Nikolai Bezroukov. Portraits of Open Source Pioneers

For readers with high sensitivity to grammar errors access to this page is not recommended :-)

Sun under the Linux siege

• Zugzwang moves: some brilliant, some not

• Solaris for Intel faces battle on two fronts: killing me softly with Linux and Microsoft

Zugzwang is a German word which is used as a special term in chess.  It occurs when you find yourself wishing you could pass your turn because every move you might make in the current position will actually make it worse! While it seems that such a position might be very rare to occur, very subtle zugzwangs are common in high-level chess games. Well, that's it, that's all I wanted to go over. What was the point? Just to make you aware that zugzwang is in fact a position that can happen with players outside chess. 

In the middle of 2000, despite all Java hoopla, Sun's top brass started to understand that combination of Linux, high end Intel CPUs and IBM created for Sun a zugzwang at the marketplace and made several desperate attempts to counter advances of Intel hardware with Linux as a Trojan horse into UltraSparc turf.

But the problem was not so much with Solaris vs Linux but with the fact that that UltraSparc CPUs were unable to compete with Intel-based CPUs in low end to midrange workstations and servers neither in price nor performance. Sun still kept leadership for high end SMP system with 8 or more CPUs and more then 4G of RAM). 

After AMD and then Intel crossed 1GHz mark Sun CPUs became essentially non-competitive for entry level and midrange servers. Quality of Solaris OS still kept many customers loyal to Sun but even here Sun became distracted by Java. The only edge they had was that UltraSparc chips can work in huge SMPs configurations (up to 64), but in raw performance a single UltraSparc CPU became no better then a half of Intel's CPU on major benchmarks.

Technological trends were also working against Sun: the cost of developing and manufacturing new processors became two much for medium-weight payers and only behemoths like Intel and IBM have enough resources to continue their processor lines. HP bail out of their RISK line approximately at this time, but Sun decided to fight on.  With TI as an outsourcer Sun has to settle with higher costs of their CPUs.

Moreover the speculation that Linux is directed against Microsoft proved to be largely an illusion. In IBM hands Linux proved to be a deadly weapon against old good proprietary Unixes and first of all SCO and Solaris. In a way it helped Microsoft's fight against Sun.  For industry observers outside Linux zealot's camp Sun has always seemed a more plausible, if not imminent, victim of  Linux techno-cult. And as for IBM they have a history of going after Sun:

Re:Seems Unlikely (Score:5, Informative) by khb (266593) on Tuesday May 11, @11:45AM (#9117138)

Humm, a very different perspective can be found at http://www.forbes.com/asap/2002/0624/044.html Where Gary Reback, famed IP attorney, says: My own introduction to the realities of the patent system came in the 1980s, when my client, Sun Microsystems--then a small company--was accused by IBM of patent infringement. Threatening a massive lawsuit, IBM demanded a meeting to present its claims. Fourteen IBM lawyers and their assistants, all clad in the requisite dark blue suits, crowded into the largest conference room Sun had. The chief blue suit orchestrated the presentation of the seven patents IBM claimed were infringed, the most prominent of which was IBM's notorious "fat lines" patent: To turn a thin line on a computer screen into a broad line, you go up and down an equal distance from the ends of the thin line and then connect the four points. You probably learned this technique for turning a line into a rectangle in seventh-grade geometry, and, doubtless, you believe it was devised by Euclid or some such 3,000-year-old thinker. Not according to the examiners of the USPTO, who awarded IBM a patent on the process. After IBM's presentation, our turn came. As the Big Blue crew looked on (without a flicker of emotion), my colleagues--all of whom had both engineering and law degrees--took to the whiteboard with markers, methodically illustrating, dissecting, and demolishing IBM's claims. We used phrases like: "You must be kidding," and "You ought to be ashamed." But the IBM team showed no emotion, save outright indifference. Confidently, we proclaimed our conclusion: Only one of the seven IBM patents would be deemed valid by a court, and no rational court would find that Sun's technology infringed even that one. An awkward silence ensued. The blue suits did not even confer among themselves. They just sat there, stonelike. Finally, the chief suit responded. "OK," he said, "maybe you don't infringe these seven patents. But we have 10,000 U.S. patents. Do you really want us to go back to Armonk [IBM headquarters in New York] and find seven patents you do infringe? Or do you want to make this easy and just pay us $20 million?" ... In corporate America, this type of shakedown is repeated weekly.

Unlike IBM and HP, Sun was and (as I am writing this in 2000) still is the leading holder of market share in proprietary Unix space and thus it more affected by the  "Linux cuckoo effect": commodization of the platforms on the base of Intel X86 architecture and high speed CPUs that Intel and AMD produced so cheaply and in such huge volumes. The process of commodization OS with the conversion of Unix kernel into a kind of super BIOS also worked against Sun. Of big troika (IBM, HP and Sun) Sun was the only one that did not sell Intel boxes. So it was the most vulnerable to the hardware and OS commodization movement represented by Linux.

As Sun leadership saw from early 1999,  Linux successfully cannibalized a low end enterprise workstation market despite low (in comparison with Solaris kernel) quality and dismal networking. Sun made a brilliant counter move by introducing SunPCi card. For some reason this brilliant product that created "dual-personality" workstation was never properly marketed.  Partially here the problem was with Sun brass attitude toward Microsoft: they fail to realize that objectively Microsoft is the strongest ally of Sun against Linux techno cult and improving interoperability can play into Sun hands as IBM cannot replicate those moves for political reasons.  Also Microsoft itself needed a Unix emulation layer and here was another lost possibility of interoperability of Windows and Solaris.

With the advances of Intel hardware, around 1999 (with Intel CPU reaching 1GHz ) Linux became more plausible for midrange servers and during the downturn that started in late 2000 many firms tried to cut costs to the bones and that means Intel hardware from such vendors as Compaq and Dell. Neither Compaq not Dell install Solaris for Intel on their boxes.

This lack of any possibilities of a coherent counterplay typical for zugzwang immediately had negative impact of Sun's presence in the Web server market: one of the most important market for Sun was essentially lost without major fight. After 2000 very few ISPs were using Solaris, most ISPs switched to Intel platform using either Linux or FreeBSD.

Large ecommerce shops and online brokers also (partially) switched their Web server farms to Linux.. Just think about eBay, Amazon, eTrade and other e-commerce players that abandoned Sun on web-farms in favor of Intel boxes and you'll get the picture.  In most cases Sun survived on backend with Oracle databases, etc, but many IT executives start thinking that it's only a matter of time before Linux kernel will be mature enough to run Oracle databases. 

SCO destiny showed quit clear that the value of strong media support of "David vs. Goliath" story of Linux should not be underestimated. With its strong, religious style PR Linux became a real threat on the low end for any proprietary Unix vendor and at the end of 2000 with the burst of  Internet bubble Sun leadership saw the writing on the wall. 

And their options were very limited. Of course Sun was to a certain extent protected by the uniqueness (and in some areas superiority) of  its 64 bit UltraSparc architecture (at this time, and generally before introduction by AMD 64-bit extensions, Intel servers did not scale well above 4G of memory and four CPUs) as well as the unique ability to tune this architecture to Solaris needs and vise versa. Despite lower CPU speeds, UltraSparc was definitely a better server architecture than Intel 32 bit architecture, but the problem was that Intel architecture was "good enough",  much cheaper and Intel CPUs reached significantly higher clock rates. UltraSparc CPUs were also less power hungry and that represented a distinct advantage as the number of servers in a rack rose dramatically with 1U and 2U servers becoming much more prominent in corporate datacenters.

Moreover Sun neglected Solaris during it's Java-hype years and lost chances (if such existed) to make it a standard e-commerce platform on Intel 86 (slight chances in this area existed due to Linux records of security vulnerabilities and lower technical quality of the kernel). That means that Sun became extremely vulnerable to any economic downturn:  Sun's stock was over inflated during the Internet bubble and the company suffered a crushing downslide (from $86 to $6) in just two years. That compares well only to Red Hat record ;-). Revenues also have taken a nosedive, down 32% from 18.3 billions in 1999 to $12.5 billions in 2001. On lower end both Windows 2K and Linux can use IBM, Dell, HP/Compaq hardware for a fraction (50% or less) of Sun's price and with 2G CPU and 1Ghz Rambus memory becoming commodity,  the lower end was not that low anymore.  Thus for Sun Red Hat represented a bigger enemy then Microsoft. Especially dangerous was Red Hat alliance with IBM ("enemies of my enemy is my friends" and vice versa),  a direct competitor that was eating Sun's hardware market share (AIX gained nicely after 2000, partially at Solaris expense; with AIX 5.2 IBM became pressing Sun for all major Oracle deals). And there is no secret that Sun major competitors like IBM and HP adopted Red Hat partially as a powerful weapon against Sun. That means that Sun has found itself in a very difficult position when it cannot adopt Linux, but at the same time Sun cannot ignore it and that's why I called it zugzwang.

Later Sun's brass realized  that that on low end the game is lost and that Sun will continue to lose market share to Intel, so they need to forge an alliance with AMD to have some breezing space in low end segment. But in 2000-2001 the only way to compensate this loss was by gaining additional market share on midrange and high-end market segments as well as in software by aligning yourself with Microsoft and stressing interoperability of two platforms and thus reusing Microsoft huge marketing muscle.

Instead Sun initially tried to bet on Java. The idea was sound: while Linux may be making a dent at the lower end of corporate networks only, Sun theoretically could continue to dominate the midrange and high-end where Java middleware will connect enterprise applications with Oracle and other databases. But IBM quickly moved into this space as it suffered from internal diversity of platforms much more then Sun and thus can benefit from Java more then Sun. Thier success with Websphere product line had shown dog like determination with which they jump into Java bandwagon.  It is due to IBM support Java quickly became a new Cobol for commercial applications.

Instead of using Microsoft as a leverage against IBM, Sun created infamous lawsuit over JVM infringements which killed J++, the most successful competitor of IBM developer tools.

Also commodization of hardware worked against Sun and due to Intel (and AMD) successes Sun soon found that its major ally Oracle is flirting with Linux in order to stop erosion of its market share by Microsoft.

At the beginning of 2001 some analysts started to suggest that Sun days are numbered and the shadow of SCO can one day enter the luxury offices of Sun's executives: Sun's brass underestimated the Linux threat and was too slow to react to RH/Intel/IBM  challenge (see below).  From Sun's brass point of view,  a lot of Linux developers are amateurs and don't understand the fundamentals or dynamics of OS market change. While that was partially true (and Linux Torvalds is far from technical genius in Unix kernel space, he is just a very successful in reimplementing somebody else ideas) that was a huge mistake. Situation slightly improved only when Linux run into SCO lawsuit rock.

But generally Sun management was not up to the situation. For all you can say about Microsoft, it remains surprisingly vigilant to new technology possibilities and threats including Linux and played Linux card nicely in its lawsuit. Later with the introduction of .Net they made Linux look like the previous century technical solution (which it actually was).

At the same time with Intel hardware moving into 3+ GHZ range IBM's bet on Linux became really dangerous for Sun. The trump card of Linux is cheap high-speed Intel CPUs and chipsets and here unfortunately Sun can do nothing...  Unfortunately only in 2003 Sun decided to use AMD Opteron as a second CPU to fight  those chances and hedge its hardware bets.

 Fighting  "Linux cuckoo effect"

The first strategy the Sun tried was the strategy of "peaceful coexistence." One of the first descriptions of  the strategy of "peaceful coexistence" with Linux was probably in 1999 Bill Joy interview (see Linux Magazine November 1999 The Joy of Unix):

LM: Have you ever considered making the Solaris source code more freely accessible?

BJ: Yeah. The difficulty is that it's got a lot of third-party stuff that's licensed under funny terms. So I think it will be really healthy for both the Solaris and Linux communities to work more closely together.

LM: Think that will ever happen?

BJ: It already is. We run a lot of Linux binaries, and we're trying to find ways to work together. Merging isn't a goal. I think Linux and Solaris have different goals. Linux is not worried about providing MVS class or VM370 or whatever IBM-class services for corporate data centers. That's not the center of the Linux community.

LM: But there are certainly areas of overlap.

BJ: That's okay. It gives people a choice, and that's not a bad thing, right? I still prefer to win. I'm not saying we're not competitive, but I'd still rather have it be Linux than NT. If there's two Unix choices and one Microsoft, that improves our chances.

Here's an similar later quote from Sun CEO Scott McNealy, "We do not intend to focus our core business on Linux. Our application development has one focus and that is Solaris. So Sun's strategy is not based on open source."

But Sun lost too much time and money with a very controversial strategy of promoting Java and fighting Microsoft proprietary extensions to the platform, while losing Java middleware market to IBM. At the same time Solaris stagnated: Solaris 8 and 9 were too late to counter Linux from the point of view of ease of administration. And AIX 5.2 largely put an end to Solaris kernel technical superiority. It can run virtual partitions while Solaris cannot (Solaris 10 countered this IBM move with zones).  In OS business you can never assume you have a 5-year reign. And too many Suns executives took huge success of Solaris 2.6 for granted.  Of course that still does not mean that Oracle developers or other developers of leading Unix applications believe in Linux marketing hype no matter what Larry Ellison might say. But during tough economic time money speaks despite the fact that there are a lot of complains about the shortcomings of the current Linux kernel:

Lance Larsh of Oracle outlined Oracle's wish list for Linux -– detailing features and additions to the kernel that would allow Oracle's high-end database servers to perform better when running Linux. Torvalds said that most of the issues brought up by Larsh are already on the "fix" list.

Many of the presenters requested improved error-reporting features in 2.5. Currently, if certain computing operations fail, the kernel doesn't know if the failed disk has a simple glitch such as a bad sector, or if the entire drive is completely trashed.

Kernel 2.5 will most likely provide more detailed information to aid troubleshooting.

More than one conference presenter lobbied for improved journaling features and enhanced control over drives' write-caching procedures.

Some presenters noted the kernel's dismaying tendency to become overwhelmed under major pressure. API's Jamal Hadi Salim said that the standard kernel doesn't do very well when it's subjected to intense network traffic primarily because it tends to do a lot of repetitive handling of each packet of information.

Jamal presented several fixes to this issue, which many felt would will improve and streamline the way the kernel deals with information overload.

The ability to name removable devices was another issue under discussion, and dynamic tracking will most likely be a feature in V2.5.

Another likely feature, judging by the enthusiasm and time devoted to discussing it, will be support for hot plugging.

Virtually all new devices that connect to a computer via USB, SCSI, or firewire connections can be hot-plugged: connected to a computer without first having to shut down the system.

The Linux kernel still can't handle hot plugging, though, and it needs to be able to offer the function to stay competitive.

This "coexistence strategy" did not work well and was eventually abandoned in 2003: both Intel and IBM were determined to make this game for Sun as difficult as possible. During the market downfall Sun learned that for most companies, the gains from converting to Intel and Linux are so large that they cannot be ignored.

For example,  E*Trade Financial in 1999 paid $12 million for 60 Sun machines to run its online trading website. In 2002 CIO Josh Levine replaced those machines with 80 Intel-based servers running Linux for a mere $320,000. That has let E*Trade bring its tech budget down 30%, from $330 million in 2000 to $200 million the same year --a big reason the company has stayed alive despite the carnage in its WEB broker business. "It's remarkable," he says. "On top of all that, website response time has improved by 30%.". Other examples include similar moves by Amazon and eBay.

Here is a pretty revealing quote about similar strategy of another big Unix player. The quote belong to an early Linux supporter who uses IBM Intel-based NUMA-Q servers (acquired with 1999 year's purchase of Sequent Computer Systems Inc.) running the Dynix/ptx Unix variant in his data center instead of Linux:

 "Dynix/ptx was always known as a killer high-end operating system -- it will run enormous user loads - but the downside [is] there have been relatively few applications written for it. That problem is now eliminated"

But the problem is that not only that Solaris was the most vulnerable to Red Hat enterprise distributions success (and UltraSparc to Intel success), but that a new "axis of evil" RH/Intel/IBM to a certain extent represent a united front against both Suns hardware and software. It is very difficult for Sun to offer a sound co-existence strategy of using Red Hat Enterprise Linux for its own benefit. As Microsoft has shown pretty convincingly, the vendor who controls the low end has a good chance eventually to controls other market segments. 

There were some realistic assumptions in the original Suns "co-existence" strategy, and the part of the strategy that tried to represent Linux as "little brother of Solaris" was not bad and had some merits.  

Growing pains slowed Linux update cycle to the level when Sun kernel developers can and should always have a technical edge over the Linux kernel developers. Solaris kernel always was, is and in foreseeable future will be better designed, more stable and more scalable than Linux kernel, especially for running databases. But the problem was that the Solaris port for Intel did not bring enough money and needed to be run on subsidies. That's difficult to do in the current economic climate and that's why AMD Opteron came very nicely into Sun's game.  some money on Opteron-based servers can position Solaris as a portable OS running on two major enterprise platforms.

Later Sun will try to position Solaris 10 on AMD as a viable alternative to expensive and bloated Red Hat Enterprise Linux. But still that did not solve that problem that in difficult economic times many corporate players might be tempted to move to the cheapest hardware supplier (Dell or Compaq for the low end servers).

The idea that Sun was trying to play can be called "Linux Personality in Solaris" -- an attempt to reproduce a more-or-less Linux-compatible  Solaris version, including a development environment with all the relevant libraries and development tools and even GUI (Gnome), but without a Linux kernel. That idea probably can work if it is sold with Microsoft "affinity" as well as in this case at least for large enterprise customers Solaris became a viable chose of a single implementation and avoid choosing between Novel Linux and Red Hat, the two major vendors that fragmented Linux enterprise market. Sun business objective is to supply a single development target for mature Linux applications with the highly scalable and stable Solaris kernel and application environment that is as close to Microsoft as one can get.

Most high level managers actually are perfectly able appreciate the  fact and cost saving due to a single Solaris distribution that contrast with Linux schizophrenic situation when there are multiple viable "personalities" for the enterprise distribution.

RH and Suse have different installers, different configuration utilities and what is important different file systems (if you use the default, which is the safest option), their own set of patches that have different side effects, own training courses and certifications. For all enterprise purposes they are two different OSes much like Solaris and AIX.   This can be called Linux Multiple Personality Disorder (LMPD) similar to human MPD which until recently they used to be diagnosed as a variety of schizophrenia ;-) . Here is the relevant quote that defined MPD:

Multiple personality disorder, or MPD, is a mental disturbance classified as one of the dissociative disorders in the fourth edition of the /Diagnostic and Statistical Manual of Mental Disorders/DSM-IV/). It has been renamed dissociative identity disorder (DID). MPD or DID is defined as a condition in which "two  or more distinct identities or personality states" alternate in controlling the patient's consciousness and behavior.

Note: Split  personality" is not an accurate term for DID and should not be used as a synonym for schizophrenia. Many patients with DID are misdiagnosed as schizophrenic because they may "hear" their alters talking inside their heads.

The following symptoms may indicate multiple personality disorder:

• Two or more distinct personalities exist within one person.
• Each personality has its own way of thinking about things and relating to others.
• At least two of the identities take control of the person's behavior.
• The person is unable to recall important personal information. Most MID patients have amnesia, or "lose time," for periods when another personality is out. They may report finding items in their house that they can't remember having purchased, finding notes written in different handwriting, or other evidence of unexplained activity.

Think about it. When companies care about competitors, it can be only because those competitors have power and because they have something unique that a lot of customers want. In case of Red Hat enterprise distribution the only attractive feature was cheap Intel hardware. People in IT industry can also to pay attention to a strong technical message from the competitor. But there was never any strong technical message from Linux.  It was purely economic message: the servers can be bought cheaper and the pool of Linux developers in offshore countries is larger/easier to grab/train/sustain, etc.

As Microsoft has convincingly had shown to IBM before in difficult economic times technical superiority is not enough. Microsoft was the first company to understand that important OSes never compete just on the basis of technical merits. They compete more as social movements. Otherwise we probably would use some flavor of  Multix, because Unix fathers managed to make some architecture mistakes that Multix designers were capable to avoid and all good Unix ideas can be ported to Multix without any problems ;-)

Yes, in comparison with Solaris kernel, Linux kernel sucks in many ways, but the Linux has huge marketing momentum due to existence of social movement around it. And this marketing momentum helps Linux to grab market share from Sun (and all other proprietary Unix vendors), not so much from Microsoft as the founders of the movement and naive evangelists like Eric Raymond expected.  Microsoft to a certain extent outsmarted Linux crowd luring them in "over complexity swamp" where they might well perish because here is the battlefield were availability of source means nothing (who cares about millions likes of code anyway; those few who really care can definitely obtain Microsoft code under NDA just as easily as Linux code). what counts is the sheer amount of  human and financial resources. Moreover there is even Microsoft "fifth column" in open source community represented by such people as Miguel De Icaza (of Gnome and  Mono fame; the former is a reimplementation of Microsoft GUI and the latter if the .Net framework respectively). All that "open source" vs "closed source" blah-blah-blah" from vendors like Red Hat is a plain vanilla hypocrisy.

From the point of view of availability of architectural information about internals Red Hat distribution is essentially as closed as Microsoft for all major subsystems. Moreover from the point of available documentation Microsoft looks better. Moreover most customers really don't care: they have other things to do then debug somebody else mistakes in kernel code. 

In this situation Scott McNealy simplistic anti-Microsoft stance became dangerous for the company and it was dropped. Objectively Microsoft is more an ally of Sun in its struggle with Intel/IBM alliance then the enemy. Scott McNealy was smart enough to understand that at the end of 2003 and the truce was signed. The other positive thing is that Solaris 10 was the second commercial Unix that provided good (lightweight) VM capabilities and unlike AIX these capability are not limited to PowerPC hardware. They are available on Opteron port of the Solaris too.

As such Solaris 10 is well positioned to regain some grounds on technical merits.  Sun and Microsoft are allies also in a sense that Linux can be considered a parasite on the Microsoft controlled Intel hardware platform specification (Microsoft controls and develops this specification and releases it to all hardware vendors for free) the same way as it is parasiting on the reimplementation of all Sun's technical contributions to Unix.

Also opening of Solaris was a nice move because while for 99.9% of customers it does not matter much whether the Solaris code is really open or not (like it does not matter for 99.8% Red Hat Enterprise Linux customers ;-) but such a move strips Intel/IBM alliance of important PR advantage. Of course they will claim that Sun's open source is inferior to Linux, but that is too subtle argument to be effective ;-) and even can be played against them as anybody who support GPL is in reality an anarchist not a capitalist :-). See Labyrinth of Software Freedom (BSD vs GPL and social aspects of free licensing debate) for more information.

The key element of  Sun's "peaceful coexistence" strategy is an attempt to increase attractiveness of Solaris platform by better integration between servers and incorporation some open source software, The first and most visible steps were adopting Gnome as GUI and opening Star Office. What  make Sun open source game more complex, if you take into account the fact that for both IBM and Intel, Sun itself is the most important target to hit in the Linux shootout ;-).  This contradictive mixture of anti-Microsoft and at the same time not very pro-Linux elements is a distinctive feature of  Sun's strategy. As Martin Vermeer pointed out in his comment on Linux today about ZDNet: One agnostic's view on open source theology 

Well... those are the main business models for money making. Then there is "learning-doing" synergy, like with Cygnus: they work on the gcc, gnu toolchain code, both improving it and learning to serve their embedded customers better. (Of course you could call this value add too.) Then there is "common code" synergy, like selling this year's, freeing last year's; freeing the light, selling the gold version; freeing the binary, selling the templates; dual GPL'ing and commercially selling a library; and many more.

But, the big one: monopoly poisoning! Why do you think Sun bought Star Office and promptly released it under the GPL, and still keeps its personnel working on it? Or why AOL/NS continues to put their own people's effort into Mozilla? Or why HP would already have opensourced OpenMail, if not for that phone call from Redmond complaining how it was "confusing customers" and reminding them of their need for Windows licenses?

Remember, a monopoly means a lack of alternatives. Provide credible alternatives (to IE, to MS Office, or to the Windows platform as a whole) and weaken the monopoly's value, both for revenue generation and for leverage.

Note that this is a business model. It is strategic and doesn't produce revenue directly. It isn't dependent on volunteer labour, but the open source licence is essential for long-term cred. It only works for big market players, that see their own home markets threatened by the monopolist's take-over, and that are not fatally dependent on the monopolist's goodwill, e.g, in licensing.

Zugzwang moves: some brilliant, some not

Still Sun was a really good "Unix chess" player and despite being in zugzwang managed to make several strong moves:

• Really brilliant move with Star Office acquisition and release it under dual licensing. That essentially means that Linux has a common office suit with Solaris and this desktop suit will work on Solaris better than on Linux. Promotion of Open Office as a de-facto standard among Office suits also makes Solaris more attractive as a desktop. But paradoxically Linux benefited from this move more than Solaris.  Still this was bold and a pretty surprising move. Then Sun released it first as freeware and then later re-released under dual licensing with GPL as one of licenses everybody except Microsoft applauded this move. that create a positive PR for Sun. Even Richard Stallman was pleased: the major peace of GNU licensed software that was missing from GNU software stack and was very difficult and expensive to create was just given to FSF by a good old Sun. What RMS probably missed is that for suits of just complexity GPL is just irrelevant. Real keys of the kingdom are in developers hands and who pays the developers controls the product.
   
• Nice move with free availability of Solaris OS for SPARC.  Due to this move Linux suddenly lost its  zero price advantage (more over Red Hat Enterprise Server  now became completely inferior product from the price/performance perspective and no sober corporate manager would ever install Red Hat Enterprise if Solaris can perform in this particular role just fine). The problem is that Linux does not compete on SPARC and Solaris still does not compete well in Intel space (although with Opteron port of Solaris this might change in late 2004). Anyway move means losing some revenue stream for Sun but neutralizes Linux's advantage as a "free" OS (Again, from skeptics point of view Red Hat Enterprise Linux is nothing but an expensive overpriced bloatware with expensive per server licensing and huge support costs)
   
• Bold move with opening source of Solaris. That by-and-large eliminated comparative advantage of having the source available. And restricted license does not matter: as Linux experiment have shown perfectly well that the size of the codebase kills open source attractiveness more effectively than any Microsoft antiLinux crusade. Very few users really need the source of large and complex operating system or application and even fewer can ever benefit from its availability.  Moreover making changes in the kernel creates a fork that need to be dealt with after each new kernel release.
   
• Brilliant move with introduction of SunPCi  card - a fully functional PC on the card that fits into PCI slot on Sun workstations. While technically brilliant it was badly executed move with dismal marketing.  Providing a cheap hardware card that can run Windows natively on a separate CPU, Solaris became more Windows friendly than Linux will ever be and that move killed one of the main attraction of Linux on the desktop for the users who have money (and that segment is understandably is the most important for Sun). But Sun's brass failed to make this card default (that would add just $200 or less to the workstation prize. Also the card as a separate product was so overpriced that few organization adopted it as an option. IMHO selling it as an option, not as a standard workstation component doomed the effort.

  Still while rather expensive for individuals, it was OK for companies and that's why in 2000 UltraSparc workstation (for example $995 Sunblade100) with  $495 SunPCi IIpro card running Solaris is probably the most interesting software development environment in existence. It enables unique level of PC integration with the Solaris, including shared files,  monitor, and keyboard (as well as CD-ROM and floppy drive).  You not only can cut and paste data between PC and Solaris applications, PC file viewer lets you instantly view and copy text from many popular types of PC files or attachments, including files created by Microsoft Word, Excel, PowerPoint, Lotus 1-2-3, COREL, and AutoCAD.  PC launcher 1.0 lets SunPCi users get seamless access and power to view, edit, and print many popular types of PC files or attachments instantly, by automatically launching the associated Windows application and file from Solaris. By incorporating PC launcher into the Solaris CDE desktop, users can share attachments and files created by Microsoft Word, Excel, PowerPoint, Lotus 1-2-3, and AutoCAD applications.
   

• Questionable move with the adoption of GNOME as a common GUI (let's them program, we will polish and use). While determined to fight Red Hat on the server Sun understands perfectly well it make sense to cooperate on the desktop front.  To a certain extent it make sense as it simplified usage of Linux applications on Solaris, but the quality of Gnome was dismal and in general Gnome (especially gnome 2.0) was such a bloatware that the it became a mixed blessing for Sun. Also regardless of all the news about Linux recently and the increasing availability of needed apps for the desktop and enterprise, the number of Linux desktops was relatively small in total Intel space in comparison with Windows desktops and percentage wise might even decrease in 2003 and 2004 after Linux became old, last century story and lost large part of its marketing appeal.  Even for this open source opened site the number of Linux users deteriorated from almost 20% in 1999 to less then 10% in 2004. That means that the percentage of Linux desktops like the number of Solaris desktops may actually be decreasing. And misery loves company. I think that  SunPCi  card represented a more interesting direction that Sun's brass did not managed to exploit to its own advantage.
   
• Brilliant attempt to provide an easy driver migration from Linux (the task of preserving a market share on Intel-based Solaris deployments has huge strategic value for the company because it  might provide some interesting openings in the future, especially if IA64 will not live according to the expectations). On Aug 22, 2000 Sun released a Linux-2-Solaris driver migration kit.
   
• Good, but badly executed attempt to provide a better interoperability with Linux by using ports of the of best GNU utilities as part of Solaris installation. Sun's record in providing decent utilities with Solaris is horrible (and I really hate stupid Sun's brass for killing TCL project) and here GNU applications, as bad as they are in comparison with best windows counterparts, might help.  But Sun failed to include most of  best on GNU utilities (like mc, vim, expect) in the default installation (only bash and less were included by default).  Of course many GNU utilities are not that impressive from the technical point of view (bash cannot compete with ksh93 in shell scripting, but it is a much better interactive shell) but that greatly increases the user-friendliness of the Solaris environment (for example mc was/is a nice addition to the Solaris command line tools. Seeing Linux applications and desktops on servers that aren't running the Linux kernel was a big surprise to  Linus Torvalds.

  "It used to be the other way around," said Torvalds, speaking with Computerworld on the LinuxWorld show floor in San Jose last week. "A few years ago, we had to find a way to run [The Santa Cruz Operation Inc.] applications on Linux."

• Attempt to sell Linux appliances. Sun bought Linux appliance vendor Cobalt and continued to sell those appliances with Linux. Bad thing is that Sun hugely overpaid for it and eventually killed them.

Etc

Society

Groupthink : Two Party System as Polyarchy : Corruption of Regulators : Bureaucracies : Understanding Micromanagers and Control Freaks : Toxic Managers :   Harvard Mafia : Diplomatic Communication : Surviving a Bad Performance Review : Insufficient Retirement Funds as Immanent Problem of Neoliberal Regime : PseudoScience : Who Rules America : Neoliberalism  : The Iron Law of Oligarchy : Libertarian Philosophy

Quotes

War and Peace : Skeptical Finance : John Kenneth Galbraith :Talleyrand : Oscar Wilde : Otto Von Bismarck : Keynes : George Carlin : Skeptics : Propaganda  : SE quotes : Language Design and Programming Quotes : Random IT-related quotes :  Somerset Maugham : Marcus Aurelius : Kurt Vonnegut : Eric Hoffer : Winston Churchill : Napoleon Bonaparte : Ambrose Bierce :  Bernard Shaw : Mark Twain Quotes

Bulletin:

Vol 25, No.12 (December, 2013) Rational Fools vs. Efficient Crooks The efficient markets hypothesis : Political Skeptic Bulletin, 2013 : Unemployment Bulletin, 2010 :  Vol 23, No.10 (October, 2011) An observation about corporate security departments : Slightly Skeptical Euromaydan Chronicles, June 2014 : Greenspan legacy bulletin, 2008 : Vol 25, No.10 (October, 2013) Cryptolocker Trojan (Win32/Crilock.A) : Vol 25, No.08 (August, 2013) Cloud providers as intelligence collection hubs : Financial Humor Bulletin, 2010 : Inequality Bulletin, 2009 : Financial Humor Bulletin, 2008 : Copyleft Problems Bulletin, 2004 : Financial Humor Bulletin, 2011 : Energy Bulletin, 2010 : Malware Protection Bulletin, 2010 : Vol 26, No.1 (January, 2013) Object-Oriented Cult : Political Skeptic Bulletin, 2011 : Vol 23, No.11 (November, 2011) Softpanorama classification of sysadmin horror stories : Vol 25, No.05 (May, 2013) Corporate bullshit as a communication method  : Vol 25, No.06 (June, 2013) A Note on the Relationship of Brooks Law and Conway Law

History:

Fifty glorious years (1950-2000): the triumph of the US computer engineering : Donald Knuth : TAoCP and its Influence of Computer Science : Richard Stallman : Linus Torvalds  : Larry Wall  : John K. Ousterhout : CTSS : Multix OS Unix History : Unix shell history : VI editor : History of pipes concept : Solaris : MS DOS :  Programming Languages History : PL/1 : Simula 67 : C : History of GCC development :  Scripting Languages : Perl history   : OS History : Mail : DNS : SSH : CPU Instruction Sets : SPARC systems 1987-2006 : Norton Commander : Norton Utilities : Norton Ghost : Frontpage history : Malware Defense History : GNU Screen : OSS early history

Classic books:

The Peter Principle : Parkinson Law : 1984 : The Mythical Man-Month :  How to Solve It by George Polya : The Art of Computer Programming : The Elements of Programming Style : The Unix Hater’s Handbook : The Jargon file : The True Believer : Programming Pearls : The Good Soldier Svejk : The Power Elite

Most popular humor pages:

Manifest of the Softpanorama IT Slacker Society : Ten Commandments of the IT Slackers Society : Computer Humor Collection : BSD Logo Story : The Cuckoo's Egg : IT Slang : C++ Humor : ARE YOU A BBS ADDICT? : The Perl Purity Test : Object oriented programmers of all nations : Financial Humor : Financial Humor Bulletin, 2008 : Financial Humor Bulletin, 2010 : The Most Comprehensive Collection of Editor-related Humor : Programming Language Humor : Goldman Sachs related humor : Greenspan humor : C Humor : Scripting Humor : Real Programmers Humor : Web Humor : GPL-related Humor : OFM Humor : Politically Incorrect Humor : IDS Humor : "Linux Sucks" Humor : Russian Musical Humor : Best Russian Programmer Humor : Microsoft plans to buy Catholic Church : Richard Stallman Related Humor : Admin Humor : Perl-related Humor : Linus Torvalds Related humor : PseudoScience Related Humor : Networking Humor : Shell Humor : Financial Humor Bulletin, 2011 : Financial Humor Bulletin, 2012 : Financial Humor Bulletin, 2013 : Java Humor : Software Engineering Humor : Sun Solaris Related Humor : Education Humor : IBM Humor : Assembler-related Humor : VIM Humor : Computer Viruses Humor : Bright tomorrow is rescheduled to a day after tomorrow : Classic Computer Humor

The Last but not Least Technology is dominated by two types of people: those who understand what they do not manage and those who manage what they do not understand ~Archibald Putt. Ph.D

The statements, views and opinions presented on this web page are those of the author (or referenced source) and are not endorsed by, nor do they necessarily reflect, the opinions of the Softpanorama society. We do not warrant the correctness of the information provided or its fitness for any purpose. The site uses AdSense so you need to be aware of Google privacy policy. You you do not want to be tracked by Google please disable Javascript for this site. This site is perfectly usable without Javascript.

Last modified: March 12, 2019