Slashdot Mirror
Ex-Sun CEO Warns Oracle of Death By Open Source
gearystwatcher writes "Former Sun CEO Scott McNealy talks to The Reg on where things went wrong, and acquisition by Oracle: 'We probably got a little too aggressive near the end and probably open sourced too much and tried too hard to appease the community and tried too hard to share,' McNealy said. 'You gotta take care of your shareholders or you end up very vulnerable like we got. We were a wonderful acquisition — we got stolen for a song at the bottom of the Dow.'"
Definitely, if all the valuable assets of your business is in software (Solaris, StarOffice, Java, etc) and you give away such software for free then your business does not make sense at all.
Ubuntu is an African word meaning 'I can't configure Debian'
Right, the only mistake Sun did was open-source too much. Like all the closed shop were doing wonderfully well too.
Thanks Sun.
I need another CEO job and I can't get one!
Have you fscked your local propeller head today?
"...while he's never read Atlas Shrugged, McNealy cites its author Ayn Rand as his mentor while he was growing up. Rand is a hero to those on the political right "
Interesting...
He who knows best knows how little he knows. - Thomas Jefferson
Sun should not be charging for their software.
They can make their money on performances and T-Shirts.
Sun's biggest problem was that its various flagship products were out-competed or unprofitable. On the high end hardware, IBM could build better mainframes. On the lower end hardware, Dell could build cheaper workstations and servers. On their Unix, Linux became as good as or better than Solaris. And Java, while nifty, had no way of turning a profit.
By open-sourcing its software offerings, Sun ensured that while its business was screwed, its legacy lives on.
I am officially gone from
> We were a wonderful acquisition — we got stolen for a song at the bottom of the Dow
Translation (spin removal) - I screwed up - and would now like to thank the Gods for my "golden parachute". Since I think a suitable time period has passed (hey, it's 2010 and people have an attention span of 2 minutes or less, besides nobody who will live much longer than another 2-3 years even knows who Bill Joy, or a SPARC let alone a 360 was), it is okay for me to now attempt to twist and distort history so the world doesn't remember me as "the bloke who fsked up, big time and killed off one of the last bastilions of real technical people who "got it"."
- Yeah, I fsked up BIG TIME, but you can' t prove it and my name isn't Julian Assange, so after tomorrow you won't remember anyway.
> with total knowledge of the company's inner workings and financial statements ...who ran the company into the ground.
> with no business/management experience whatsoever are.
Of course that is blatantly untrue. We are not without experience. We have
the most relevant experience of all. We're the ones who have been in the trenches
watching as Sun did this to itself.
We are Sun's customers.
Of course in your sort of MBA-cult mentality the actual customers don't care.
A Pirate and a Puritan look the same on a balance sheet.
That's strange. Red Hat does all via Open Source and is about to pass the $1 Billion mark. Sounds like to me McNeally was a very poor CEO and it had nothing to do with the things they Open Sourced.
My karma is not a Chameleon.
Right, the only mistake Sun did was open-source too much. Like all the closed shop were doing wonderfully well too.
The summary is incomplete. Somewhere else in the interview he mentions that one of his regrets is not open sourcing Solaris earlier, claiming it was better than, and could have beaten Linux. His point is that they didn't have a good business model and didn't make enough money from the open source, but he also clearly still believes open source can be profitable, and open source was the right direction for Sun.
Oracle is not nor has ever been about opensource. It's about making money. Unlike Sun, if it doesn't make money (either directly or indirectly), Oracle will dump it. Oracle knows why it's in business, who its customers are and they are not developers, the community, consumers, or small businesses. They are large companies, the government, and other big institutions.
"The problem with socialism is eventually you run out of other people's money" - Thatcher.
Except, this CEO didn't exactly manage things to a positive result, did he? And if I were asked to guess who understands open-source business models more; Slashdot readers or CEOs of companies that failed to capitalize on open-source business models? I'd go with the Slashdot readers every time.
I notice that your Slashdot UID is 1378985. You have clearly concluded that there are almost 1 and 1/2 million Slashdot readers, all of whom have no business experience and troll from their basement. This in spite of numerous recent articles about the fact that the mass of linux kernel and other major Open Source project developers are paid developers, many of whom work for very big name companies in the high technology industry. You have clearly forgotten who has the experience to analyze McNealy's position, and whom the guy without the experience opening his mouth out of turn is.
News Flash: There are many, many, many people in the world more qualified to analyze where Sun went wrong with their approach to Open Source than Larry Ellison and Scott McNealy, and some of them are right here on Slashdot. In fact, the decline of Sun could be viewed as specific evidence that there was a lack of understanding about Open Source on his part. HP, IBM, Intel, and many other big name hardware/software companies seem to have managed to keep on without losing their shirt in the process, for example.
Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
Google is doing it right. Its google-docs does not do much, in terms of bells and whistles it pales when compared to Ms-Office. But it is well positioned based on a simple truth. 90% of the people need only 10% of the features of full fledged Ms-Office. Give that 10% free and effectively deny Ms-Office the mind-share of 90% of the people. Force Microsoft to interoperate with a significant part of this 90%. Give customers of Microsoft some ammunition in price negotiation. Anything that will make Microsoft play defense in the Office arena, is the resource it can not spend in fighting Google. It is ably helped by Microsoft that has promoted to leading positions people who won the corporate desktop market. Like Civil War generals fighting the war using Napoleonic tactics against machine guns, or the WW-I generals fighting that war using Civil War lessons, the management of Microsoft is fighting the consumer market war using corporate desktop war tactics.
Coming back to Sun, it was effectively done in by amortization. The cost of development and research of intel chips was spread over so many more customers compared to the sparc chips. The same way cost of development of Windows was spread over a much larger number of customers. When there is an order of magnitude difference between you and your competitor in terms of potential for amortization of cost of R&D, you should have the vision to react early and react decisively. For all the high salaries paid to these MBA types, they did not see it coming.
I'll grant you I am Monday-morning-quarter-backing. But I not getting Sunday-after-noon-quarterbacking salaries either. Scott McNealy got paid to see this coming. He failed. Miserably.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
What about death by alienation, because you shit where you eat? (ie on your open source developers)
As for Scott's comments, I don't see Redhat in the same free fall Sun was in. Of course Sun was passive aggressive towards the OS community in many aspects. Redhat for the most part hasn't been.
Just don't forget who the CEO was who failed to adapt Sun after the dot com bust. What's your point? That basement dwellers with no business or management experience would not be able to lead a company to its destruction like McNealy did?
It seemed to me that Java was one of Sun's biggest problems. Sun managed Java badly, and that bad management was very bad public relations.
Just don't forget who the President and then CEO is, who oversaw the nose-dive of Sun shares to that being scooped from the bottom for a bargain price that he mentions. In fact, Oracle paid too much: Sun was already only worth some 3 billions at the time. Between November 2007 alone and November 2008, the total worth of Sun shares dropped by 80%. And Schwartz was compensated to the tune of over 11 million dollars for that nose-dive Sun took in 2008. 'Cause, you know, nothing says "the CEO deserves a big fat bonus" as driving the company in such a nose-dive where it lost 80% of its worth. The guy who oversaw in 2004 the cancellation of Sun's CPU designs and most R&D, followed predictably by a spike in profits and then by the crash we all saw.
Just about the only thing that can be said in Schwartz's favour is that at least it wasn't as bad as Scott McNealy's strategy of just foaming at the mouth against Microsoft and bipolar swings between "we love teh LINUX!!!" and "Linux is teh EVIL! DIE! DIE! DIE!" often within the same day, instead of telling customers why they'd want to buy Sun gear. No, big companies don't give you money just to fight Microsoft, Scott. Which strategy resulted in the drop in Sun share value by a factor of TEN TIMES between 2000 and end of 2003. (With an even deeper dip in 2001.)
Even giving ample allowance for just having the deck stacked against Sun, Schwartz just didn't prove that he's the genius CEO who can pull it out. If the guy had taken Sun and doubled its share value, ok, I'd listen to what he has to say and take notes. But being the guy who sat there while the company continued going downhill is hardly some kind of credentials. My cat could sit there and watch the stock roll downhill. And my cat wouldn't even need 11 millions compensation for that job.
So you're asking me to do, what? Trust him just because... what? Is he some kind of nobility that us peons have to trust and never question? Really, WTH is he? The High Priest of the Sun? Well, ok...
Can I believe that some random nerd could know better? Actually, yes. Heck, I'd even trust my barber to have a thing or two to teach Schwartz.
A polar bear is a cartesian bear after a coordinate transform.
I'm not an expert and I don't follow these things so take my opinion as is.
I remember people talking in the early 2000s when Sun was trying to be the "dot in the dotcom" that Sun lost to linux. Many companies simply didn't need a big Sun box and now that they could get something unixy on a box more appropriate to their business that is what they did.
That could have happened if a proprietary business one day saw a niche for a unix operating system on a small machine.
Sun was all about selling big machines and their OS for those big machines.
Aside from that, I think he has a good point about thinking carefully when aiming the open source gun.
Many companies make it hard to buy something...at least for programmers.
If somebody can download something for free and have it work fine, they will not bother to go through the channels in their company to get them to buy something.
Sun succumbed to the same thing that felled SGI, namely the boom in commodity computing. Sun made some really great products, the problem was that they also that made products that were really expensive. Back in the day when the difference between the high end and commodity was significant enough that a lot of companies were willing to shell out the money for the primo shit. However so called "commodity" computing(both hardware and software) has eventually caught up and a lot of companies could no longer rationalize the difference between Sun's stuff and the much cheaper products.
For instance 2 years ago we were looking for a new RAID and were considering Sun's ZFS storage appliance but the $10k for 2 tb was just waaaay to much money for the tiny extra bit of redundancy we could get. It was cheaper to just buy a much bigger raid, split it in 2, and do an rsync. Not the greatest situation in the world, but ultimately it saves a lot of money. Sun just could not compete for anything but a relatively tiny niche market while having massive amounts of capital tied up in labor and facilities.
Monstar L
Even after death Sun still doesn't get it. Here's where you went wrong Sun: 1) You were expensive! REAL expensive. Your hardware, your support, and your software were too pricey. You never adjusted as Linux came on the scene. 2) You were WAY behind! Look at how long you keep the bare-bones and clunky OpenWindows as your desktop. Look at how pathetic the "new and improved" CDE was compared to the other desktops of the time. 3) My opinion is that Sun didn't hire enough new engineers. They were too content to keep computing have the "dusty" old feel that companies like Apple had the foresight to shed. Too bad. It was really bad leadership, and this proclamation shows that they still don't get it.
Oh, my bad, turns out that the muppet in the summary actually is McNealy not Schwartz. Sorry. Now that's even more fun. As I was saying, McNealy is the guy who was just foaming at the mouth against Microsoft and was having split personality fits about Linux and OSS, while Sun was pretty much imploded. There were people jumping ship _because_ Sun had abandoned almost any pretense of having a product to sell, and was just telling everyone why they should give it money to fight Microsoft.
And who alienated the very same OSS gang he now talks about, with his schizophrenic swings between professing his love for OSS and Linux in the morning, and foaming at the mouth against it in the evening. I don't think that guy ever really either understood OSS or embraced it half as much as he tries to make it sound.
It's like, dunno, take my criticism of Schwartz above and make it times ten.
A polar bear is a cartesian bear after a coordinate transform.
We're still using older Sun hardware in our data centers as Oracle servers. When Oracle upped the licensing prices for Oracle on Sun SPARC multi-core chips, we decided that upgrading the hardware would be too expensive due to the increase in Oracle licensing costs and stuck with the old gear even while upgrading Solaris and Oracle on the hardware. We also began looking at new, replacement hardware. Our new servers that I'm working on now are Dell R710's running Red Hat 5.4.
Red Hat does quite well giving away its open source OS and apps as fast as it can. That's not what devalued Sun.
What devalued Sun was that its CEO, McNealy, was unable to run such a company. He kept proprietary products like Solaris propped up for years longer than they had a market among competitors like Windows and Linux, even as primary competitor IBM deprecated its proprietary OSes to embrace them both. Then McNealy punted on Solaris, opening its source only when there was no demand for it. Sun's Solaris business didn't get taken by competitors copying Solaris' source or anything like that. In fact, opening the source kept it going for years, even if it was too little, too late to save it. Especially with the CEO failing to actually embrace open source, but rather seeing it as a dumping ground for nonproductive assets instead of a hothouse to grow those assets into productive centers to be monetized.
McNealy is like any failed CEO whose failure was trying to control something better developed by letting it go more: blame the "liberals" ("liberal" means "free from control"). If McNealy can blame open source for his own failures, he might find new income from the many other incompetent businesses that need a scapegoat like open source to hide their own failures. And in today's corporate world, especially America's, there is no higher demand for anything than for a scapegoat.
--
make install -not war
Let me get this straight: Ayn Rand was his "mentor", yet he's never read her what is considered by most (and Ms. Rand herself) to be her "flagship" novel? The Fountainhead was merely a warming-up exercise to Atlas... Personally, I think both novels are awful, but I've never been a big fan of polemic, no matter which side of the political spectrum it falls on. (And too many good authors fall into the trap of thinking I give a $hit about their political views and let their books suffer greatly as a result; e.g. Clancy after "Cardinal of the Kremlin".)
Maybe he has political ambitions, and professing admiration for Ayn Rand is just a checkbox he felt obligated to fill out...
SirWired
former Fed Chairman Alan Greenspan was a big fan of her
That's quite a laugh, seeing how Greenspan was precisely at the top of the power pyramid that Rand was vehemently opposed to. Without a doubt, Rand would be 100% against a situation where consolidated power (i.e. government) is master of the economy, where an elite few have the ability to pull financial strings to benefit some at the expense of others, or impede the free market via one central point of failure.
More likely, Greenspan merely cherry-picked from her philosophy where he thought it would aid him politically. Calling him a "big fan" is naive to say the least. A "big fan" of Ayn Rand wouldn't be in the business of government in the first place.
So when he said "We probably got a little too aggressive near the end and probably open sourced too much and tried too hard to appease the community and tried too hard to share" he was just lying like a weasel, as his contradictory hindsight also says he should have done more of what he did "too much".
Open source wasn't the problem, as he freely admits. Doing it too late was the problem. By the time it was "near the end" it was too late to "take care of the shareholders" by doing anything different. Open source was the only thing keeping Sun relevant near the end, and therefore the only thing taking care of the shareholders.
McNealy screwed up, as everyone watching Solaris sink could tell. He should have opened the Solaris source, ported it to Java running on every CPU but optimized for highest speed on Sparc - and then maybe Xeon. Should have made Java applets actually work on every CPU/OS/browser, the way Adobe did Flash, and bought Macromedia instead of Adobe getting it - or just competing with it. So many things he could have done if he'd managed for the 2000s instead of the early 1990s. Now he's just a whiner whose day is long gone.
--
make install -not war
I don't think it's even about rich or non-rich. What Ayn Rand does isn't as much a defense of being rich, as a defense of psychopathy and of not giving a damn about the others or their well being.
And while in her writing she does somewhat tone it down, in her diary she was going all fangirl over people like William Edward Hickman. That was her ideal of superman and she loved a quote from him saying "what is good for me is right."
Just to make it clear, what William Edward Hickman was famous for was kidnapping a schoolgirl and mailing her father taunting ransom notes signed with names like "Fate" or "Death". Then when the father came with the money, and thought he saw his girl sleeping in the abductor's car, she got thrown out of the car... dead. Hickman had cut off her limbs -- by his own testimony, _alive_, as the blood was coming out in small spurts, i.e., the heart was still beating -- hollowed out her torso and strewn her inner organs all over town. Actually living out an earlier fantasy he had told a former accomplice about, to take someone apart and chuck bits of them all over town.
Ayn Rand thought Hickman was some kind of dashing romantic adventurer whose only "crime" was rejecting the unreasonable conformism of society. (Like, you know, not taking live children apart.) She pretty much foamed at the mouth against those boring sheeples who dared so self-righteously criticize her hero. A bit later she blames society for basically not offering him anything better to do than gut and dismember a little girl. I mean what was the poor guy supposed to do? Get a boring job and a boring wife and all that? No, really. That's her justification for Hickman.
And really, that's what her writing is about. Even the economic angle is Bullshit with a capital B. I mean, her utopia needs an infinite free energy source to even function. But she manages to do a heck of a job in lionizing the psychopaths who doesn't give a damn about anyone else, and calling those "statists" and "collectivists" names, and fantasizing about their destruction.
Now consider that a large number of those at the top _are_ psychopaths. See, for example: Is Your Boss A Psychopaths?
If you were one, wouldn't you just _love_ a philosophy that says it's just normal to not give a damn about anyone else, and that it's an _objective_ (or Objectivist) fact that it's all about caring for number one?
A polar bear is a cartesian bear after a coordinate transform.
In 2007 Sun had that thingie, if you would fill in a form, they would send you solaris. It's 2010 now. Almost 2011. Despite numerous mails from my side, I am still waiting for Solaris.
IBM's success with Java pretty much proves that it was Sun's management of java rather than Java itself that was the problem. On the same note, IBM's success with Linux pretty much proves that McNealey's whole rant makes little sense.
But why would anyone follow his advice after he ran Sun into the ground?
There is a war going on for your mind.
The privacy (of source code) is dead, get over it.
'cause that's a big concern... That Oracle will open source too much code.
this signature has been removed due to a DMCA takedown notice
You keep on repeating this "basement dweller" nonsense like some article of faith.
We're the ones that have been in the trenches for the last 20 years.
A Pirate and a Puritan look the same on a balance sheet.
Except, this CEO didn't exactly manage things to a positive result, did he?
Scott McNealy started a company 30 years ago and was CEO for 22 of those. When he resigned Sun employed 38K ppl. Sun was bought this year for $7.4 billion. You may not like that it was bought, you may not like who bought it, but I would call that a positive result. If you can do better, please do.
While Slashdot certainly has its share of ignorant blowhards, there is something to be said for distance as an analytical tool...
Until androids replace humans in management, the people with the most intimate knowledge will also be the people with the greatest emotional investment. Sometimes somebody with less detailed knowledge; but less personal investment, can be useful...
Sun was the leading provider of workstations and Unix servers and could do no wrong - until the Unix wars. Rather that unifying the market, the Unix wars fragmented the vendors and opened the door for someone else to walk in with a hardware-neutral OS. That was Microsoft with Windows NT.
I was deeply involved in the CAD world and all the good apps ran on Unix. PCs were low-power toys. Customers were tired of supporting different platforms for different apps and wanted a single OS for all platforms. Just as it looked like it was close to happening (the Year of Unix), it turned into a battle between AT&T/Sun and OSF (Open Software Foundation - DEC, HP, SGI, IBM and others). Sun chose the (annoying to me) OpenLook interface, while Microsoft offered OSF a Windows-lookalike interface called "Motif". Microsoft supported OSF, while working on Windows NT to replace Unix. Upon adding SGI's Open GL interface to NT, CAD vendors began to see NT as an alternate to Unix. When Solidworks appeared on NT, the race to port Unix apps to NT was on.
So I think Sun wasted their energy in trying to beat their Unix rivals while Microsoft focused on what customers wanted (one OS for all) and won.
(Yes - in those days MS focused on customers.)
The funny thing is the Unix wars were won by an open-source product - GNU/Linux. Image where Sun could have been had they united the Unix world. McNeely lacked vision and let Gates take it all. Pretty stupid. Sun was an asshole company at the time and treated us like jerks. Karma is a bitch, ea?
Place nail here >+
I left Sun back in the early 1990's - it was quite apparent to me even then that McNealy was going the wrong way and ought to have been replaced by the board of directors.
There are a large cadre of former Sun people who felt that Sun could have been a contender but that it was stuck in quicksand created by its own top management.
Forgotten also is the damage done to Sun by its relationship with AT&T back in that era - all of that caused years and years of internal strife over machine architectures [anybody remember Sun OS on the AT&T 3-B Uselesses?] and System V versus BSD. Nobody ever seemed to notice Microsoft growing and taking customers.
Sun was already dying before the opened up Java or Solaris. But lets blame something else than poor management.
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
The dot com boom, and Jonathon Schwartz. When the dotcom bubble broke, Sun was suddenly in the position of having to compete with an enormous inventory of used SPARC servers coming on the market. They didn't cut their overhead enough to deal with the double-whammy of reduced sales and competition from much cheaper intel-based servers. Add to that, a CEO who was tragically out of his depth, and the result was inevitable. The one thing I can give Sun's management credit for is selling the rotting carcass to Oracle for at least double what it was worth.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
A while ago, Polaroid and, I guess, DEC before it, all had the same problems. They had a great body of work, but they could not capitalize on it. At Polaroid, I worked on their electronic still camera. The project took a distant back seat to their "instant film" division because that's where the income was. The current cash cow was too powerful within the organization to allow it to develop and market potentially competing technologies. The problem is, if you don't modernize your product line and, perhaps, risk exiting products, your competitors will, and in the case of Polaroid, they did.
Polaroids "ESC" camera we were working on was the best of its time. The color was good, the resolution was great (for the time), and the speed good too. Polaroid starved the project because they didn't want jeopardize their instant film. So, polaroid died when digital cameras took over.
Similarly, Sun had great technology, but it seems as though they could not decide if they were a hardware company or a software company. Their hardware was good, but not competitive. Their software strategy was flawed. They could not commit to being "open source" or being "closed source," instead tried to straddle the middle ground decimating the value proposition for their closed source, and generating distrust on the open source offerings.
Maybe it isn't exactly like polaroid, but these were two companies with a huge amount of intellectual capital that were unable to monetize it.
I sometimes wonder what would have happened if Sun had published Solaris under the GPLv3 in 2007. With the Torvalds set on keeping the Linux kernel under GPLv2, would Free Software and Open Source have split as the more Libre-centric folks put the Gnu environment on the Solaris kernel? Would Debian Gnu/Solaris have really taken off?
It would have been late for Sun, but it sure would have been interesting.
Right, the only mistake Sun did was open-source too much. Like all the closed shop were doing wonderfully well too.
not open sourcing Solaris earlier, claiming it was better than, and could have beaten Linux.
LOL. This is why their company failed. Their CEO is an idiot.
Oracle paid too much:
Possibly but that is debatable. Sun did have a lot of pretty useful assets. Not all of them profitable but I think that had to do more with Sun's particular business model than the parts of the business. IBM sells similar hardware and software profitably and Sun's software assets are valuable at minimum as strategic weapons. In the right hands these parts of the business could be quite valuable.
Sun was already only worth some 3 billions at the time.
By what measure? You are approximately right if you are only talking book value of assets. But no one values a company that way except in liquidation. Market cap? Unless the company is in even worse shape than Sun was the management won't sell without some premium over the stock's current value. Valuation is very inexact. You can't simply say a company is worth $X billions. It doesn't work that way. At best you can give a range and even then it depends on who the buyer is and what the circumstances surrounding the sale are. When finance geeks (and I am one) value a company they use several different methods (NPV, multiples of EBITDA or revenue, etc) and even then the valuations they get almost never match.
'Cause, you know, nothing says "the CEO deserves a big fat bonus" as driving the company in such a nose-dive where it lost 80% of its worth.
He might deserve a big bonus if the alternative was that the company would lose 100% of its value. I'm not convinced that is the case here but the circumstances matter. Sometimes saving a company from bankruptcy is positively heroic. Personally I think McNealy's team (it's never just the CEO who is responsible) screwed the company and Schwartz really never had much of a chance; the damage had already been done. I'm not sure anyone could have saved Sun as an independent company but Schwartz certainly wasn't the guy to do it. End of the day selling the company was probably the right move but they should have done it sooner.
Which strategy resulted in the drop in Sun share value by a factor of TEN TIMES between 2000 and end of 2003. (With an even deeper dip in 2001.)
Pretty much every tech company's stock tanked between 2000 and 2003 after the dotcom bubble burst. Amazon went from over $100/share down to less than $5. Ebay went from over $30 to around $8 (split adjusted). Even Microsoft lost about half its market cap in that period. Sun just had more exposure than most so they benefited more during the bubble and crashed harder afterward. Hard to compete with your own gear being sold on eBay for pennies on the dollar.
That's strange. Red Hat does all via Open Source and is about to pass the $1 Billion mark. Sounds like to me McNeally was a very poor CEO and it had nothing to do with the things they Open Sourced.
If you are going to use dollars as a benchmark then Sun was far more successful than Red Hat even at Sun's lowest point; Sun was sold to Oracle for over $7 billion. Don't mistake this for a criticism of Red Hat. I'm only challenging using dollars as a benchmark.
Hah, McNealey's blaming FOSS for his own management shortcomings. The bad thing is people not in the industry (not in IT and more in Financials) will read this crap and come to the conclusion that FOSS == BAD for Business. I think its more a case of McNealy == BAD for business.
Python: 'And then suddenly you have a language which says "we're all stuck with whatever the whiniest coder wants".'
That statement sent shivers up my spine.
The company died on the vine with McNealy (and his ego) at the helm. Sun never fully recovered from all the equipment they virtually gave away (leased or financed) to companies during the telecom/Internet bubble. When he finally decides to step aside, he allows the promotion of Schwartz, an engineer with little successful business experience, to CEO while he continues to pick up a paycheck as chairman of the board. The idea by Schwartz (with McNealy agreeing) to start giving away their software and make it up in support has got to be the most asinine. My experience with companies is they usually do not pay for support if there is no compelling reason, and post bubble, they had little incentive to do so.
Sun suffered from a NIH syndrome throughout its life, pushed primarily by McNealy. The company also had a history of acquiring companies only to EOL products months later; it is amazing how many purchases were poorly timed.
Open sourcing Solaris is a red herring. The biggest problem was the Intel version played second banana and its constant on-off-on development was not helping matters. If they slashed the price of the Intel version to a reasonable level, while offering unrestricted use on multiple processors and no user limits, it would have been a win. Instead, they tried to get people to buy their Sparc equipment.
Sun is a good case study on a business that refused to adapt.
The one thing I will say about Ellison and Oracle, they can squeeze blood out of a rock.
Slashdot - The great and glorious cluster fuck of Internet wisdom.
At one point after the dotcom bust, SUN had the second most engineers dedicated to designing CPUs. More than IBM and certainly more than AMD, only behind Intel. Those engineers in the Valley aren't cheap and what does SUN have to show for all the work of those engineers? Back in 2000/2001 time, I had a friend that worked for SUN and he didn't think that they would last another 5 years. Their overhead was just killing them with little progress in converting to a service company.
It's a case of steering the ship before stopping the leaks. Eventually, all the leaks will sink your ship.
Did anyone ever try to buy things from Sun?
No other company I ever worked with made it so hard. Unless you were a megacustomer, it was actually fairly difficult to actually buy anything from them.
In contrast, buying RedHat on the small scale is click, click, done.
Here's a summary of Ellison's rant on why Sun died, notice the complaints are mostly about sales and engineering decisions, open source had very little to do with it:
http://blogs.barrons.com/techtraderdaily/2010/05/13/oracles-ellison-sun-execs-were-astonishingly-bad-managers/
Blessed are the pessimists, for they have made backups.
IBM has seen open source as a means of creating "solutions" for customers not a money maker in of itself. Ibm views linux as a sturdy and inexpensive tool that it can spend money to become very expert at. Sun sells expensive tools, IBM sells solutions to customers needs using inexpensive tools. That is why IBM is very very rich.
He's just an old fart.
He's been not getting it for many years now.
All he can do is to make the retirement time fun.
Interestingly, IBM and Oracle got more value out of using Java than Sun did by writing it. That appears to have been one of the decision points for Oracle.
One of the extra advantages of Linux for IBM was that it offered a new OS for the 390s, and of a very popular flavor. Sun already had a Unix OS for SPARC, so they didn't get the added value.
Sun was, IMHO, always a "BSD vs Bell" shop: they understood the struggle to free BSD, and learned how to deal with Bell and the commercial world, but that's where they stopped.
--dave
davecb@spamcop.net
"... people become rich and powerful because they're scum-sucking leeches who like to steal from everybody else"
[citation needed]
AFAIK, this kind of language Marx reserved only for other contemporary philosophers of his time (critics and competition). He actually claimed the other way around: that, to paraphrase your pseudocitation, people become scum-sucking leeches who like to steal from everybody else because they are (and would do anything to remain) rich and powerful.
So, there you are, it is not one but two-dimensional difference between Ayn Rand and Karl Marx: first dimension is whether character precedes and causes social situation or the other way around, and second is whether correlation of personal worthiness to social standing is positive or negative. Personally, I think both POVs are gross oversimplifications, and there is obvious problem of defining "good and valuable person" without circular argument, although their reasoning is worth studying.
So, if she's a hero of the political right, then the political right approves of adultery and malicious behavior.
Most conservatives would probably claim they don't approve of these things, yet your theory would explain why so many of them still seem to like Newt Gingrich.
Space game using normal deck of cards: http://BattleCards.org
IBM's success with Java pretty much proves that it was Sun's management of java rather than Java itself that was the problem.
On the same note, IBM's success with Linux pretty much proves that McNealey's whole rant makes little sense.
Is that the success of running Linux on high-margin hardware like POWER and mainframes? And also charging consulting fees for suggesting these solutions? And also the profits from licensing the application software that runs on these systems?
How is this related to Linux, and not to the old adage "no one got fired for buying IBM"?
He isn't, if you RTFA.
He's saying that Sun saw the writing on the wall for closed systems, and jumped on the bandwagon of opensourcing all they could.
Unfortunately, it would seem they did so without any cohesive plan as to how they'd turn that into a decent source of revenue.
When Scott McNealy went on an open source rampage there were many warning him that it would harm Sun if too much key software was released. Maybe open source can work for a major business, but McNealy was warned several times to slow down. Two years later Sun gets sold for a bargain and his legacy is Oracle's problem now.
Unfortunately, it would seem they did so without any cohesive plan as to how they'd turn that into a decent source of revenue.
Which is a management failure is it not?
The point is that Sun didn't "open source too much" as McNealy says. They screwed everything up by thinking that just throwing the code in the wild once in a while is good enough. It's not, that's why Novell forked OO.o and other open source projects from Sun did as they did.
I couldn't disagree more with that statement, and not just for ideological reasons. By the time that Sun started really embracing open-source, Sun had been abandoned by everybody except for its longtime customers who already had a very substantial investment in the platform.
A lot of Sun's open-sourced projects attracted a great deal of attention by myself, and many others in the Sysadmin world. Projects like ZFS, and Xen/xVM made it pretty clear that Sun had some of the best people in the industry working for them, and that Solaris was probably worth reconsidering, even if it meant being coupled with Sun's expensive hardware (which is a drop in the bucket compared to the extra staffing costs associated with a high-maintenance server platform). Also, the existence of OpenSolaris meant that we could take the platform for a "test drive" on some old hardware before taking the plunge.
By the time that Sun had won us over, the writing was already on the wall w.r.t. the Oracle acquisition, and nobody would go near Sun with a 39 1/2 foot pole (which, as we've found out, was a perfectly justifiable paranoia).
-- If you try to fail and succeed, which have you done? - Uli's moose
It is, but he's not blaming open source for it. If anything, he's blaming himself.
IBM had the same problem in the 91 recession Sun had recently. But IBM had enough succesful lines to make the switch.
Just because IBM has a lot of open source support doesn't mean they don't sell a lot of commercial software (like Dassault, a partner they had invested heavily in for a long time)- they have gone from a mostly hardware company to mostly a services company with hardware offerings (HP is similar now that they bought EDS). Services and support is the real cash cow for them, as it was for Control Data back in its day (before their management put the company in a tailspin of spinning off profitable divisions to keep stock above junk status... and gee, EDS did that too...).
Well, Sun did TRY to make applets work on every platform. They failed. Java as a whole was, much like the IBM PC, a piece of junk. The key was that it was a piece of junk that had big corporate backing. This got enough people using it that it remained relevant long enough to improve.
At the time of it's release, it was just a crappy emulator without a reference platform to point to to say "that is correct behavior", so it ran differently, not only on different systems, but even on the same system depending on what it was embedded in. I remember running an applet on a system in IE, Netscape, and Lotus Notes. On the same system, it looked different in each application.
It's not, that's why Novell forked OO.o and other open source projects from Sun did as they did.
And look at where Novell is now.
Why should he have ported Solaris? Sun was a machine company that sold high end boxes. There was and still is quite a nice market for large systems. There was and still is quite a market for reliable systems and strong hardware service arms.
Huge numbers of companies are moving to virtualization, Sun was a leader of Unix virtualization. Today I think they could be selling huge numbers of boxes if they offered a compelling value. What is the value of Solaris outside of Sun hardware?
All this noise and nobody mentions copyright assignment. Everybody knows the star wars quote by pricess Leia to Tarkin "the tighter you grip the more star systems slip through your fingers" or something to that effect. So far Oracle looks like absolutely destroying the Sun legacy.
Yet Adobe succeeded.
I think you exaggerate both the difficulty of that task, and the degree to which Sun tried. The hardest part was probably dealing with hardware UI differences, like Apple's 1-button mouse (and Unix/Linux's 3-buttons), but that kind of thing hasn't proven prohibitive in the years that so many outsiders have worked on improving it. Thus, by the way, demonstrating the lost value of open source to Sun.
--
make install -not war
Open Source business model works well when your company model is a consulting based business. Suns model was product based business. IBM doesn't do many products any more, they offer services and consulting this works with Open Source. Sun made products this didn't work with open source. So McNealey view is apt.
If you are in business of making products to sell to others Open Source is a threat.
If you make a service or consulting on using products then Open Source is an asset.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Because then people would have kept buying Solaris, not just the Sun boxes (and asking for Linux on them). Sun could have been the leader of Unix virtualization on an open source Solaris, and would have blown Linux away. Open Linux, open Java, OpenOffice.org , every developer captured in "the Sun way" - and Sun simply ensuring the quality of those products, selling Sparcs (and PCs) in virtualizing racks, selling site licenses with support, training, all branded "Sun". But it tried to cling to proprietary software as open source was devaluing that (unless you had a Microsoft type monopoly, which Sun never did), then threw up open source too late, and not enough, and in ways that alienated the developer community instead of integrating with it. The opposite of the right strategy, except in lip service. And not just in retrospect: IBM has done exactly all that, with Java.
The point is that Solaris could have been Linux: the hugely popular open-source OS that ran on CPUs and machines from high to low, because it captured the exploding population of developers. Instead years passed while Linux eclipsed (pun intended :) Solaris. Then Sun, which made a big deal out of porting it to x86 but missed that market entirely, finally open sourced it as a halfway step to dumping it. And effectively dumped it.
--
make install -not war
Solaris was Linux. Read /. from the 1990s or even the early 2000s. The goal of Linux was to offer something like Solaris.
I don't know of anytime that people bough Sun boxes and asked for Linux on them. Even as sales were dropping off fast everyone agreed Solaris on Sun hardware was better than Linux on Sun hardware. And if they did it would be like Apple's attitude towards someone who wants to run Windows on their hardware, "have at it".
In the early 2000s Sun was still selling. The problem in 2005 was that Sun hardware wasn't much better. IBM beat them on the high end and HP, Dell on the low end. They were squeezed into a narrow and narrower hardware range.
Most Linux software compiled into Sun well. They offered CDs loaded with Open Source apps. I think they should have done this earlier but it was the hardware that killed them. Arguably they might have done far better by keeping Solaris as a niche OS for high end end hardware that was close enough to Linux to run Linux apps (recompiled). QNX has done well for 12 years competing with Linux as a nice for people who need a little more on the embedded side than Linux can offer.
IBM didn't create Java or Linux.
So because the goal of Linux (you say) was to offer something like Solaris, therefore Solaris was Linux?
Linux was open source, a kernel added to the GNU userland, owned by no one, sold by many, developed, distributed, marketed and supported by many more. Solaris was none of those: it was a proprietary product developed and available from only Sun, and until it was too late to matter running only on Sun HW.
I'd point out that McNealy could have transformed Sun into a real competitor to MS (plus with a Sparc/HW biz up its sleeve), but you just tried to argue that Solaris was Linux. Absurd anyway, but backed up by "read /. from the 1990s or even the early 2000s" (which of course I did; I've been on Slashdot since about 1998). If you're going to say stuff like that, I'm not interested in convincing you of what I'm saying.
--
make install -not war
And while in her writing she does somewhat tone it down, in her diary she was going all fangirl over people like William Edward Hickman. That was her ideal of superman and she loved a quote from him saying "what is good for me is right."
Instead of arguing the idea, find something about the person in their freakin diary and then argue that. This is what passes for debate these days.
If everyone looked out for the interests of themselves and their family properly and as their first priority, then society would be left with the truly needy that need charity. Charity would be more than able to take care of their needs. Her point is that need can be manufactured, learned, and lied about, thus the basis for forcing one person to support another because of this need is unjust.
My mother can give me all her money, and now she can pass the "poor and needy test". She is now declared needy and society must care for her.
My neighbor can choose to watch American Idol and every other piece of garbage on TV all night while I study, read and improve myself. When they get fired for incompetence they are now needy. That need was a result of choosing not to improve. He is now declared needy and society must care for him.
My other neighbor can be on unemployment and paid for over a year and never look for a job, while I work 16 hours days in fear of being fired. He has been declared needy and society must care for him.
Promising the needy more money from the non-needy is what allows politicians to gain power and influence and votes at the expense of productive members of society. It further encourages more people to become needy and less people to become productive. That is, until there are no longer enough productive people left and society collapses.
This is the essence of Ayn Rand. Self sufficiency and an undying desire to do your best at whatever you are gifted enough to do.
Scott,
There's a reason why Oracle is #2 today. Larry knows how to take care of his shareholders. Duh.
You, as it appears, learned that lesson after the fact.
Hindsight is 20/20, you didn't know back then and likely don't get it still. Sun lost from it. Guess what, Oracle's been in the OSS game for awhile (since 2003) and competing with IBM, HP and CIsco (other F/OSS contributors) they get it, heck IBM gets it, which is scary--its the F/OSS community that needs to deal with it now, which will be interesting.... much like a chess game.
Thanks for your wisdom too little too late.
Former Sun employee.
Yes, McNealy's argument is precisely that Sun's management of Java was wrong (i.e. failing to capitalize on it) and the situation with IBM is Exhibit A: Sun made all the investment, and IBM reaped all the profits.
Are you arguing that Sun should have just done what IBM did, i.e. wait for somebody else to come along and give away a technology they could use to make money with services? (That's largely true of IBM in the case of linux too, although IBM has invested a little bit there).
Look, we all hate Oracle and Microsoft, but let's face it, they're charging money for proprietary software and making a good living at it. (That doesn't mean Sun could have survived by demanding lots of money for Java, but then again, what Sun did do didn't work either).
Mod parent up. Those of us who actually have worked at the top software megacorporations know exactly how hiliariously stupid and ignorant /.s readership is.
It seems his point was that the company should have open sourced things at a correct time and did not. They should have open sourced some products more aggressively early on and NOT open sourced many products they did at the end.
Open source wasn't the problem, as he freely admits. Doing it too late was the problem. By the time it was "near the end" it was too late to "take care of the shareholders" by doing anything different. Open source was the only thing keeping Sun relevant near the end, and therefore the only thing taking care of the shareholders.
What does relevance have to do with anything!? Income is what is relevant to a shareholder. NONE of the projects that Sun open sourced in the last few years increased income for the company. If anything, because of the way Sun did things, they decreased the income.
He should have opened the Solaris source, ported it to Java running on every CPU...
!?! Do you know what Solaris and Java are? I literally, have no idea what this is supposed to mean.
It's the "NOT open sourced many products they did at the end" that is not supported by either his argument, or your lack of one, or the independent facts.
The point is that McNealy's way of running Sun was wrong, not simply his open source strategies that were too little, too late. Plenty of other companies have shown that running open source right is quite the income generator, both directly and indirectly.
What relevance has to do with is marketing. Without relevance, computer tech companies can't attract either developers or customers - or keep them. Without growing customers, shareholders don't get profits - which is what shareholders care about. Indeed, tech corp shareholders care about ever increasing stock prices more than they care about income (on which stock prices are only loosely based). Stock prices are more directly based on relevance than on income, as stock analysts are their real market.
Of course I know what Solaris and Java are. If you don't know how to open the source of an OS, then port it to a new language, that can be compiled to a portable binary running on an embedded VM, then you evidently don't know any more about computers than you do about business.
--
make install -not war
"Open Source business model works well when your company model is a consulting based business."
Yes. But let's make "consulting" a wider word and it will become even truer. If your business is about contracting (I'll pay you to *do* this) instead of licensing (I'll pay you to be *allowed* to do this) then it makes quite a lot of sense.
It's not only consulting but developing on-demand too (and everything where "doing" is involved instead of "licensing").
"Suns model was product based business."
The problem is not what Sun's model *was* but that it wasn't able to adapt.
"IBM doesn't do many products any more"
So they *had* a business model and they *adapted* it to a changing environment.
One thing the article discusses quite clearly was McNally's desire to move the low end to Intel in the late 1990s. It appears he regrets not becoming an Intel vendor much earlier. Its interesting I wonder how they would have handled having a Sun branded line which was much cheaper when it really would have been undercutting sales of their Sparc workstations and low end servers.
Doc --
You are missing the point. Solaris was better than the Linuxes of the 1990s. Open Source was behind Solaris for those years. "Solaris was Linux" was a comment regarding Linux's development path, it was quite deliberately targeting Solaris features.
Now can you answer the question. When were people putting Linux on Sun hardware in large numbers?
I'm not saying that it was difficult. I do believe that Sun tried. They kept reinventing the graphics classes. You have AWT, SWT, and Swing. The hardware UI differences are a red herring. There are only a couple of different classes of display, and other systems have been able to handle that without trouble.
It was, and is clear that the group in charge of Java was in over their head when they created it. There were some things they got right, but they got a lot more wrong. They were simply far behind the curve in the emulator field.
What it is used for, and what it is good for today was not their target when it started. Java was a failed solution for a problem, that eventual found a different problem it could solve.
No. Worker's refused to work for free. Stupid workers.
Sleep your way to a whiter smile...date a dentist!
Shut the fuck up Scott McNealy.
Yeah, that's right. I said it.
Where the fuck were you when in 1999 when SUN was selling SPARC 20's for over 20K?
At the same time commodity hardware, running Linux, was going for 1-2k, tops.
Where the fuck were you when SCO sued Linux and you made sure to send SCO their blood money.
What were you trying to say, you stupid fuck. Look at you now.
Nickel and dime the customer and charge outrageous fees.
SUN was over in 2001. Java kept it going for while more.
In 5 years there will be no more SUN and Linux will completely dominate.
And you sat there with no vision and watched it all happen.
You could have had it all. But you chose to be an arrogant asshole and
now you are left holding your dick.
What if you made Solaris Open Source? What if you lowered the prices
of hardware? What if you went to INTEL. Yeah, you did but too little
too late. You could have owned all UNIX.
You stupid motherfucker. You ruined SUN. You ruined Solaris.
And now you are like: Booohooo OSS ruined my life, waaaaaaaa.
Shut the fuck up and take some responsibility. You fucking asshole.
You guys think I am too harsh. Maybe use "softer" words?
Well guess what? I am a Solaris UNIX SA. I make my living
from Solaris and this asshole ruined it.
I love Solaris. It is a beautiful thing. It should have been
the dominant OS today.
This guy is to blame.
Fuck him and his fucking horse.
Except for the part where it's actually relevant to what she wrote. When we have a chick who lionized sociopaths in her novels, made rape a sort of declaration of love in The Fountainhead ("had she meant less to him, he would not have
taken her as he did; had he meant less to her, she would not have fought so desperately") in spite of her own describing the victim as in terror and pain between the reassurances that that's really what she wanted, made a whole pseudo-intellectual philosophy in defense of psychopathy, gave interviews in which she railed against any kind of altruism or conformity to social norms, and modelled at least one character after Hickman... maybe her rabid fascination with psychopaths is actually relevant. I mean, we're not talking the relevance of her favourite dish or brand of car to Objectivism, but that the chick preaching why you shouldn't give a damn about other's needs... actually went as far as their need to live in her views. Maybe, just maybe, it's all part of the same picture.
Except that's what got me to dub Objectivism a cult of sociopathy before even knowing about Rand's fascination with murderers. It's that inability to conceive or discuss anyone else's problem than as some elaborate scam to rob them of their preciouss money. That old lady with a crap pension must be somehow hiding her money somewhere and only faking it. That guy fired in a downturn must have been just a leech anyway. That guy unemployed for a year -- never mind that a certain unemployment level is actually wanted and part of / tied to regulating inflation, via the Philips curve -- must be not even looking for a job. Never mind that he wouldn't be even counting as unemployed if he didn't.
Out of three people, you counted three who are just some kind of leeches and parasites, and at least two being scammers too. That's the kind of thing I'm talking about. That someone could actually have a problem, or really any kind of empathy, doesn't even get a passing nod as a possibility.
That or some kind of extreme scenario, where the Objectivist jumps to some fictive totalitarian state where giving to all beggars is mandatory, and the Altruism Gestapo can kick their door in for not buying milk for the neighbour's kid. If a talk to an Objectivist doesn't degenerate in the above list of parasites conspiring to scam him of his money via the state, it degenerates in such a black and white situation where there are no shades in between not giving a fuck and being legally forced to give more fuck than they possibly can.
The more normal modi operandi of an actual human society or community don't seem to even register as possibly existing anywhere.
A polar bear is a cartesian bear after a coordinate transform.
The reason Sun died mainly is that 100 dead-cheap PC blades with linux and "cloudy" deployment
systems and "cloudy" work-sharing and fault-tolerance middleware
makes a more reliable server infrastructure than 5 or 10 Suns.
Where are we going and why are we in a handbasket?
just competing with it
They did in the end do JavaFX, but I think by then it was too late.
Man, none of you guys have a clue. Have you read Rand or are you just regurgitating what you read on Wikipedia?
I've read The Fountainhead, Anthem, and a number of Rand's essays, as well as Wikipedia and other biographic articles. I also, of course, know who John Galt is. I think some of her work has some depth it's not given credit for (unfortunately, as much by her apparent followers as her critics).
But I don't think her critics here have no clue. They might be ignoring how noble the protagonists in her fiction are, but they're not incorrect that her philosophy as policy would lead to as many (if not more) James Taggarts and Ellsworth Tooheys as Reardens and Roarks (not to mention the problems with Roark, however romantic a character he is -- sure, he had a contract with Keating that enabled him to blow up the building, but did Keating have a contract with the land owners/developers/ect that gave him property rights that he could transfer to Roark?). And as has been pointed out, when it's come to practical recognition of real-world individuals, Rand has endorsed some individuals and behavior that resembles psychopathy.
On the other hand, the Karl Marx philosophy is about theft. Those who need take precedent over those who produce.
So, speaking of "Have you even read _____" criticisms.... how much Marx have you read? Because while needs are part of the philosophy (and the fact that this is a target of criticism says something about the critics, I'd say), there's a hell of a lot more to Marx than that -- it is, in fact highly concerned with "a fair reward for [the] inspiration and sweat" of laborers and craftsmen. I'd recommend, for starters, this slashdot comment about the implications of a competitive market for labor-as-commodity.
Tweet, tweet.
IBM are very cunning in supporting and developing open source systems with an eye to making sure that no one else is able to make any cash in a particular market. Where the barrier to entry is high IBM will be there investing and charging many dollars. Where it is low you can be sure they will push open source so that there is no market for anyone and minimal investment required from them.
--------------------------------------------- "In the end, we're all just water and old stars."
The problem isn't open source. It's that there's very little room for large companies in the ever shifting world of computer technology. If you can't shift strategies on a dime, you're gone.
For example, there was Java ME, which Sun tried to monetize but failed, partly due to Android, which is why Oracle is suing Google. When discussing this lawsuit, I encountered many commentators on various sites forgetting the difference between desktop and mobile Java, even though it was very important.
IBM's success with Java pretty much proves that it was Sun's management of java rather than Java itself that was the problem. On the same note, IBM's success with Linux pretty much proves that McNealey's whole rant makes little sense.
No. IBM proves when you make > $25 Billion in Mainframe clients you can afford to invest in FOSS within other areas of your businesses. In IBM's case, they have dozens of businesses.
You came, you conquered, landscapes changed, you got conquered and then you and your company tried anything to stay afloat. Personal opinion, but if Sun open sourced Solaris and the Java family when they could compete with the current forerunners, I doubt oracle would p-0wn you guys.
Sun was successful because in the 80's they beat the mainframe. Instead of some monstrous IBM or DEC thing, you could get Unix workstations for your CS department.
Then the 90's came, and Sun became the mainframe which was eaten by cheap commodity hardware and better operating systems.
Mistake #1 was taking System VR4 and turning out the pile of crap known as Slowlaris.
Down hill from there.
You mean still alive and still a big player in the IT industry?
Granted, they don't have the position they once had, but neither does IBM, or HP, or Microsoft (or Apple, going the opposite direction).
Things change, adapt or die. Novell adapted, Sun died.
Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
Wait. A company who was just sold for less then 2 billion is worth more then a company who was just sold for less then.. 7?
How, exactly, is 2 alive, but 7 is dead?
-- I'm the root of all that's evil, but you can call me cookie..
He's going to do the exact opposite... Seriously, you think Larry Ellison is going to take advice on how to run his business from that loser?
Of course I know what Solaris and Java are. If you don't know how to open the source of an OS, then port it to a new language, that can be compiled to a portable binary running on an embedded VM, then you evidently don't know any more about computers than you do about business.
I've written professionally in Java (amoung many other languages) for more than a decade. If you think that A) re-writting a OS written in C in Java or B) ANY OS should ever be written in Java; then you clearly have no idea about either.
For starters, the strength of Java is that you don't need to compile it to native machine code. This is almost 100% true. Conversely for an OS to function it MUST be compiled and assembled into machine code. If you are going to do that why use Java? There are many other languages better suited. C for example, which is what the vast majority of Solaris is written in.
Next, do you have any idea how much time, how many developers it takes to write an OS. Even re-writing verbatim? That does not even count testers. We are talking about more than the total yearly cost of running Sun. It is in the hundreds of millions if not billions or tens of billions.
All for what? A product that runs on a few more machines? Not to mention Solaris runs on 90+% of all machines out there now anyways. Your suggesting they add a insanely huge workload for almost zero return!?
Lastly on that subject, if you wanted Solaris running on an embedded device, why not just INSTALL IT. That is it, you don't need any VM, you don't need any third part intermediary. Just have the device run Solaris natively. It is how it is done now. You can order such things from Sun/Oricle now and have been able to for some time.
As for your suggestion that open sourcing was to late/to little. I disagree. It was done wrong. There were no income streams tied to open sourcing most of the projects. They simply said, "ok, this is open source now". What they should have done (as suggested in other posts) is offer to help customize their systems to specific clients. Or offer consultation work. Anything really. But just shouting "hey, its free now" is never going to do anything to bring in income for a company.
I've worked with IBM providing services at a previous employer.
Their "solutions" look great to managers wanting a plug-in fix. When you're trying to get the job done, they don't know what the hell they're doing.
IBM = Indian Bureaucratic Mess. All you have are poor Indian techs on the phone desperately trying to grasp what it is you're trying to do.
And then they charge $15k / hour for that.
The issue is that for all those big "IT Services" companies, there are too many levels of managers, accountants, lawyers etc than there are actual techies.
If your main product is IT consulting/servicing, and half of your employee base is NOT fully qualified or trained to do that, you have issues. As far as I remember, Sun really didn't have much going for it anymore. And Oracle's approach has only made things worse.
Of course Sun had a problem marketing - by definition. Evidently you don't really know what marketing is. "Knows who they are and what products they have" is not marketing. It's not "= advertising". It's connecting the market to the company for maximum access. Sun's products were mostly locked up within a server vendor. Outside of the geeks buried in IT, nobody even knew what StarOffice was, or even really what Java is - the two products Sun needed to build its marketing on to compete instead of dying.
If it were Sun, with its greater investment and expertise in Java, revving Solaris would have a lot more Java and a lot less C and C++. While it is not necessary to compile Java to native code, it is still necessary to compile it to a virtual native code, bytecode, for a JVM. Or entirely to native code instead of to bytecode. Java is a language, with a VM that runs its own code. Sun could have delivered a Sparc CPU with a "JM" implementing bytecode in HW, or just optimized the compiler for Sparc, or optimized it for x86 or whatever CPU it wanted Sun machines to run the full stack best on.
An OS needs to be compiled to machine code to perform well (much of the OS could be bytecode interpreted by a JVM in machine code embedded in a machine code OS), but Java can be compiled to machine code - or run in a CPU that executes bytecode - as I already pointed out. I've been writing SW professionally for over 30 years, and professionally in Java (originally on Sparcs) for over a decade myself, since the first beta JDK. It doesn't cost $billions for an effort like that, rather dozens or a few hundred million. But Sun had the money, and could have stolen Linux's growth, accelerated by Sun's other assets. Indeed Google would likely be putting a "picoSolaris" onto all these mobile and embedded devices which will be the large majority of OS installs in the world within 5 years. Instead Sun spent its money McNealy's way, and the rest is history.
If you want Solaris running on an embedded device, you have to compile it to run on that device's CPU. Sun barely ported Solaris to x86 while it was worth doing (before Linux fully owned that segment). To port not just to the dozens of CPUs already necessary this decade, but to the new ones popping up every year or month, Sun would need the OS written in a language suited to such varied deployment, which is neither C nor C++ - more like Java. Going halfway with a JVM until a CPU matured as a (profitable) platform, then optimizing the OS better than JIT and other autogeneration does. Or you could just run Solaris on a tiny few platforms, betting the farm on them, as Linux and now Android run circles around you.
Open sourcing was too little, too late, but just opening more earlier is of course not the answer. The answer was to open more, earlier, and more strategically. Which doesn't mean everything opened has to have a revenue stream as if it were proprietary, which is both practically impossible and proven unnecessary by Red Hat, IBM and others. And even someone like me, who's not applying for the Sun CEO job, knows it takes some big leaps, some of which I've pointed out here - and some of which opportunities and risks a Sun CEO would know much better than I do. And indeed some of my suggestions might not have worked in their business - though technically and at first business glance it's better than what Sun did try. It was well beyond McNealy to see that, though.
--
make install -not war
You missed that Novell just got sold and gutted for a small fraction of what Sun was picked up for?
No, Sun failed to see the writing on the wall, in that they continued to develop and attempted to sell expensive "high performance" SPARC processors and SPARC-based servers.
Unfortunately, they had their own version of the PowerPC problem, in that like Apple, Sun just couldn't couldn't keep up with Intel's R&D, and as such SPARC's performance fell further and further behind.
There was also the minor problem of commodity OS's and servers that grew to eat their lunch. (And breakfast and dinner.)
The only reason they embraced open-source lay in an attempt to keep their hardware relevant. (Java works best on Sun SPARC!)
Didn't work.
Any sect, cult, or religion will legislate its creed into law if it acquires the political power to do so.
he
Well Ellison gets one thing wrong. It wasn't 12 inches of cooling fans, it was 12 inches of solid copper heat sink topped by a fan. Sun took a failed Darpa project and tried to make it into a commercial venture. I was ready to dump my life savings into Lambert stock.
Things change, adapt or die. Novell adapted, Sun died.
Any examples of companies changing without adapting?
I've long said that the difference between IBM and Oracle is one of approach.
If you invite Oracle into your shop, Oracle will come in, look around, tell you all your current hardware and software is crap, and if you rip it all out and standardize on solutions from Oracle your business will run much better. Oracle then makes a killing selling you software licenses and support contracts and you have many of the same headaches you had before, only now the only company you can call is Oracle.
If you invite IBM into your shop, IBM will come in, look around, tell you what a great job you've been doing and how, with a little tweak here and there, it can get all your existing infrastructure running along like never before, with everything talking to everything else and all your problems solved. IBM then makes a killing billing you countless hours of consulting and selling you middleware to make that dream scenario happen, though it never really will.
Sun did neither of these things. Sun sold stuff. Sun did not make a killing.
Breakfast served all day!
That's the GP's point. Say instead that IBM didn't have to create Java or Linux (and incur the cost of doing so) -- instead, it could just step in and immediately profit from both, while Sun, which didn't seem to understand how IBM was pulling it off, managed to profit from neither.
Breakfast served all day!
Garbage collection interacts very badly with swap. Once your Java program starts hitting the disk, it will stand still for minutes. Bigger memory sizes are solving this problem nowadays.
swap allows on OS to move UNUSED applications / etc off to disk, so they can be restored later to main memory when they need to run. if your system is trying to run an application that's located in swap, that's called thrashing and any application is going to perform like crap.
You are wrong in two subtle but crucial ways here:
I once read an interesting paper on an experiment where they modified a Java VM's garbage collector and a Linux kernel to be more aware of each other and improve performance: Garbage Collection Without Paging. Worth looking at. (There's also an amusing story of how the authors of the papers submitted a patch to get their GC-oriented advisory kernel calls into the Linux kernel, and got rejected by folks who didn't give a damn.)
From 10,000 feet, I'd say that one of the issues is that operating system kernels have been designed for C-style languages with manual memory management, and don't provide any services designed for garbage-collected languages.
Are you adequate?
Sun is history because they failed to adapt to the fast commodidation of the computing marketplace. They failed to embrace Linux when everyone else was jumping onto it around year 2000. They failed to embrace Intel/AMD platform at the time when IMB, HP, and Compaq were more than glad to sell you either RISC or CISC system, as long as you bought their stuff. They canceled Solaris on x86 at the time when it was probably last moment to prop up Solaris on x86 and thus make the platform more mainstream. Instead, they bought storagetek (does buying a tape backup company seem like a way to become a leader in unix market place? no!) Now they reaping the results. They weren't worth more. They spent a decade with not much results. Even on the high end, they couldn't beat IBM or any serious storage vendors. It's entirely deserved that Oracle picked them up for a song. Sun couldn't have been worth more.
I was just reviewing this stuff a bit, and came across this paper which is just as interesting and relevant (PDF file): CRAMM: Virtual Memory Support for Garbage-Collected Applications (Ting Yang, Emery D. Berger, Scott F. Kaplan, J. Eliot B. Moss).
Are you adequate?
Remember that McNealey got fired because it did not sell enough high end expensive systems that Wall Street demanded. The shareholders are idiots. SGI saw this and tried to enter the wintel market as they saw the competition. It killed them.
What could he have done? Wall Street demands proprietary expensive slow hardware for optimal premiums and consumers will not bother with an arcane cheap unix since Joe Six Pack can't run his MS Word or Counterstrike on it.
http://saveie6.com/
To put it another way, IBM were clever and came up with a business model that works, Sun were stupid and thrashed around trying things and the ex-CEO is now using open source as a scapegoat.
Red Hat have a pretty successful open source model as well.
That my friends, is what is wrong with capitalism. When you are punished for giving back to the community, and praised for being selfish you know something is seriously fucked up about how this country operates.
Not sure where you've spent last five years, but IBM is pushing its proprietary systems again, and with pretty good results. Did you notice that for big machines, their always publish server application benchmarks under AIX, not Linux? High-end Linux benchmark results from IBM are limited to things like HPC.
The mistake they made was that they forgot (or didn't know how) to monetize the open source solutions they had. .
Ok, Pet Peeve time... "Monetize" does NOT mean 'make money out of something'
It is the act of creating currency - literally minting coins or printing notes or banknotes. It can also mean establishing a national currency.
There are hundreds of way to say "profit from", "make money from", "exploit financially", but pretty much one way to talk about monetizing currency.
So QUIT misusing "MONETIZE"!!!
And get off my lawn, dammit!
http://en.wiktionary.org/wiki/monetization
Same here... still waiting for my Solaris...
Hope Oracle is nice enough to send me the DVD....
Dear Scott
The poster is right you and Jonathan Schwartz were the ones that killed it. I was a Solaris Engineer for over 15 years. Our shops back end ran on Solaris (around 300 machines) but we are now in the process of phasing out Solaris for Linux thanks to you. All the training I have with your system is now worth nothing.
I read the article and you missed what killed your company. You didn't do any marketing. Now I am one that hates the marketing department but still it does have it place especially when playing with the big boys that also use it.
When we would try to sell your products our customers would ask "Sun who?". Look at Cisco everyone knows the name. The average Joe's knows the name Cisco. They have no idea what Cisco does or sells but Joe knows the name from the cute little media ads "The Human Network".
I asked many times to your sales and engineers people "Why not get your name out?" and was told "Management doesn't think it is a good idea." Well no one knew your name so why buy Sun? Sun who????
Damn guys why didn't you run ads showing where you all shined. Little robots on Mars designed to run 90 days and RAN FOR YEARS! The fact that your old tag line was still true you guys where "The Dot in Dot Com" and still are to a degree. If you get on the Internet you touch a Sun box somewhere Everyone uses you but the average person don't know who you are even if they were using your products. I have talked to people that have bought Cisco stock and have NO IDEA what Cisco makes or sells. They have just heard the name over and over again and heard from there broker they are a good investment. How can anyone buy from you if they don't know your name asshole. Hell it was easier selling Linux more people have heard of it more than Sun and you've been around longer and Linux has no marketing budget and Sun had one. Seems the problem lies with you asshole. You should have done your job and got your good name to the public.
The truth is you could have made money with Sun if you tried. I fully believe you've been setting up this sell for awhile to line your own pockets with gold and to hell with your customers (oh wrong word "Consumers") and to hell with your employees.
You said in your article you did it for the stockholders and the employees. Well most of the good employees didn't get to keep their jobs. They quit. They couldn't work for an asshat company like Orcale. So did that work for you??? No. so shut up about jobs. Yes your "Customers" the stockholders did ok. You and Jonathan will never hurt for anything sitting on your pile of cash.
You wrote 'You gotta take care of your shareholders or you end up very vulnerable like we got." This is where you went wrong. You should have taken care of your NAME and branding and the people that used your products and sold them to the public. People like me. No matter how much money shareholders put into a company it isn't going to be for shit if your NOT SELLING PRODUCTS! People can't sell your name if no one knows it you fuck up.
So quit your crying and quit doing interviews its all you fault so go cry to your money and just shut up and go away. Your done here. No one likes you no one wants to hear from you. just go and spend you money on someone like yourself a two dollar whore and have a good time on us.
Oh yes please tell your buddy Jonathan that if I see him I will cut that pony tail off of his head. He doesn't deserve to wear it. He's no pony tailed geek. He's just another corporate asshole.
b0
Damn I do feel better now!
Novell is getting sliced, diced, sold to Microsoft, and sued by shareholders. They did same as Sun, gave lip service to open source while doing exactly what community hates. You need people to like you if you want to succed in open source. Both Sun and Novell failed at that. Sun created Corporate Decision about Destroying Linux (CDDL) and put stuff behind walled garded saying "here, it is free, take it... except you must ditch GPL and let us take everything we want". Novell was even more malicious and more stupid. They decided to be Microsoft drone and make fauxpensource clones of MS crapware so Microsoft can claim openess and latter sue Linux users. Sun and Novell both deserved to go extinct.
I know that Alan Greenspan personally knew Ayn Rand. But the article was not about Alan Greenspan, it was about Scott McNealy, who almost certainly did not know her as he was growing up.
This reminds me of a blog post by Hani Suleiman from 2008. Today, that post looks like a prophecy, first Tim Bray actually went to Google and now McNeally admits that "you gotta take care about your shareholders". A real pity that Hani stopped posting.