision to kill Solaris x86.
In the last few months Sun has killed the IPlanet initiative; made
decisions about Java licensing that benefit only IBM (and hurt everyone
else, including BEA and Sun); pretty much killed or started to kill all
of their Open Source initiatives; and culminated with the decision to
kill Solaris x86.
These are the actions of a company in contraction. Contraction in the
wrong direction. Sun is a company with bigger money and market problems
than most of us outsiders were aware of until these actions. To
understand the problem and appreciate how we got to this point, there's
some history that needs reviewing.
I remember the first Sun Workstations. At the time, the term
"workstation" was synonymous with Apollo, the company that invented the
market. Sun's entry into the world, like that of DEC before it, was
based on offering equal or better capability at better price. Unlike
DEC, they used "off the shelf" technology for just about everything: the
CPU was a Motorola chip, the OS was based on Bill Joy's grad work on
BSD, etc. In effect, Sun didn't Invent or Build anything, they were
Systems Integrators. Apollo's arrogance in the marketplace and
trivializing of the threat represented by Sun ultimately put them out of
business as Sun ate their market share and expanded into markets Apollo
had never been able to because of their price/performance ratio. This
was despite the fact that Apollo had some technologies that were clearly
superior to anything anyone else had. Like DEC before it, Sun
established a new value point, and with it a unique ecological niche.
Thus proving once again that initial acquisition cost is =the= big
determining factor in IT spending, not technological superiority. Then
Sun started to get confused by their own success.
Because it was cheap HW that gave Sun life, Sun's top management thought
of the company as a HW company. Thus while Sun was pursuing innovations
like SMP Operating Systems, they were also underwriting the development
of the first commercial RISC CPU, the SPARC. As Sun's market share
grew, top management felt justified that the HW, leavened by SW
innovation, was responsible for Sun's success. The truth was and is the
exact opposite: None of Sun's HW is that good without the SW that Sun
created to make it work well. The SPARC series CPU has always been
underwhelming, etc while SW like Solaris and Java were what really made
people Buy Sun. Sun management did not listen when other people tried
to correct their world view, even people who had big line item purchase
responsibility or industry experts. In so doing, they unconsciously
began to copy the arrogant attitudes of companies like DEC and Apollo
before them: they stopped listening to the customer. Sheer momentum
made Sun a multi-billion dollar company, again validating top
management's world view. Then the real problems began.
The days of easy growth were over and Sun found that growing a
world-wide enterprise IT company was =not= easy. Tied to their vision
of Sun as a HW company, management tried to compete with the likes of
DEC and IBM by creating HW that could compete with the capabilities of
that from DEC and IBM. Some of these things, like HW failover, were
absolute requirements in the space Sun wanted to play in. Having such
features meant you could "ante up" for the game, it did not guarantee
that you'd get a good hand. And try as they might, Sun's HW strategy
did not yield the rewards management thought it would. Again they were
told: "It's the SW, stupid", and again they did not listen.
During all of this there was a SW effort that Sun management thought was
unimportant. Originally codenamed "Apache", it was the port, originally
to the 80386, of the Sun core OS and associated utilities. Recognizing
that 40+ million commodity ants (especially ants that get 2x stronger
every year)were inevitably going to topple just about every proprietary
CPU that anyone could make, Industry Insiders like myself applauded the
product. Sun hated it from Day One. The best fate that a fanatic
following could arrange was for a strategy of Benign Neglect: "It isn't
Real Sun, but we'll leave the toy around so you can play until you are
convinced and mature enough to Buy Real Sun". The parallels between
this attitude and IBM's about a little Loss-Leader they made during a
rush project in Boca Raton were pointed out to Sun, but fell on deaf
ears. So we Faithful shrugged our shoulders and went back to our day to
day routine. Again.
At one point, Sun decided to "go around" rather than through the
competition: they pushed for market share amongst all of the startups
of the .com boom, "putting the dot in .com" as it were. We all know how
that turned out. Sun took some big loses because of their leadership
position during that fad.
Sun's CPU work never yielded the world leadership position they aspired
to. Consultants like myself were telling them to buy the Alpha or make
something truly innovative or they'd be in real trouble as long as they
considered themselves a HW company. For one shining moment, it looked
like Sun had pulled it off. On paper, MAJC was everything we had told
Sun they needed, and we began to think we might have to change our tune.
Sun even called the chip a "bet the company" decision. Then MAJC died.
By then it was obvious Sun was never going to acquire the Alpha, so we
began the old chorus: "SUN IS NOT A HW COMPANY. Sun is an IT Solutions
company. Change your Business Strategy or we shall all suffer." Nope.
So here we are. Sun is contracting into its self perceived core
business: their HW, and abandoning or contracting everything else.
Exactly the opposite of what they should be doing. And this time the
iceberg is very close to the ship. Sun can still turn the rudder and
avoid it, but given history I doubt they will. I've been a die hard Sun
loyalist from Day one. I've put my career on the line at times to back
them, and they're about to commit suicide. PLEASE listen to us this
time, or at least I am going to be backed into a corner with no choice.
If Solaris x86 dies, I'm going to have to tell all my IT contacts to Buy
Something Else on the next upgrade/replace cycle. I'm going to quietly
start selling off my stock, and I'm going to have to tell the young IT
professionals not to waste their time learning a dead end product
family. Sun's death will do neither your customer's nor The Industry
any good. Listen to the lookout and avoid the iceberg. Please?
YTC#1
>
> DO NOT ASK FOR WHOM THE BELL TOLLS, IT TOLLS FOR SUN
>
> The decision by Sun Microsystems to "kill" Solaris x86 means their days
> as a viable business unit are numbered. Don't get me wrong, I'm not
Don't talk such utter bo****s
--
Bruce Porter
"The internet is a huge and diverse community
and not every one is friendly"
http://www.ytc1.co.uk
Ron
Sent: Thursday, January 17, 2002 2:31 PM
Subject: Re: Do Not Ask For Whom The Bell Tolls
| Ron wrote:
| >
| > DO NOT ASK FOR WHOM THE BELL TOLLS, IT TOLLS FOR SUN
| >
| > The decision by Sun Microsystems to "kill" Solaris x86 means their
days
| > as a viable business unit are numbered. Don't get me wrong, I'm not
|
| Don't talk such utter bo****s
^^^^^^^
Eh?
I'd love to respond intelligently to your comment, but I can't parse
it...
YTC#1
I think you will find that I am implying you have managed to spew forth
a pile of drivel suggesting that the only thing that keeps Sun afloat is
X86.
You obviously have no understanding of thier business and where it
stands in the great scheme of things.
Yes X86 is/was good and handy, yes it does have a place, yes I agree it
should not be axed. No I don't think it will overly dent Sun profits and
future growth.
Sun have axed producst in the past and they have re-appered with a
vengance (SDS, and x86 !)
Ron
"YTC#1" <ytcREMOV...@ytc1.co.uk> wrote in message
news:3C4848F0...@ytc1.co.uk...
| Ron wrote:
| >
| > From: "YTC#1" <ytcREMOV...@ytc1.co.uk>
| > Sent: Thursday, January 17, 2002 2:31 PM
| > Subject: Re: Do Not Ask For Whom The Bell Tolls
| >
| > | Ron wrote:
| > | >
| > | > DO NOT ASK FOR WHOM THE BELL TOLLS, IT TOLLS FOR SUN
| > | >
| > | > The decision by Sun Microsystems to "kill" Solaris x86 means
their
| > days
| > | > as a viable business unit are numbered. Don't get me wrong, I'm
not
| > |
| > | Don't talk such utter bo****s
| > ^^^^^^^
| > Eh?
| > I'd love to respond intelligently to your comment, but I can't parse
| > it...
| I think you will find that I am implying you have managed to spew
forth
| a pile of drivel suggesting that the only thing that keeps Sun afloat
is
it
| stands in the great scheme of things.
a disturbing list of strategically scary decisions on Sun's part, not
the only one. It =is= true that I find the abandonment of commodity
class HW to be strategically the worst item on that list. History has
shown that the high volume commodity HW is going to "win the war".
Being in the low volume proprietary HW business relegates you to the
status of niche player, and then extinct, unless you are something
exceptional in the HW field. SUN IS NOT A HW COMPANY. They never have
been. SUN IS AN IT SOLUTIONS COMPANY. They always have been. When
contracting towards their core in bad times, they must contract along
this rule or they risk "contracting" themselves out of business.
| Yes X86 is/was good and handy, yes it does have a place, yes I agree
it
| should not be axed. No I don't think it will overly dent Sun profits
and
| future growth.
We are, of course, free to disagree with each other ;-)
| Sun have axed products in the past and they have re-appeared with a
| vengeance (SDS, and x86 !)
If The Titanic avoids the iceberg, it can change course and speed at
will later. If it hits the iceberg, no navigation decision after that
matters. Sun has =never= been in as much danger as they are now, and
they are acting frighteningly like Apollo, and DEC, and Stratus, and
etc, did under similar circumstances. If Sun wants to survive, =they
have to get it right this time=. Commodity HW is finally coming of age.
Like those generals who were waging war when major technological
advances took place, the choices are Learn and Live vs. Don't.
Ron Peacetree
Paul Floyd
Using a variable width font newsreader, are we?
It might help to know that there's one * missing, and that the starred
letters should be llock.
A bientot
Paul
--
Paul Floyd http://paulf.free.fr (for what it's worth)
Mail as URL, replace 1st . with @
What happens if you have lead in your pants as well as lead in your pencil?
Benny Kleykens
Sun is doing great. Most of its competitors have either given up or they
are trying to get togheter (in a more than clumsy way) to get some kind
of grip on Sun.
Scott McNeally is right when he says that IBM is the ONLY real
competition left in the sever business. As for OSes only Linux really
has a shot of ever getting where Solaris is at the time.
Frankly what more do you need ?
ALL competitors of Sun have decided they needed to support an Intel
processor that did not exist and an OS that is still trying to mature,
doesn't that really say it all ?
Obviously the times are scary for everyone atm, so they are scary for
Sun too but people that panic like you only make things worse, screaming
"FIRE !" where there is none.
Have a nice year, personally I think 2002 is going to be a GREAT year.
Benny
YTC#1
And you seem to think Sun is in the low volume area I take it ? I think
you need to have a trawl of many many major companies.
> status of niche player, and then extinct, unless you are something
> exceptional in the HW field. SUN IS NOT A HW COMPANY. They never have
ROTFL
What planet are you on ? A tin shifter, allways will be. PS keep a lot
of revenue up, but ultimatly Sun has always been and allways will be a
box shifter.
> been. SUN IS AN IT SOLUTIONS COMPANY. They always have been. When
Now my sides are spitting.
> contracting towards their core in bad times, they must contract along
> this rule or they risk "contracting" themselves out of business.
>
> | Yes X86 is/was good and handy, yes it does have a place, yes I agree
> it
> | should not be axed. No I don't think it will overly dent Sun profits
> and
> | future growth.
> We are, of course, free to disagree with each other ;-)
>
> | Sun have axed products in the past and they have re-appeared with a
> | vengeance (SDS, and x86 !)
> If The Titanic avoids the iceberg, it can change course and speed at
> will later. If it hits the iceberg, no navigation decision after that
> matters. Sun has =never= been in as much danger as they are now, and
Oh, not in as much danger of never succeding with that very first order
?
> they are acting frighteningly like Apollo, and DEC, and Stratus, and
You are on drugs, arn't you ?
> etc, did under similar circumstances. If Sun wants to survive, =they
> have to get it right this time=. Commodity HW is finally coming of age.
If any company wants to survive , they have to "get it right"
HP ? Compaq ?
IBM are living on thier name as usual, they *give* kit away because it
is the only method they have of beating Sun. Interestingly customers
will often come back to Sun when thier alternative platform did not do
as promised, despite being free.
> Like those generals who were waging war when major technological
> advances took place, the choices are Learn and Live vs. Don't.
Rant on.
You are Bill Gates AICMFFX
--
Bruce Porter
XJR1300SP, XJ900F, GSX750W, GS550, GSX250, CB175
WUSS#1 , YTC#1(bar), OSOS#2 , DS#3 , IbW#18
Ron
"Benny Kleykens" <benny.k...@pi.be> wrote in message
news:3C498BF8...@pi.be...
| The server business is not a niche market.
market", and it's only going to become more so as time goes on.
Perceiving oneself as a proprietary HW manufacturer is a very risky
thing to do. History has shown the HW is =not= Sun's strength.
Bang-for-the-buck wise, there are increasingly more and better HW
choices out there, and the performance gap continues to become more and
more lopsided in favor of the non Sun solutions. Sun's =SW=, OTOH, is
world class and second to none.
IT Organizations buy =solutions=, and the but the best solution at the
lowest price they can find. Sun's SW on proprietary HW causes debates
over the HW vs. the SW. Sun's SW on the best HW in the industry is a no
brainer. That HW tends to not be Sun HW, so Sun has to make sure their
SW runs on that HW for maximum competitive advantage. Abandoning their
cross platform capabilities and hobbling their Open Source efforts is a
way to commit suicide.
| Sun is doing great. Most of its competitors have either given up or
they
| are trying to get together (in a more than clumsy way) to get some
kind
| of grip on Sun.
Sun's revenue's were $3.11B last year with a loss of $.13/share. The
year before revenues were $5.12B with a =profit= of $.13/share. RedHat
has increased in profitability both years. IBM is so big they burp to
get over losses that would put almost anyone else, including Sun, out of
business. Sun had it's first mass layoff in the history of the company:
44,000 (~9% of worldwide headcount) in October of 2001. Sun is =not=
"doing great". Market penetration does you no good if you hemorrhage to
death, as lot's of .coms just proved.
| Scott McNeally is right when he says that IBM is the ONLY real
| competition left in the sever business. As for OSes only Linux really
| has a shot of ever getting where Solaris is at the time.
| Frankly what more do you need ?
"Waiter? Reality Check Please?" Sun doesn't even have =close= to an
answer to IBM's POWER4. No one does. That's the high end. Any lower
than that, and AMD and Intel are going to eat everyone's lunch. Solaris
=does= do SMP better than anyone else is likely to for the foreseeable
future, but that just proves my point that it's the SW that's Sun's UVP.
| ALL competitors of Sun have decided they needed to support an Intel
| processor that did not exist and an OS that is still trying to mature,
| doesn't that really say it all ?
By Q4 2003, both AMD and Intel will have 64b CPUs and 4-8 way SMP
solutions (Intel has a 32 way that is currently hobbled by the lack of
an OS. Unless you run S-x86 on it...) that will cover just about every
IT need. What they can't handle, the shrink and repackage of POWER
will. Unless Sun pulls off a MAJC like miracle, there's no HW niche for
them to live in. And that "OS that's trying to mature" has =the= best
OS Arch and Dev group in the world (unless Dave Cutler and Co leave M$
in disgust. Would that it were so.) SQL Server is now completely
competitive with Oracle. IIS outperforms Apache on a regular basis,
etc. Win2K/WinXP =is= growing up, and faster than some want to admit.
It's a mistake to underestimate your enemy. M$ is a =dangerous=
monopolistic entity that has managed to prove that The Sherman
Anti-Trust Act doesn't apply to them.
| Obviously the times are scary for everyone atm, so they are scary for
| Sun too but people that panic like you only make things worse,
screaming
| "FIRE !" where there is none.
I'm not screaming "fire!", I'm screaming "change direction and avoid
that iceberg, or we're all going down!". We have to set direction
correctly =now= to avoid the iceberg =then=.
| Have a nice year, personally I think 2002 is going to be a GREAT year.
It always is for someone, I'd just like that someone to be on my side.
Having to throw out all of my infrastructure and convert to M$ because I
have no other choice is =not= my idea of "great". And if Sun makes the
wrong choices in 2002, we all might be backed into such a corner by the
end of 2004.
Ron
| Ron wrote:
| > As I stated in the post, the killing of Solaris-x86 is the final
item in
| > a disturbing list of strategically scary decisions on Sun's part,
not
| > the only one. It =is= true that I find the abandonment of commodity
| > class HW to be strategically the worst item on that list. History
has
| > shown that the high volume commodity HW is going to "win the war".
| > Being in the low volume proprietary HW business relegates you to the
| And you seem to think Sun is in the low volume area I take it ? I
think
| you need to have a trawl of many many major companies.
engaged, or worked for most of the Fortune 2000. How about you? IT
Managers are =nervous= about the future at the moment.
| > status of niche player, and then extinct, unless you are something
| > exceptional in the HW field. SUN IS NOT A HW COMPANY. They never
have
| ROTFL
| What planet are you on ? A tin shifter, always will be. PS keep a lot
| of revenue up, but ultimately Sun has always been and always will be a
| box shifter.
|
| > been. SUN IS AN IT SOLUTIONS COMPANY. They always have been. When
| Now my sides are spitting.
Clearly we disagree. I provided the historical basis for my viewpoint,
what's the justification for yours?
| > If The Titanic avoids the iceberg, it can change course and speed at
| > will later. If it hits the iceberg, no navigation decision after
that
| > matters. Sun has =never= been in as much danger as they are now,
and
| Oh, not in as much danger of never succeeding with that very first
order
| ?
No, because unlike Apollo, M$ doesn't "write off" or "ignore" any
competitor. I was there. I saw Apollo decide that Sun was a "no count,
cheap HW provider without technology that's anywhere near ours". M$ is
the exact opposite, to the point that they killed the first company to
make a windowing DOS shell using market manipulation when they had
=nothing= like it. Not a single line of code. That set of actions
delayed a DOS GUI by 3 years and a stable DOS GUI by 5-6, stifling
innovation that could benefit the consumer by that amount of time. But
maintaining M$ market position.
| > they are acting frighteningly like Apollo, and DEC, and Stratus, and
| You are on drugs, aren't you ?
I'm trying to save =your= job Mr. "I work for Sun". If you don't
believe my analysis, go look up the business cases I'm hinting at and
Independently Investigate The Truth. Or at least, be polite to people
who are acting in your interest's whether you believe them or not.
| > etc, did under similar circumstances. If Sun wants to survive,
=they
| > have to get it right this time=. Commodity HW is finally coming of
age.
| If any company wants to survive , they have to "get it right"
No, sometimes market forces are such that you just need to be in the
right place at the right time and it won't matter what mistakes you
make. Being a network services provider was like that during the 90's
because demand so outstripped supply that =anyone= who could do it was
guaranteed work. That analogy does not hold for the current IT market
in general and Sun in particular.
| HP ? Compaq ?
| IBM are living on their name as usual, they *give* kit away because it
| is the only method they have of beating Sun. Interestingly customers
| will often come back to Sun when their alternative platform did not do
| as promised, despite being free.
IBM's POWER4 is the highest performance CPU design in existence as of
this writing (go look at www.top500.org to see who makes the fastest
machines on the planet. "White", "Blue", and "Red" are IBM Big Iron)
Compaq no longer exists. HP is a niche player in the machine business
and a major force in the peripherals business. Sun makes sales because
they have better SW, not HW. Unfortunately, the gap between Sun's HW
and anyone else's at any price/performance point in the market is
becoming worse and worse for Sun. There's only so long that superior SW
can cover for inferior HW at higher prices.
| > Like those generals who were waging war when major technological
| > advances took place, the choices are Learn and Live vs. Don't.
| Rant on.
"Those who do not learn from History are doomed to repeat it's
mistakes."
"Know The Enemy and yourself and you will win every engagement. Know
only The Enemy or yourself and you will win every other engagement.
Know Neither The Enemy nor yourself and you will never win an
engagement."
| You are Bill Gates AICMFFX
By all means, enlighten me as to what "AICMFFX" means?
Dennis
You have a most persuasive arguement and I tend to agree with it- my MIS
manager is having fits because we recently have been starting to go with Sun
SW and now he's wondering if we should stay with M$ or figure out what
flavor of Linux is going to come out on top...
Damn sad it is...
I'd pick you to go meet with those Sun "executives" when the time comes. You
have my vote :)
Dennis Price
Gain Communications
"98.2% of a Win2kAD Shop"
"1.8% Sun x86" (which is now probably going away due to Sun's
well-thought-out poor "business" decisions instead of growing like we had
planned less than two months ago)
dprice@gci-net*dot*com
"Ron" <purp...@earthlink.net> wrote in message
news:9Pr28.9140$OS5.7...@newsread1.prod.itd.earthlink.net...
YTC#1
>
> "YTC#1" <ytcREMOV...@ytc1.co.uk> wrote in message
> news:3C498F0B...@ytc1.co.uk...
> | Ron wrote:
> | > As I stated in the post, the killing of Solaris-x86 is the final
> item in
> | > a disturbing list of strategically scary decisions on Sun's part,
> not
> | > the only one. It =is= true that I find the abandonment of commodity
> | > class HW to be strategically the worst item on that list. History
> has
> | > shown that the high volume commodity HW is going to "win the war".
> | > Being in the low volume proprietary HW business relegates you to the
> | And you seem to think Sun is in the low volume area I take it ? I
> think
> | you need to have a trawl of many many major companies.
> I've had P&L responsibility in the $10M+ range, and consulted with,
> engaged, or worked for most of the Fortune 2000. How about you?>
What about me? What does P&L have to do with it?
> IT Managers are =nervous= about the future at the moment.
IT managers have always been nervous.
>
> | > status of niche player, and then extinct, unless you are something
> | > exceptional in the HW field. SUN IS NOT A HW COMPANY. They never
> have
> | ROTFL
> | What planet are you on ? A tin shifter, always will be. PS keep a lot
> | of revenue up, but ultimately Sun has always been and always will be a
> | box shifter.
> |
> | > been. SUN IS AN IT SOLUTIONS COMPANY. They always have been. When
> | Now my sides are spitting.
> Clearly we disagree. I provided the historical basis for my viewpoint,
> what's the justification for yours?
What historical basis ? You wrote something and it becomes historical
basis ?
Next you will be saying Intel is a SW company.
>
> | > If The Titanic avoids the iceberg, it can change course and speed at
> | > will later. If it hits the iceberg, no navigation decision after
> that
> | > matters. Sun has =never= been in as much danger as they are now,
> and
> | Oh, not in as much danger of never succeeding with that very first
> order
> | ?
> No, because unlike Apollo, M$ doesn't "write off" or "ignore" any
> competitor. I was there. I saw Apollo decide that Sun was a "no count,
> cheap HW provider without technology that's anywhere near ours". M$ is
^^^^^^^^^^^ Shouldn't that be SW ?
> the exact opposite, to the point that they killed the first company to
> make a windowing DOS shell using market manipulation when they had
> =nothing= like it. Not a single line of code. That set of actions
> delayed a DOS GUI by 3 years and a stable DOS GUI by 5-6, stifling
> innovation that could benefit the consumer by that amount of time. But
> maintaining M$ market position.
Whoosshh, lost me with this one, I thought we were talking Sun here?
>
> | > they are acting frighteningly like Apollo, and DEC, and Stratus, and
> | You are on drugs, aren't you ?
> I'm trying to save =your= job Mr. "I work for Sun". If you don't
If the security of my job was down to you , I would be worried.
> believe my analysis, go look up the business cases I'm hinting at and
Which I don't.
> Independently Investigate The Truth. Or at least, be polite to people
> who are acting in your interest's whether you believe them or not.
If I believed every one who said they were acting in my interest, when
they are spouting crap based on the X86 decison (rememebr your 1st post
?) then I would be a very dim and poor person.
>
> | > etc, did under similar circumstances. If Sun wants to survive,
> =they
> | > have to get it right this time=. Commodity HW is finally coming of
> age.
> | If any company wants to survive , they have to "get it right"
<snip>
>
> | HP ? Compaq ?
> | IBM are living on their name as usual, they *give* kit away because it
> | is the only method they have of beating Sun. Interestingly customers
> | will often come back to Sun when their alternative platform did not do
> | as promised, despite being free.
> IBM's POWER4 is the highest performance CPU design in existence as of
> this writing (go look at www.top500.org to see who makes the fastest
I think tests on newer boxes would be a tad fairer. The ESx500 is old
hat.
Have a look around for some real world/ app tests.
Ask yourself why Oracle develop on Sun, why IBM do thier quantum
computing on Sun.
Find out what chip manufacturers design on ...
> machines on the planet. "White", "Blue", and "Red" are IBM Big Iron)
> Compaq no longer exists. HP is a niche player in the machine business
At last we agree on something
> and a major force in the peripherals business. Sun makes sales because
> they have better SW, not HW. Unfortunately, the gap between Sun's HW
> and anyone else's at any price/performance point in the market is
> becoming worse and worse for Sun. There's only so long that superior SW
> can cover for inferior HW at higher prices.
Why would we want to run our software on inferior platforms ? We let
everyione else do that.
>
> | > Like those generals who were waging war when major technological
> | > advances took place, the choices are Learn and Live vs. Don't.
> | Rant on.
> "Those who do not learn from History are doomed to repeat it's
> mistakes."
Thos taht don't take thier heads out the sand and stop saying "me too"
go nowhere.
>
> "Know The Enemy and yourself and you will win every engagement. Know
> only The Enemy or yourself and you will win every other engagement.
> Know Neither The Enemy nor yourself and you will never win an
> engagement."
>
> | You are Bill Gates AICMFFX
> By all means, enlighten me as to what "AICMFFX" means?
As in
And I Claim My Five Free (X boxes )
I thought a man of your obvious experience would have recognised the
standard troll response.
--
Bruce Porter
Ron
| > I've had P&L responsibility in the $10M+ range, and consulted with,
| > engaged, or worked for most of the Fortune 2000. How about you?
| What about me? What does P&L have to do with it?
would call a pertinent response. I'm one of the people that your bosses
try to sell HW to every year. I've had what some would call large
budgets under my control for that purpose. I'm routinely asked for
advice on IT issues by others in similar situations, and I'm one of the
people who's been doing SW development in this industry since the days
when Sun first arrived in the market. Now, since you questioned =my=
bona-fide's, it seems only appropriate that you return the courtesy and
provide =yours=.
I'm fairly certain that at this point, our discussion is rapidly
becoming, if not already, not appropriate for a public forum. Please
post a valid email address for future private discussion if you wish to
continue this. I like fire in people, and you obviously have that. But
you need to be careful about misusing that fire, or you're going to end
up burning yourself.
| > IT Managers are =nervous= about the future at the moment.
| IT managers have always been nervous.
That's a horribly cavalier attitude to have about the customers who
provide your company the $$ to pay your wages.
| What historical basis ? You wrote something and it becomes historical
| basis ?
I've given historical facts. They can be checked for validity. If you
think any of what I said was a lie, you have the onus to provide a
source that disputes my claimed facts. I've claimed that those facts
form a set of precedents that should be taken as guideposts as to what
works in this kind of situation and what doesn't. If you disagree and
want to say so, you have the responsibility to provide a rationale as to
why my logic is flawed or why those historical facts don't matter in
this situation. I've never claimed to be infallible, by all means prove
me wrong. but =prove= me wrong. Shouting invective at me doesn't help
further the discussion and makes you look immature or foolish.
| > No, because unlike Apollo, M$ doesn't "write off" or "ignore" any
| > competitor. I was there. I saw Apollo decide that Sun was a "no
count,
| > cheap HW provider without technology that's anywhere near ours". M$
is
| ^^^^^^^^^^^ Shouldn't that be SW ?
I stated Apollo's viewpoint as I remember hearing it at the time.
| > the exact opposite, to the point that they killed the first company
to
| > make a windowing DOS shell using market manipulation when they had
| > =nothing= like it. Not a single line of code. That set of actions
| > delayed a DOS GUI by 3 years and a stable DOS GUI by 5-6, stifling
| > innovation that could benefit the consumer by that amount of time.
But
| > maintaining M$ market position.
| Whoosshh, lost me with this one, I thought we were talking Sun here?
We =are= talking Sun here. Who do you think is Sun's biggest long term
threat? Most professionals in the IT industry would say M$. Therefore
M$'s attitude toward what they perceive as potential competitors is
germane, don't you think? History has shown M$ to be a ferocious
competitor. They don't take prisoners, they'll do anything to win, and
they don't believe in mutual co-existence.
| > I'm trying to save =your= job Mr. "I work for Sun". If you don't
| If the security of my job was down to you , I would be worried.
Get worried. I and others like me are The Customer. Without us, you
have no job.
| > believe my analysis, go look up the business cases I'm hinting at
and
| Which I don't.
Fine. Then bring something to the table to further the debate.
| > Independently Investigate The Truth. Or at least, be polite to
people
| > who are acting in your interest's whether you believe them or not.
| If I believed every one who said they were acting in my interest, when
| they are spouting crap based on the X86 decision (remember your 1st
post
| ?) then I would be a very dim and poor person.
I never asked you to believe my argument, I asked you to believe that
I've raised it in order to help you. Whether or not you think an ally's
attitude is foolish, you owe them the courtesy due an ally. By all
means, be skeptical. That's healthy. But responding emotionally does
not help you separate fact from fiction or find flaws in logic. In
addition, it can lead you down self-destructive courses.
| > IBM's POWER4 is the highest performance CPU design in existence as
of
| > this writing (go look at www.top500.org to see who makes the fastest
| I think tests on newer boxes would be a tad fairer. The ESx500 is old
| hat. Have a look around for some real world/ app tests.
IBM holds the performance crown on that web site with old HW, just
released much =better= HW, and you don't see how important that is?
| Ask yourself why Oracle develop on Sun, why IBM do their quantum
| computing on Sun.
Oracle and Sun have a historical alliance of a strategic nature. I
afraid I don't understand your reference to "Quantum Computing" as that
to my knowledge refers to a research area of logic and circuit design
orthogonal to this entire thread. If you choose to continue this
discussion privately, please clarify then.
| Find out what chip manufacturers design on ...
mostly PC's.
| > and a major force in the peripherals business. Sun makes sales
because
| > they have better SW, not HW. Unfortunately, the gap between Sun's
HW
| > and anyone else's at any price/performance point in the market is
| > becoming worse and worse for Sun. There's only so long that
superior SW
| > can cover for inferior HW at higher prices.
| Why would we want to run our software on inferior platforms ? We let
| everyone else do that.
The degree to which that echoes the attitudes I've heard in the past
from failed IT companies is frightening.
| >
| > | > Like those generals who were waging war when major technological
| > | > advances took place, the choices are Learn and Live vs. Don't.
| > | Rant on.
| > "Those who do not learn from History are doomed to repeat it's
| > mistakes."
| Those that don't take their heads out the sand and stop saying "me
too"
| go nowhere.
And that is germane to this discussion how?
| > | You are Bill Gates AICMFFX
| > By all means, enlighten me as to what "AICMFFX" means?
| As in
| And I Claim My Five Free (X boxes )
| I thought a man of your obvious experience would have recognized the
| standard troll response.
I've never been one to keep track of most of the cute contractions. And
I never assume I'm talking to someone who will respond like a troll.
Paul W. Stoecker, Ph.D.
> In the last few months Sun has killed the IPlanet initiative; made
> decisions about Java licensing that benefit only IBM (and hurt everyone
> else, including BEA and Sun); pretty much killed or started to kill all
> of their Open Source initiatives; and culminated with the decision to
> kill Solaris x86.
I don't think that Sun has killed the IPlanet initiative...in fact, I
believe that they have just rolled it into their own operations and
brought it back to the mothership. Netscape no longer exists, so
there was no real alliance. I believe that iPlanet is one of the
basic building blocks of the SunOne initiative. The collaboration,
application, directory and other servers are still alive.
I don't see Sun killing open source initiatives either. JAVA is a
good example of a community process that works. They continue to
develop StarOffice and have given it to the open source crowd to do as
they wish.
Is killing x86 Solaris a mistake? Maybe, but to be perfectly honest,
it is a real dog on the Intel processor. Sun ends up spending
precious resources on making sure that all sorts of peripherals,
cards, etc. have compatible drivers just to get it to run.
I really don't find the SPARC to be a bad platform. They have more
processing power than most Intel chips, but more importantly, my Sun
Hardware is the most reliable computer hardware that I have ever
owned.
It seems to make perfect business sense to postpone or ax the project
if there is no ROI in a reasonable time. I would rather have Sun
spend its resources, making the Solaris OE more enterprise aware,
easing the administrative workload.
Of course, I can see we will have to agree to disagree, but I don't
see Sun folding up shop just because they stop the development of the
x86 version of Solaris. I also don't think they see themselves as a
HW company only, I am sure Scott McNealy is fully aware that all HW
needs great SW.
Think about some of the new things coming out:
The Liberty Alliance Project, JavaTV, SunOne, JXTA, Iplanet products,
etc. They are still hard at work improving their HW, but at the same
time releasing great SW products.
To be honest, I am a big supporter of Sun because they have excellent
Technical support, are responsive when I have issues and their systems
with SW included are the most reliable.
They won't be able to please everyone, no company can....maybe there
will be some extra stock around in coming months for me to pick it up
at a good price.
Regards,
Paul
r...@toad.rmkhome.com
:You have a most persuasive arguement and I tend to agree with it- my MIS
:manager is having fits because we recently have been starting to go with Sun
:SW and now he's wondering if we should stay with M$ or figure out what
:flavor of Linux is going to come out on top...
Just remember, M$ software runs on junk hardware. While Solaris x86 also
runs on that same hardware, one can scale up to carrier class sparc systems
easily.
I've always deployed Solaris x86 for small servers, DNS, NIS, mail, and
some web servers. For larger servers I would always go with sparc. Intel
servers have fast cpu and low throughput. Sun sparc servers have high
throughput under heavy load.
--
Rick Kelly r...@rmkhome.com www.rmkhome.com
Philip Brown
>...
>Is killing x86 Solaris a mistake? Maybe, but to be perfectly honest,
>it is a real dog on the Intel processor.
What are you talking about?
"dog" usually refers to speed. As long as you have decent hardware
(eg: *SCSI* DRIVES) it runs great.
Even with ATA drives, its pretty good, as long as you remember to enable
DMA.
--
[Trim the no-bots from my address to reply to me by email!]
[ Do NOT email-CC me on posts. Pick one or the other.]
S.1618 http://thomas.loc.gov/cgi-bin/bdquery/z?d105:SN01618:@@@D
http://www.spamlaws.com/state/ca1.html
Alan Coopersmith
|In the last few months Sun has killed the IPlanet initiative; made
|decisions about Java licensing that benefit only IBM (and hurt everyone
|else, including BEA and Sun); pretty much killed or started to kill all
|of their Open Source initiatives; and culminated with the decision to
|kill Solaris x86.
iPlanet is alive and well - the only recent major change is it's now
part of Sun, instead of a joint venture of AOL/Netscape & Sun.
And I don't know of a single Open Source project at Sun that's been cut
recently - in fact, if you download the latest Solaris 9 EA you'll see
more open source than ever in the OS, and I'm personally working with a
project that just got approval to open its source soon.
(I can't comment on the Java licensing issues, since I have no idea what
you're referring to, but I don't pay much attention to Java licensing
issues in general.)
--
________________________________________________________________________
Alan Coopersmith al...@alum.calberkeley.org
http://soar.Berkeley.EDU/~alanc/ aka: Alan.Coo...@Sun.COM
Working for, but definitely not speaking for, Sun Microsystems, Inc.
Paul W. Stoecker, Ph.D.
> On 20 Jan 2002 16:04:34 -0800, sto...@research.panasonic.com wrote:
> >...
> >Is killing x86 Solaris a mistake? Maybe, but to be perfectly honest,
> >it is a real dog on the Intel processor.
>
> What are you talking about?
> "dog" usually refers to speed. As long as you have decent hardware
> (eg: *SCSI* DRIVES) it runs great.
> Even with ATA drives, its pretty good, as long as you remember to enable
> DMA.
Comparatively, Linux is faster than Solaris on the Intel Platform. I
am not saying that Linux is better only faster. Actually, I am more
of the opinion that linux is like AIX....ain't unix. :-)
Have you ever tried to install Solaris for Intel on a machine with a
RAID controller card and those fast SCSI drives you are talking about.
Any OS is fast on "decent hardware", my point is that it is slower
than other alternative OSes, it lacks support for a lot of different
hardware, the price of SPARC hardware is considerably lower and the HW
is better (I do believe a SunBlade 100 is going for about the same
price as a "good" Intel PC.), the Intel version is less stable, and it
takes a lot more engineers and researchers to support multiple
platforms. So, in the end, killing the Intel version makes good
business sense.
The business of Sun is to make money. I would rather they killed
Solaris on the Intel platform and concentrate on the hardware and
software that works the best. By now, Apple should have been in its
grave a million times over and years ago, everyone predicted that,
didn't they? But it is still a thriving company.
Sun is similar to Apple in the sense that they have a hardware and
software solution that is all their own. Sun has the advantage in
that it does work with the open source community and does make its OS
and many software titles free for public consumption. Also, Sun does
not have to rely on many other companies to make software for its
platform.
The question raised is if cancelling the continued development and
research into the Intel Platform of Solaris going to send Sun into
oblivion? (I have already refuted many of the other claims raised
about iPlanet and so forth). I don't know...because unlike Ms. Cleo,
I can't see into the future, but my bet is that Sun will be around for
some time.
The most credible threat to Sun is Linux, but when people realize that
the TCO for Linux can be higher than that for Unix or Windows, you may
see a change. A recent study showed that the ads by IBM about the
server consolidation was not completely honets. Linux was not even
cheaper than Windows until it reached 25,000 email accounts, then it
quickly became cheaper creating a real cost savings at 50-75,000 email
accounts (now, how many people need that many accounts.) I hope Sun
does the same study for Linux vs. Solaris, we may all be surprised.
Oh well, enough of my musing and pontificating.
Like I said before, we have to agree to disagree, much like
discussions of politics or religion.
Paul
Ron
"Alan Coopersmith" <al...@CSUA.Berkeley.EDU> wrote in message
news:a2ghhr$9f4$1...@agate.berkeley.edu...
| "Ron" <purp...@earthlink.net> writes in alt.solaris.x86:
| |In the last few months Sun has killed the IPlanet initiative; made
| |decisions about Java licensing that benefit only IBM (and hurt
everyone
| |else, including BEA and Sun); pretty much killed or started to kill
all
| |of their Open Source initiatives; and culminated with the decision to
| |kill Solaris x86.
| iPlanet is alive and well - the only recent major change is it's now
| part of Sun, instead of a joint venture of AOL/Netscape & Sun.
only. All the resources doing real work are coming from the
Netscape/AOL side of the table, with Sun planning to or already removing
their resources from the project. Can you provide more details as to
the reality?
| And I don't know of a single Open Source project at Sun that's been
cut
| recently - in fact, if you download the latest Solaris 9 EA you'll see
| more open source than ever in the OS, and I'm personally working with
a
| project that just got approval to open its source soon.
Unfortunately, all of that is for =SPARC=, a proprietary platform that
is less and less part of the mainstream. If all of that was still being
made available for x86, I wouldn't have started this. Remember my
hypothesis, right or wrong, is that the majority of the IT world
ultimately belongs to commodity HW, with specialized niches for special
purpose high end stuff. If this is correct, the killing of S-x86 is a
strategic mistake that cedes most of the IT industry to those who =do=
continue to support commodity HW.
| (I can't comment on the Java licensing issues, since I have no idea
what
| you're referring to, but I don't pay much attention to Java licensing
| issues in general.)
The information I was given was that Sun's Java licensing scheme has
changed to a format where anyone who wants to see any Java code provided
by Sun that provides functionality must pay for a license in the $100K+
range. And that once you've seen that code you can't share anything
about that code to anyone else who hasn't paid such a fee. This is
death to all of the cheaper App server development efforts, etc. It
also threatens BEA since part of their UVP is the ability to buy their
stuff and see a very rich Java implementation at every level of detail.
"Black boxing" their stuff is counter to the way they historically make
$$. IBM, of course, can pay the fee for the parts they need in
WebSphere without blinking...
Ron
Sent: Monday, January 21, 2002 8:41 AM
| phi...@bolthole.no-bots.com (Philip Brown) wrote in message
news:<slrna4n6dg....@bolthole.com>...
| > On 20 Jan 2002 16:04:34 -0800, sto...@research.panasonic.com wrote:
| > >...
| > >Is killing x86 Solaris a mistake? Maybe, but to be perfectly
honest,
| > >it is a real dog on the Intel processor.
| >
| > What are you talking about?
| > "dog" usually refers to speed. As long as you have decent hardware
| > (eg: *SCSI* DRIVES) it runs great.
| > Even with ATA drives, its pretty good, as long as you remember to
enable
| > DMA.
|
| Comparatively, Linux is faster than Solaris on the Intel Platform. I
| am not saying that Linux is better only faster. Actually, I am more
| of the opinion that Linux is like AIX....ain't Unix. :-)
Using the 2.4 or higher kernel, Linux is faster for many, but not all
tasks, as long as the box in question has less than 3 CPU's. For SMP >=
4, Solaris comes into its own. If using the legal definition of "what
is Unix?", you are correct that neither Linux nor AIX are Unix, they are
"Unix-like". I happen to agree with you based on that definition. Most
people don't care...
| Have you ever tried to install Solaris for Intel on a machine with a
| RAID controller card and those fast SCSI drives you are talking about.
Yes. Successfully on multiple occasions. By "fast" I mean Seagate
Cheetah's (and Barracuda's Back in The Day, and ..., etc) The real
problem is that HW RAID support for a specific drive tends to lag that
drives introduction to the marketplace by 9-12 months. By which time
better drives are available that the people who =need= the fastest
possible stuff want =those= instead of the now supported stuff. Which
is why I've stopped being a believer in HW RAID...
| Any OS is fast on "decent hardware", my point is that it is slower
| than other alternative OSes,
I'll see if I can get some real world benchmarks to answer this. The
gap is not as big, nor as one sided as you seem to believe. And my
comment about higher orders of SMP is at the beginning of this.
| it lacks support for a lot of different hardware,
That's a Sun choice, not a design limitation.
| the price of SPARC hardware is considerably lower and the HW
| is better (I do believe a SunBlade 100 is going for about the same
| price as a "good" Intel PC.),
$ for $, I can build more capable solutions that are just as stable
using commodity HW than I can by buying SPARC (or anybody else's
proprietary HW for that matter.) The first time I demonstrated to a Sun
Sales Rep and A Field Engineer that I could do everything their $45K box
could do with the $15K box I'd built, and just as quickly, stably, etc
they turned white. I then demonstrated the analogous thing at the
~$150K Sun price point. I was running S-x86 on server class PC HW. At
the high end (and that high end keeps getting further up the pyramid,
and therefore of smaller volume, every year), proprietary HW is the only
thing that has the horses to do some of the jobs. It's value for the $
tends to be nowhere near as good, but in that space that's no longer a
consideration.
| the Intel version is less stable, and it takes a lot more engineers
| and researchers to support multiple platforms. So, in the end,
| killing the Intel version makes good business sense.
SW Engineering and IT has been my business for a little while ;-)
Common source should =not= require more than 5% of the functionality in
the source tree be dedicated to HW specificity issues. If it is,
Something's Wrong with your policy and/or process.
| The business of Sun is to make money.
Couldn't agree with you more.
| I would rather they killed Solaris on the Intel platform and
concentrate
| on the hardware and software that works the best.
By your own statement, "works the best" is not the standard, "makes the
most $$ overall" is the correct metric. Increasingly, that isn't SPARC.
It's commodity HW. Nonetheless, I'm not advocating anything as extreme
as dropping support for either line. Just that we not kill off the more
promising branch in favor of the less so...;-)
| By now, Apple should have been in its grave a million times over
| years ago, everyone predicted that, didn't they? But it's still
thriving.
"thriving" is an "interesting" term to describe a company with less than
9% market share, and one whose primary CPU provider doesn't consider
them important enough to develop new chips for...
| Sun is similar to Apple in the sense that they have a hardware and
| software solution that is all their own. Sun has the advantage in
| that it does work with the open source community and does make its OS
| and many software titles free for public consumption. Also, Sun does
| not have to rely on many other companies to make software for its
| platform.
Apple had most of those advantages as well at one point in it's life.
And infrastructure companies like Sun, Apple, IBM, etc will always need
3rd party application developers.
| The question raised is if canceling the continued development and
| research into the Intel Platform of Solaris going to send Sun into
| oblivion? (I have already refuted many of the other claims raised
| about iPlanet and so forth). I don't know...because unlike Ms. Cleo,
| I can't see into the future, but my bet is that Sun will be around for
| some time.
I've asked for clarification as to the details regarding any
"refutations", so the jury still out there. And I was explicit that I
did not think Sun was suddenly going to Keel Over Dead. However, like
large trees, large companies can take a long time to die after receiving
mortal damage. "Sun being around for some time" is not my issue. My
issue is that if they start to resemble one of those mortally wounded
large trees, I have to cut them from my and my client's lives for
safety's sake. I and others warned Apple at their market height. They
didn't listen. They are paying the price. Now Sun is doing things that
set off similar alarms, and they're getting the same warning. I don't
want to be dragged kicking and screaming into a IT industry without Sun
as a major player. But if they insist, I want =less= to put my business
at risk.
| The most credible threat to Sun is Linux, but when people realize that
| the TCO for Linux can be higher than that for Unix or Windows, you may
| see a change. A recent study showed that the ads by IBM about the
| server consolidation was not completely honest. Linux was not even
| cheaper than Windows until it reached 25,000 email accounts, then it
| quickly became cheaper creating a real cost savings at 50-75,000 email
| accounts (now, how many people need that many accounts.) I hope Sun
| does the same study for Linux vs. Solaris, we may all be surprised.
Unfortunately, history shows that TCO metrics don't matter nearly as
much as Initial Acquisition Cost and Capital Expenditure in the IT
industry. Unless a revolution occurs in General Accounting Practices, I
don't think that pattern is likely to change (...and I, and other people
in this business, have a sore head from butting this wall repeatedly
over the years)
| Like I said before, we have to agree to disagree, much like
| discussions of politics or religion.
Sometimes it comes to that, but most of the time evidence and logic can
be weighed in a discussion and consensus achieved. I'd love to be able
to justify changing my stance on this issue. I'm not spending my
valuable time writing these to hear myself rant, I =want= a logical
discussion on the issue.
Ron Peacetree
Benny Kleykens
> the TCO for Linux can be higher than that for Unix or Windows, you may
> see a change.
Why would Linux be a threat to Sun ?
Think about it : if Linux becomes AS good as Solaris on SPARC then
companies will not have ANY problem finding people to administer their
systems. Right now they all to often have to sell out completely to
Global Services from IBM.
Sun does not make much money from Solaris on SPARC, the main source of
income is selling servers. Service agreements are taken for hardware
service, software aid comes with it but is not the reason. Consultancy
as a form of income is less imporant to Sun but consultancy on Linux may
be as profitable or even more than
consultancy on Solaris anyway.
And if linux ever becomes good enough to run on Sun servers then a whole
new market will open up for Sun.
Linux is not a treath to Sun, it is a "threat" to Solaris but not to
Sun.
Simple facts prove me right :
www.sunfreeware.com
www.openoffice.org
kind regards,
Benny
Timothy J. Lee
Paul W. Stoecker, Ph.D. <sto...@research.panasonic.com> wrote:
> Linux was not even
>cheaper than Windows until it reached 25,000 email accounts, then it
>quickly became cheaper creating a real cost savings at 50-75,000 email
>accounts (now, how many people need that many accounts.)
In the context of email, how does Windows get to be cheaper than
Linux (or Solaris or BSD) by any measure, even for just 5 email
accounts?
If nothing else, Windows servers often require more expensive hardware
than the competing x86 OSes. Then the email server software for Windows
tends to be more expensive (don't forget the per client license fees
as well), since the free stuff is not as well developed (as it is with
sendmail/qmail/postfix/etc. + various IMAP/POP choices available for
the competitors).
--
------------------------------------------------------------------------
Timothy J. Lee
Unsolicited bulk or commercial email is not welcome.
No warranty of any kind is provided with this message.
YTC#1
>
> | > | you need to have a trawl of many many major companies.
> | > I've had P&L responsibility in the $10M+ range, and consulted with,
> | > engaged, or worked for most of the Fortune 2000. How about you?
> | What about me? What does P&L have to do with it?
> You questioned my knowledge base and experience. I gave you what many
> would call a pertinent response. I'm one of the people that your bosses
> try to sell HW to every year. I've had what some would call large
> budgets under my control for that purpose. I'm routinely asked for
> advice on IT issues by others in similar situations, and I'm one of the
> people who's been doing SW development in this industry since the days
> when Sun first arrived in the market. Now, since you questioned =my=
> bona-fide's, it seems only appropriate that you return the courtesy and
> provide =yours=.
>
emailed
<snip>
Actually, have you thought of putting the origional post in
comp.unix.solaris ?
--
Bruce Porter
XJR1300SP, XJ900F, GSX750W, GS550, GSX250, CB175
WUSS#1 , YTC#1(bar), OSOS#2 , DS#3 , IbW#18
Paul W. Stoecker, Ph.D.
> only. All the resources doing real work are coming from the
> Netscape/AOL side of the table, with Sun planning to or already removing
> their resources from the project. Can you provide more details as to
> the reality?
>
Well, Sun has completely taken over support of the products. I have
iPlanet Messaging and Calendaring servers. I now submit my support
request via the web and was told the it will soon be rolled into the
Sun Online Support Center. Also, Sun continues to bundle iPlanet
applications with Solaris. If anything I believe Sun is the one
committing the resources.
> Unfortunately, all of that is for =SPARC=, a proprietary platform that
> is less and less part of the mainstream. If all of that was still being
> made available for x86, I wouldn't have started this. Remember my
> hypothesis, right or wrong, is that the majority of the IT world
> ultimately belongs to commodity HW, with specialized niches for special
> purpose high end stuff. If this is correct, the killing of S-x86 is a
> strategic mistake that cedes most of the IT industry to those who =do=
> continue to support commodity HW.
>
Let's see...openoffice, Jxta, Java, StarOffice...only run on
SPARCs???? I believe that many of their efforts run on multiple
platforms.
> The information I was given was that Sun's Java licensing scheme has
> changed to a format where anyone who wants to see any Java code provided
> by Sun that provides functionality must pay for a license in the $100K+
> range. And that once you've seen that code you can't share anything
> about that code to anyone else who hasn't paid such a fee. This is
> death to all of the cheaper App server development efforts, etc. It
> also threatens BEA since part of their UVP is the ability to buy their
> stuff and see a very rich Java implementation at every level of detail.
> "Black boxing" their stuff is counter to the way they historically make
> $$. IBM, of course, can pay the fee for the parts they need in
> WebSphere without blinking...
Can you provide your source? Seems to fly in the face of the JCP.
Paul
Paul W. Stoecker, Ph.D.
> Why would Linux be a threat to Sun ?
>
> Think about it : if Linux becomes AS good as Solaris on SPARC then
> companies will not have ANY problem finding people to administer their
> systems. Right now they all to often have to sell out completely to
> Global Services from IBM.
>
> Sun does not make much money from Solaris on SPARC, the main source of
> income is selling servers. Service agreements are taken for hardware
> service, software aid comes with it but is not the reason. Consultancy
> as a form of income is less imporant to Sun but consultancy on Linux may
> be as profitable or even more than
> consultancy on Solaris anyway.
>
You could be right, maybe that is why they want to postpone or stop
development of Solaris on x86. Maybe they have made the strategic
decision that they can perform Linux consultancy on the x86...after
all it is not a threat to Solaris.
> And if linux ever becomes good enough to run on Sun servers then a whole
> new market will open up for Sun.
> Linux is not a treath to Sun, it is a "threat" to Solaris but not to
> Sun.
> Simple facts prove me right :
>
> www.sunfreeware.com
> www.openoffice.org
>
Hmm! Sun still develops its own modified versions of open source
programs and their own version of StarOffice. In fact, you forgo
support if you replace their versions of BIND, Sendmail or Apache with
the latest open source versions.
All in all, Sun has a huge investment in Solaris especially in the
Sparc Platform, can you imagine the losses that would amount if Linux
does become "good enough". Therefore, it is still a threat...but
opinions vary.
Paul
>
> kind regards,
>
> Benny
Philip Brown
>phi...@bolthole.no-bots.com (Philip Brown) wrote in message news:<slrna4n6dg....@bolthole.com>...
>> "dog" usually refers to speed. As long as you have decent hardware
>> (eg: *SCSI* DRIVES) it runs great.
>> Even with ATA drives, its pretty good, as long as you remember to enable
>> DMA.
>
>Comparatively, Linux is faster than Solaris on the Intel Platform. I
For something other than desktop work? Not noticably.
>Any OS is fast on "decent hardware", my point is that it is slower
>than other alternative OSes
Yes, it runs at only 95% or whatever of linux.
That would be because it has the extra overhead of safety features that
stop it from blowing up under heavy load.
Paul W. Stoecker, Ph.D.
http://consultingtimes.com/Serverheist.html
They include groupware not just email accounts and TCO figures.
Paul
remo...@sonic.net (Timothy J. Lee) wrote in message news:<a2i0ve$gm9$1...@newbolt.sonic.net>...
Timothy J. Lee
Paul W. Stoecker, Ph.D. <sto...@research.panasonic.com> wrote:
>The study can be found at:
>
>http://consultingtimes.com/Serverheist.html
>
>They include groupware not just email accounts and TCO figures.
It is also a comparison using different hardware:
Microsoft on Intel PCs
Linux on IBM G6 mainframe
Obviously, the G6 will be more effective for huge sites, but
rather expensive for non-huge sites.
Paul Floyd
<sto...@research.panasonic.com> wrote:
>>
>> In the last few months Sun has killed the IPlanet initiative; made
>> decisions about Java licensing that benefit only IBM (and hurt everyone
>> else, including BEA and Sun); pretty much killed or started to kill all
>> of their Open Source initiatives; and culminated with the decision to
>> kill Solaris x86.
>I really don't find the SPARC to be a bad platform. They have more
>processing power than most Intel chips, but more importantly, my Sun
>Hardware is the most reliable computer hardware that I have ever
>owned.
At home I use a PC (a bit over 2 years old) and at work an Ultra 10
(about a year old). Even though the PC is older, and cost half as much,
it is still a fair bit faster. As for reliability, other than a $10
sound card that I fried with a bad driver, I've never had a problem.
>It seems to make perfect business sense to postpone or ax the project
>if there is no ROI in a reasonable time. I would rather have Sun
>spend its resources, making the Solaris OE more enterprise aware,
>easing the administrative workload.
>
>Of course, I can see we will have to agree to disagree, buton't
>see Sun folding up shop just because they stop the development
>x86 version of Solaris. I also don't think they see themselves as a
>HW company only, I am sure Scott McNealy is fully aware that all HW
>needs great SW.
Let's turn the clock back about 5 years. How many CPU architectures were
there (I'm not counting microcontrollers/DSPs)? Intel + 2 or 3 clones,
MIPS, HP-PA, SPARC, POWER + PowerPC, Alpha. 3 of those 6 are dead. Which
one is going to be next, SPARC or POWER? My bet is SPARC. Remember only
10 years ago, HP/Apollo was a serious rival to Sun, and no they're gone.
What will become of Sun if they don't develop/sell hardware? They could
trade on their brand name and become a services company. Or perhaps
migrate to Intel or Power CPUs.
Paul Floyd
<sto...@research.panasonic.com> wrote:
>phi...@bolthole.no-bots.com (Philip Brown) wrote in message news:<slrna4n6dg....@bolthole.com>...
>> On 20 Jan 2002 16:04:34 -0800, sto...@research.panasonic.com wrote:
>> >...
>> >Is killing x86 Solaris a mistake? Maybe, but to be perfectly honest,
>> >it is a real dog on the Intel processor.
>>
>> What are you talking about?
>> "dog" usually refers to speed. As long as you have decent hardware
>> (eg: *SCSI* DRIVES) it runs great.
>> Even with ATA drives, its pretty good, as long as you remember to enable
>> DMA.
>
>Comparatively, Linux is faster than Solaris on the Intel Platform. I
>am not saying that Linux is better only faster. Actually, I am more
>of the opinion that linux is like AIX....ain't unix. :-)
I've not noticed much of a difference between Linux and Solaris, though
I've never spent the time trying to get better performance out of Linux
(turning off all those services that seem to keep launching and
thrashing the disk...).
>Have you ever tried to install Solaris for Intel on a machine with a
>RAID controller card and those fast SCSI drives you are talking about.
The only time I had a problem was with an unsupported video card.
>Any OS is fast on "decent hardware", my point is that it is slower
>than other alternative OSes, it lacks support for a lot of different
>hardware, the price of SPARC hardware is considerably lower and the HW
>is better (I do believe a SunBlade 100 is going for about the same
>price as a "good" Intel PC.), the Intel version is less stable, and it
>takes a lot more engineers and researchers to support multiple
>platforms. So, in the end, killing the Intel version makes good
>business sense.
I'd say the Blade 100 ($1000) has the performance of a $4-500 PC.
[Apple]
Yes, but Apple uses an IBM/Motorola part, so their situation is not
quite the same as Sun.
Alan Coopersmith
|| part of Sun, instead of a joint venture of AOL/Netscape & Sun.
|My understanding was that Sun is now "part of the joint venture" in name
|only. All the resources doing real work are coming from the
|Netscape/AOL side of the table, with Sun planning to or already removing
|their resources from the project. Can you provide more details as to
|the reality?
You've got it almost completely backwards - it's AOL/NS who has pulled
staff, and Sun who has picked them up:
http://news.com.com/2100-1001-274009.html
http://www.vnunet.com/Features/1126302
Many of the iPlanet products are a core part of the SunONE stack,
which is Sun's biggest marketing push at the moment.
Ron
| "Ron" <purp...@earthlink.net> writes in alt.solaris.x86:
| || iPlanet is alive and well - the only recent major change is it's
now
| || part of Sun, instead of a joint venture of AOL/Netscape & Sun.
| |My understanding was that Sun is now "part of the joint venture" in
name
| |only. All the resources doing real work are coming from the
| |Netscape/AOL side of the table, with Sun planning to or already
removing
| |their resources from the project. Can you provide more details as to
| |the reality?
|
| You've got it almost completely backwards - it's AOL/NS who has pulled
| staff, and Sun who has picked them up:
|
| http://news.com.com/2100-1001-274009.html
| http://www.vnunet.com/Features/1126302
|
| Many of the iPlanet products are a core part of the SunONE stack,
| which is Sun's biggest marketing push at the moment.
or at, the time of Sun's first mass layoff (44,000) back in October of
2001. Does anybody have a definitive source as to the reality of this
point that is more timely? Preferably one from =this= year if possible.
AOL/TimeWarner has become a lot more aggressive about the IT market
since last October, and they currently have better financials than Sun.
In short, there's reason to believe those sources no longer represent
reality.
Ron
Date: Mon, 21 Jan 2002 21:50:06 +0000
Hello Bruce,
Apologies for the delay in replying, I've had some other things require
my attention.
As long as this conversation remains civil and germane to the general
interest, I'm going to keep it public unless there's an objection.
| Yup, you got me all fired up. I really objected to the death toll you
| are trying to hand out.
You misunderstand my intent. I want Sun to survive and thrive. As a
knowledgable member of The Faithful I'm seeing signs that they are
making decisions that will have negative consequences to them and
customers.
| I do, as you say, work for Sun (PS in the UK). As a Project
| Engineer/Consultant. I have been working on Sun equipment since 1989 ,
| first as a customer and more recently as an employee.|
|
| As you can tell I feel really strongly about Sun and the products, I
had
| years of people saying how great M$ was and watching their PCs and
apps
| fail all over a factory when the Sun boxes kept going ( I was in
| support/admin of both camps).
I feel equally as supportive of Sun, and I completely concur with your
experiences as related here. Unfortunately, even after being burned,
even after being shown TCO numbers, those same companies keep buying the
solutions with the best Capital Acquisition Cost/Performance ratio
(read: M$ solutions) except in those niches where they absolutely have
no choice. Those niches are shrinking at an exponential rate. If Sun
is only successful in the specialized niches, Sun will cease being a
major market force.
| When I get a bit of time I will be glad to trawl off some performance
| stats. I know a raw cpu test (no app loaded) will show IBM chips as
| faster, but the more important stats have to be the application loaded
| ones and scalability.
Let's not play benchmark games. At the high end, where performance is
all that matters, the IBM POWER4 stuff blows everyone away. Period. In
any space where price/performance ratios matter, Sun HW is =not=
competitive, hasn't been so in a long time, and is less so every year.
You have to go to #70 on www.top500.org before you see any Sun HW (TBF,
this is a list of "supercomputers", which is not a category Sun tends to
compete in). The only Sun HW on any of the www.tpc.org "top 10" lists
is for 1T and 3T TPC-H results (1st and 3rd respectively). www.spec.org
doesn't show any Sun HW to be particularly outstanding either. Now pick
a performance point by any of those metrics (or any other reasonably
objective metric) where there is both a Sun and a non Sun HW offering.
The Sun stuff is inevitably much more expensive. =Sun's HW is not why
people buy Sun=.
It's getting increasingly harder to justify it's purchase on the basis
of superior SW (again, except in specialized niches). Now as we all
know this is old news. The =new= news is that the commodity HW arena is
finally going 64b and SMP, and at relative price points Sun has never
been able to even come close to historically. Unless a miracle occurs,
the days of the vast majority of Sun's proprietary HW market are
numbered.
| Apologies for insinuating you were a troll, but from the first post
| onwards you did extol some troll like behavior.
A strongly held opinion politely put forward and defended is not
troll-like. I never suggested that "troll like" behavior was desirable,
admirable, or that anyone else should act like a troll. Which is what
"extol" implies. I'm providing support for my hypothesis every step of
the way in this discussion. Anyone else who wants to weigh in on this
topic, pro or con, should do the same.
I look forward to you, and others, contributing more to this discussion.
Ron Peacetree
YTC#1
>
> "Alan Coopersmith" <al...@CSUA.Berkeley.EDU> wrote in message
> news:a2mpl5$2jfv$1...@agate.berkeley.edu...
> | "Ron" <purp...@earthlink.net> writes in alt.solaris.x86:
> | || iPlanet is alive and well - the only recent major change is it's
> now
> | || part of Sun, instead of a joint venture of AOL/Netscape & Sun.
> | |My understanding was that Sun is now "part of the joint venture" in
> name
> | |only. All the resources doing real work are coming from the
> | |Netscape/AOL side of the table, with Sun planning to or already
> removing
> | |their resources from the project. Can you provide more details as to
> | |the reality?
> |
> | You've got it almost completely backwards - it's AOL/NS who has pulled
> | staff, and Sun who has picked them up:
> |
> | http://news.com.com/2100-1001-274009.html
> | http://www.vnunet.com/Features/1126302
> |
> | Many of the iPlanet products are a core part of the SunONE stack,
> | which is Sun's biggest marketing push at the moment.
> Both of those sources are dated from a time period just slightly before,
> or at, the time of Sun's first mass layoff (44,000) back in October of
Can I just say it *wasn't* 44,000 staff ! That would be more like 100%
not 9%
--
Bruce Porter
Paul W. Stoecker, Ph.D.
January 4th CNN -- AOL/Time Warner had diluted its earnings based on
problems with integrating AOL Europe and the Termination of the
Sun/Netscape Alliance.
And...that would be 4,000 not 44,000.
It is definite that Sun will absorb all of the iPlanet stuff.
AOL/TimeWarner will be taking the Netscape brower portion. After all,
they already have their own software infrastructure for running AOL.
Paul
Ron
force, in which 9% of the company's 44,000 workers lost their jobs, ..."
from this story
http://www.smartmoney.com/news/ON/index.cfm?story=ON-20020116-000747-180
6.
I slipped a decimal point. The percentage reduction, and the numbers
about Sun's earning statements, are accurately quoted from
http://www.smartmoney.com/news/ON/index.cfm?story=ON-20020118-000309-093
6
My apologies for the mistake.
Ron Peacetree
cjt
Go to the investor section at www.sun.com and listen to the conference call
from a few days ago.
YTC#1
>
> From: YTC#1 ytcREMOV...@ytc1.co.uk
> Date: Mon, 21 Jan 2002 21:50:06 +0000
>
> Hello Bruce,
> Apologies for the delay in replying, I've had some other things require
> my attention.
Snap, I find work and life are like that :=}
> As long as this conversation remains civil and germane to the general
> interest, I'm going to keep it public unless there's an objection.
>
> | Yup, you got me all fired up. I really objected to the death toll you
> | are trying to hand out.
> You misunderstand my intent. I want Sun to survive and thrive. As a
> knowledgable member of The Faithful I'm seeing signs that they are
> making decisions that will have negative consequences to them and
> customers.
Granted, but I still feel your origional claim was way over the top. All
companies make decisons that affect customers and investors , be it
negativ eor positive.
>
> | I do, as you say, work for Sun (PS in the UK). As a Project
> | Engineer/Consultant. I have been working on Sun equipment since 1989 ,
> | first as a customer and more recently as an employee.|
> |
> | As you can tell I feel really strongly about Sun and the products, I
> had
> | years of people saying how great M$ was and watching their PCs and
> apps
> | fail all over a factory when the Sun boxes kept going ( I was in
> | support/admin of both camps).
> I feel equally as supportive of Sun, and I completely concur with your
> experiences as related here. Unfortunately, even after being burned,
> even after being shown TCO numbers, those same companies keep buying the
> solutions with the best Capital Acquisition Cost/Performance ratio
> (read: M$ solutions) except in those niches where they absolutely have
> no choice. Those niches are shrinking at an exponential rate. If Sun
> is only successful in the specialized niches, Sun will cease being a
> major market force.
Granted there is a "niche" rpoble, but if one of those niches happens to
be the best platform to run Oracle on , is iut that bad ?
>
> | When I get a bit of time I will be glad to trawl off some performance
> | stats. I know a raw cpu test (no app loaded) will show IBM chips as
> | faster, but the more important stats have to be the application loaded
> | ones and scalability.
> Let's not play benchmark games. At the high end, where performance is
> all that matters, the IBM POWER4 stuff blows everyone away. Period. In
> any space where price/performance ratios matter, Sun HW is =not=
> competitive, hasn't been so in a long time, and is less so every year.
I did get some stats for you on this late Friday, but forgot to forward
them to myself. But they show that price/performance Sun is knocking IBM
all over the show.
Also I am currently involved in a benchmark test for a customer. It is a
SF6800 v an S680. We are blowing it out of the water.
>
> You have to go to #70 on www.top500.org before you see any Sun HW (TBF,
> this is a list of "supercomputers", which is not a category Sun tends to
> compete in). The only Sun HW on any of the www.tpc.org "top 10" lists
And as I pointed out , it was based on old equipment. Sun have completly
revamped the line in the last 18 months. USIII is now at 900mhz, 1.05
due out soon. USIV and USV are in the pipeline.
> is for 1T and 3T TPC-H results (1st and 3rd respectively). www.spec.org
> doesn't show any Sun HW to be particularly outstanding either. Now pick
> a performance point by any of those metrics (or any other reasonably
> objective metric) where there is both a Sun and a non Sun HW offering.
> The Sun stuff is inevitably much more expensive. =Sun's HW is not why
> people buy Sun=.
Well, it was the spec stuff I was going to show you, I read it somewhat
differntly
IBM is known to adjust test results, I have a friend who has IBM/Sun in
for some testing. He was telling me how well the IBM box was running. I
asked how how many processors he had in the SF4800, he said "4 , hats
what we asked them to supply on test".
And IBM ? "The same".
Strange I told him, I thgink you will find the IBM is running on 6 CPUs.
It was, I was right , IBM cheating again.
>
> It's getting increasingly harder to justify it's purchase on the basis
> of superior SW (again, except in specialized niches). Now as we all
> know this is old news. The =new= news is that the commodity HW arena is
> finally going 64b and SMP, and at relative price points Sun has never
> been able to even come close to historically. Unless a miracle occurs,
> the days of the vast majority of Sun's proprietary HW market are
> numbered.
But Sun is ahead of the game in 64bit, the OS has been 64bit for longer
than most.
Incidently 64bit does not always equal faster, just more addresable
memory. In fact some apps run positvly faster in 32bit.
I can see we are nevr going to agree :=]
--
Bruce Porter
XJR1300SP, XJ900F, GSX750W, GS550, GSX250, CB175
WUSS#1 , YTC#1(bar), OSOS#2 , DS#3 , IbW#18