Red Hat Set To Surpass Sun In Market Capitalization

mytrip writes "In what may come to be seen as a deeply symbolic moment in the history of operating systems, Red Hat is on the verge of surpassing Sun Microsystems' market capitalization for the first time. Sun, perhaps unfairly, represents a fading Unix market. Red Hat, for its part, represents the rising Linux market. Given enough time for its open-source strategy to play out, Sun's market capitalization will likely recover and outpace Red Hat's."

Comments

The sum of Linux vendor capitalization

[Facetious]

The makes me curious. If all Linux vendors had an equivalent of publicly traded market capitalization, what would their sum total be? Naturally it would be lower than Microsoft's $153B (as of this morning), but that isn't bad considering Linux can be had for free. (BTW, I remember back when msft's market cap was over $400B).

Thank you Sun

[rlp]

I for one hope that Sun not only survives, but prospers. Sun has greatly contributed over the years to the development community, particularly FOSS developers.

Re:Thank you Sun

[aliquis]

Personally if I got to choose one of either all of Suns knowledge, experience, code and products or Redhats I'd for sure go with Sun.

Obviously the market works differently =P

Re:Thank you Sun

[aliquis]

(differently = who is more likely to make money of their knowledge, experience, code and products :D)

Re:Thank you Sun

[tnk1]

(differently = who the stock market thinks is more likely to make money of their knowledge, experience, code and products :D)

Corrected that for you.

Re:Thank you Sun

[InlawBiker]

Sun has been a great innovator, but when they were the only game in town they charged obscene prices for their products and services. It helped open the door for Linux and Sun has only itself to blame.

When you walked into a data center ten years ago all you saw were Sun servers. Where I work now I'm hard pressed to find a single Sun box anywhere.

Re:Thank you Sun

[the eric conspiracy]

I for one welcome our new Linux Overlords

Re:Thank you Sun

[rbanffy]

If you refuse to make a profit, you can't blame those who don't when they make one by offering useful services to a large community.

Re:Thank you Sun

[ChienAndalu]

If you think the stock market is wrong, you can earn money.

Re:Thank you Sun

[sqldr]

It's not just price that puts people off sun, it's difficulty of use. There's all sorts of silly pointless things like why you have to put a 01 in front of mac addresses when configuring their ldap client to jumpstart stuff (apparently the 01 means "ipv4" or something, but after learning what it was, i figured it was too irrelevant to bother remembering what it was for), or how when you configure ldapclient it defaults to trying to do name lookups over ldap to look up the name of the ldap server (come on guys, that's a schoolboy error of a bug).

We got 2 x4150s in. Sun hardware, sun operating system.. what could go wrong? 2 weeks later, I still haven't got the bloody Sun operating system installed on Sun hardware over jumpstart, and the vendor providing tech support hasn't got it running either. Why? Turns out it's because the baud rate of the iLom runs at 115khz, and the install image defaults to 9600, and the only way to fix that is to MODIFY THE IMAGE (!!!!). Lost days over that one.

Sun should stop wasting time making pointless crap like "webstack" which is basically apache and mysql, and concentrate on what's missing from their OS - get pkg running, stop making commands difficult to use, and fix jumpstart.

Re:frist psot

[Hognoxious]

A thousand and one posts saying that it's illegal/immoral/impossible to make money from open source software will be along soon.

They'll be followed shortly after by sevaral thousand more complaining that all corporations are evil and should be banned.

In turn those will be followed by several million arguing that google are/aren't evil, or disputing the subtle nuances between doing evil and being evil.

In other words: normal service will be resumed as soon as possible. The tuna salad is off, by the way.

A relatively unimportant event

[fishwallop]

I fail to see why this is a "deeply symbolic moment in the history of operating systems" and not merely a moderately interesting moment in the corporate history of the respective companies (or, more specifically, in Red Hat's corporate history). Red Hat may represent Linux, but it's not Linux, and market capitalization, being a function of share price, is a less interesting metric then any measurement of the actual use of the operating systems these companies produce. Anyone who remembers the Red Hat IPO will know that share price is more closely tied to hype than to particularly signficant tecnical advances.

Re:A relatively unimportant event

[fm6]

I fail to see why this is a "deeply symbolic moment in the history of operating systems" and not merely a moderately interesting moment in the corporate history of the respective companies (or, more specifically, in Red Hat's corporate history).

It's symbolic because Sun was one of the leaders of the big change that occurred at the beginning of the microcomputer era, when Unix started to replace the old mainframe OSs. And it's symbolic because Sun is the last major player to consider Unix part of its core strategy. Other Unix vendors have become insignificant (SGI, SCO), disappeared, or changed their emphasis to Linux and Windows (IBM, HP).

Also, if you want to run a supported, commercial Unix on commodity hardware, Solaris is really your only option. Which is why both HP and Dell offer Solaris preinstalled. Though I don't suppose they sell a lot of those systems.

As a benchmark of the rise of Linux and the fall of Unix, yeah, it's not that big a deal. But symbols and benchmarks are different things.

This may all be a big yawn to somebody whose career started after Linux began to take over. But to those of us who spent most of our professional lives working with Unix (I started in the early 70s, before Unix was even available commercially; I've worked for 5 different Unix vendors, including Sun) it's as big a symbol as the takeoff of a certain helicopter on Tuesday.

Wrong.

[AltGrendel]

Definition - Market capitalization:
an estimation of the value of a business that is obtained by multiplying the number of shares outstanding by the current price of a share

Re:Wrong.

[QuantumRiff]

I understand that. I was asking if looking at the market capitalization really said more about how the company was doing than its sales numbers? Do we now judge the success of tech companies by looking at what non-technical financial people think the company might be worth in the very short term?

Really, I think it says more of the "investors" that they think a company with sales of 700M a year should be worth (in market capitalization terms) the same or more than a company with sales of 13Billion a year.

Re:Wrong.

[Estanislao Mart�nez]

That is largely right, except for one detail: Red Hat is producing profits, but Sun recently posted a big loss, mostly due to a 1.45B impairment of goodwill charge. (In English: they revised their estimate of the value of some of the companies they've purchased, down 1.45B).

To answer the original thread poster's question: it's not how much you sell, it's how much your keep.

Re:Wrong.

[ceoyoyo]

It's almost as silly as judging a company by it's gross sales.

What you're looking for is PROFIT. Sun's profit was 88 million, or 11 cents per share in the last quarter (down 73%) while Red Hat's profit was 24 million, or 12 cents per share, up about 7%. Sun is forecasting that they will lose money over the next year, while Red Hat is forecasting (and analysts agree) that they will continue making money.

Sun still made more money than Red Hat, but even Sun agrees that's going to change.

Re:Wrong.

[jadavis]

I was asking if looking at the market capitalization really said more about how the company was doing than its sales numbers?

It's easy to get high revenue -- just sell stuff at or below cost. You won't make a profit, but you'll have a lot of sales.

Do we now judge the success of tech companies by looking at what non-technical financial people think the company might be worth in the very short term?

There are many ways of evaluating a business as a whole, some are subjective, e.g. something "only a technical person can understand why this is good", and some are objective, e.g. something that any financial expert can understand.

You may like to think that you have a lot of insight about the value of a company because you know something about the technologies they use or produce. And that's true, to some degree.

But your ignoring the numbers outright. And the numbers mean a lot. If Sun continues to lose money, and never turns a profit, then its value is precisely the liquidation value of its assets. I'm not saying that's what will happen, but profit is clearly important to the value of a business.

And you're also ignoring all the subjective aspects that you have no ability to evaluate. How well are they targeting their value proposition in the latest advertising campaign? How competent are the executives? Are they using the right pricing model?

I don't buy for one second the idea that non-technical people are clueless about the technology business.

And I also don't buy for one second this "very short term" mentality. Investors choose investments that have some intrinsic value that they believe will hold in the long term. If they don't, they know they might be caught holding the bag, and lose everything.

There are exceptions, "irrational exuberance", etc. But most money is tied up in things with intrinsic value. Those investments may go up or down, but they generally won't go to zero.

Re:Wrong.

[tinkerghost]

And I also don't buy for one second this "very short term" mentality. Investors choose investments that have some intrinsic value that they believe will hold in the long term. If they don't, they know they might be caught holding the bag, and lose everything.

I am going to point you to the current financial meltdown of some of the biggest banks & trading houses in the US(World) as proof that a large number of "Investors" are choosing investments that have intrinsic value. Most of these "commoditiezed mortgages" were junk, that's why they were being sold. However "investors" saw a chance at a quick buck & bought them up like a 3 year old buying penny candy.

I use "investor" loosely here as they aren't investors, they are speculators. The majority of the money being moved around on any given day isn't about long term investment. It's about short term speculation - hoping to grab a 1-3% growth in 14 days or less, followed by bailing & reinvesting hoping for an annual return of 20% or more.

Investors buy for the long haul. IBM stock isn't a huge mover & isn't likely to make you rich. But it usually shows steady growth and has a dividend.

The majority of the money in the stock market may in fact be invested in long term stocks and bonds, but what people see on the news at the end of the day is how the day traders and speculators are moving. And they are always moving to maximize todays profit - even if that means destroying the long term sustainability of a company to do it.

Shareholders have ousted several boards for favoring long term profit over the quarterly report, so it's not surprising that boards & CEO's are pushing more for the quick buck & less for the sustainability for that profit.

Re:Wrong.

[Chris Burke]

Is there anyone who thinks that market cap is a good metric to judge a company's overall success by at this particular moment in time? Pretty much every stock has been hit and hit hard, and the degree to which the size of that hit has anything to do with the company itself is highly questionable. The relationship between stock price and a company's success is ephemeral in the best of times, but now?

Re:Wrong.

[civilizedINTENSITY]

I think you're right about their confusion. RH's net income last year was $403,000,000. This is down from their net income of the year before: $473,000,000. What I think is confusing people, perhaps, is their cash flow statement, which shows a negative value of $1,3B. Where did all that cash go if they have net income? Treasury Stock. A negative $2,6B went to stock repurchase.

If a corporation reacquires some of its stock and does not retire those shares, the shares are called treasury stock. Treasury stock reflects the difference between the number of shares issued and the number of shares outstanding. When a corporation holds treasury stock, a debit balance exists in the general ledger account Treasury Stock (a contra stockholders' equity account).

If the corporation were to sell some of its treasury stock, the cash received is debited to Cash, the cost of the shares sold is credited to the stockholders' equity account Treasury Stock, and the difference goes to another stockholders' equity account. Note that the difference does not go to an income statement account, as there can be no income statement recognition of gains or losses on treasury stock transactions. (This, of course, is reasonable since the corporation operates with total "insider" information.)

The trend looks better, then: Net Tangible Assets (in 1,000s): $1,808,000 $4,032,000 $2,805,000
if we took into account the buyback (as though they had bought IBM instead of themselves)
$1,808,000+2.6B=4.4B $4,032,000 $2,805,000, which is UP not down from the previous year.

Treasury stock causes the Balance sheet and Cash Flow to look negative, since the stock purchased come out of cash flow (as Sale Purchase of Stock (2,587,000)) and also shows up as a negative on the Balance sheet. Since the Income Statement doesn't show Treasury Stock, it looks good.

Further evidence...

[AKAImBatman]

...that marketing trumps technology. Sun has some incredible tech and even delivers x86 servers at highly competitive prices. Yet because Sun's marketing sucks worse than a black hole, generating new customers is a huge issue for them. As far as I can tell, the vast majority of their business is still through customer reps with little attention paid to the market as a whole.

I personally think that Sun could be successful in quite a few areas of the market. Not the least of which is as a serious competitor to Dell's server business. But first, Sun has to figure out how to communicate with the average customer. Giving their software complex prefixes like "Sun Java System", branding everything with "SPARC" even when it isn't SPARC, changing their market ticker to JAVA, and giving up on new markets before they've made inroads aren't exactly painting Sun in a positive light.

Dear Mr. Schwartz: Please hire a real marketing department and see to it that your product line makes sense to the average consumer. KTHXBYE.

Re:Further evidence...

[eln]

I agree, but it's not only marketing. Sun has apparently decided to go into the support and commodity hardware market. In commodity hardware, margins are razor thin, so they really have to distinguish themselves. In my recent experiences with Sun x86 systems, quality has been something of a problem. They say it was a temporary issue with one of their plants in Mexico, but when we ordered a ton of x86 boxes about a year ago, it took much longer than it should have to get to us, and the failure rate was unusually high.

Also, when they released the x4100 Mk2, they claimed it was virtually identical to the Mk1, and there would be no issues. However, it turns out they made some fairly significant changes such as changing the vendor of the on-board network cards to one that the OS image we were using at the time had no support for. It also had a different type of PCI port (PCIe versus PCIx, IIRC), which meant all of the extra NICs we had lying around were suddenly useless. Had they told us of these changes, it would have been no problem. Instead, they just told us our order was being changed to the "virtually identical" Mk2, and we had to scramble when we got them. Not great customer support there. After that incident, we actually stopped using Sun for x86 hardware entirely.

Going back to marketing though, they are really pushing this "Open Systems" thing, which is nice and all, but their salespeople don;t know how to sell it. At a recent presentation, the Sun sales guy was talking up Open Systems, and a member of the audience asked, "If everything is open and interchangeable, why shouldn't I just use your free open source software and go buy a cheaper system from Dell? What is the advantage of your box, if it's commodity like the rest? Why should I buy from you?", and the sales guy had no answer for him! He actually stumbled over his words for about 30 seconds, at one point actually saying there was "no reason" before one of his colleagues finally pipes up with something about "end to end support".

Maybe they need to be touting the end to end support first, and the open systems stuff second. Suits tend to like open source because it's a lot cheaper, not because they're big on the philosophy, so stop pushing the "open source" thing so hard when the open source bit is the part you're giving away for free. Market the entire platform as an end-to-end solution, and throw in the open source part as an aside. Sun's marketing team doesn't seem to get that.

Anyway, that was a bit long-winded, but the point is that Open Source isn't going to save Sun by itself. They have more problems, and I see them surviving as a much smaller and less interesting company than they are today if they stick to the path they're currently on.

Re:Further evidence...

[Samschnooks]

That's what happens when you manage from an engineering mind set. When SUN started, they were leaders in technology: no one was doing what they were doing and there was a need for that technology that they developed. The market only rewards better engineering when it satisfies their needs. Now, computing power is a commodity and generic software is a commodity.

The market that is left in SAP type of stuff. Software that allows a company to become more competitive. There's more new markets, but I'm too old to know where it is - and I'm ten years younger than the senior management of SUN. The next software big thing is going to come from an 18 year old in his parent's basement and I don't think it's going to be PC based. It's going to be some device (I don't know, a toaster that takes a whole loaf of bread, slices it, toasts the bread, and cooks the bacon) and it'll become "necessary".

Either SUN gets some young blood or they will turn into - I don't want to say it. They have MANY patents and they invented MANY of the things that run the internet (thanks Joy!) and I'm sure there are folks violating those patents. Just saying....

Re:Further evidence...

[asv108]

Its not just marketing, it being able to purchase products online at the actual price. You can't just login to sun.com and buy servers via a online portal with a corporate discount like Dell or HP. You have to talk with a rep to get the actual real price, not the phony MSRP price.

Re:Further evidence...

[C_Kode]

With the current way Sun does business hardware wise it cannot compete with Dell directly. Sun sells higher end hardware and Dell sells lower end servers very cheap. You can buy two Dell servers loaded with the same memory and SCSI drives instead of SATA drives, with 3 years support for what a single Sun server would cost you. I have several JBoss clusters and I just throw Dell PE1950s at them. If one crators, the rest of the cluster just hums along. Sun competes with HP, IBM, and the like. They DO NOT compete with Dell. Although, Dell can definitely compete (and steal market share) from Sun. Especially now with the current state of economy and everyone being so cost conscious.

Re:Further evidence...

[AKAImBatman]

For quite some time, Sun was undercutting Dell on AMD64 servers. I have been told that you can still get the servers cheaper if you have a rep. The problem is that Sun gave up on the rest of the market after only a short push. (You might remember the "rhymes with hell" ads here on Slashdot.) And dealing with a central sales rep is a pain and a half when any segment of a large company can order a server through Dell.com.

So I'm not surprised that you think Sun doesn't compete with Dell. As I said, they have a massive failure in their marketing department and no real commitment to expanding their business.

Re:Further evidence...

[DesScorp]

"Dear Mr. Schwartz: Please hire a real marketing department and see to it that your product line makes sense to the average consumer. KTHXBYE."

Schwartz is part of the problem at Sun. When McNealy named Schwartz as his successor, a collective "huh?" was heard all over the tech world.

Schwartz's gamble seems to be "give every piece of software away, and sell commodity hardware".

This is, in a word, foolish. IBM doesn't give everything away. Nor does HP, or Apple. They carefully balance their open source obligations against the need for exclusivity in some areas. Sun should be trying to emulate Apple in many ways (and IBM in some others), but instead, is trying to remake itself into a Red Hat that sells cheap X86 hardware, and this is a recipe for doom.

If Linux is how much can be made free...

[thtrgremlin]

... then Linux market capitalization is how much software that previously cost money was made free, so if Linux can be considered directly responsible for killing Microsoft, which I think is some peoples objective, that puts their market capitalization at $400B - $153B = $247B. That means Linux has 1.6x the market capitalization of of Microsoft just in Operating Systems! That doesn't even begin to include all the other great FlOSS out there.

Add to that the average wage of a software engineer times the number of man hours contributed to FlOSS, and you can quickly see how Microsoft is getting its butt kicked!

I love the new math!

Re:If Linux is how much can be made free...

[hairyfeet]

I think it is more that they forgot where their bread was buttered. I mean, there is a REASON why you see websites all over the place showing you how to turn Win2K3 and 2K8 server into a desktop OS. That is because the business user, MSFT's bread and butter, has been ignored while they got into this multimedia pissing contest with Apple.

It is like all the business guys at MSFT was replaced by a marketing drone screeching "We can be as hip as Apple and as cool as Google! Really, we can! Stop laughing at me!" and the worst part is they STILL don't get it. Folks got used to using MSFT Operating Systems because that is what they used at work. Businesses like boring as shit, simple low resource OSes so they don't need to buy a gamer rig for their secretaries. Before it was all good- You had Win9X for home and WinNT/Win2K for business. But now they are determined to stuff us all in this multimedia bloated nightmare of an OS instead of giving business users a simple boring bloat free work environment.

So I have no doubt that MSFT will continue to decline. That is what happens when you ignore a large chunk of your customers. And I am sure that there will be many Linux distros willing to try their hand at the business market, just as Apple is wedging their foot slowly but surely into the home market. Maybe if Win7 turns into another Vista failure(which I personally predict it will) then they will listen to their customers. Until then I see good things for companies like Red Hat that focus on the needs of business, instead of trying to go with a "one size fits all" approach to every PC market.

Don't give that much credit.

[jellomizer]

SUN's market was traditionally on High End equipment. Standard PC hardware has been getting to the Good enough category, and replaceing the need for the high end stuff. Even if the high end stuff today is that much more high end, we are reaching a point where we need less high end equipment.
the 80's almost every major university had its own super computer. 90's they had a mainframe, 2000's they have high end microcomputer based servers.
SUN product line has been between mainframe and microcomputers. Now their new stuff is either to much for what people need to too expensive for what you get.

Linux growth has always been fasted with the Unix Corps who are upgrading to a new network, and it is way cheaper for a Unix corporation to switch to Linux (or Old Unix to New Unix) then to Windows. (Windows to Linux costs a lot more).

It might be interesting...

[zogger]

..if Redhat sold netbooks, laptops and desktops and servers pre loaded with linux that "just worked", all of it, no hardware gotchas anyplace.

Re:Riiiight . . .

[tnk1]

Thoroughly unlikely. Sun would be bought out long before they needed to declare bankruptcy. Their star may not be in ascension, but the company has real value.

Sun has always been very bad in business

[Anonymous Coward]

It's not that Sun would have done the wrong things. Their problem for the last 15 years has been that they have been over-careful of commitment into their own business. From business perspective they have been apathic at best, and I iterate the same I said in 90s already: I would not invest a penny into them.

They have not taken any steps to fix their real problems so far, so likely they will just keep sinking slowly until the company will be bought by someone who knows how to actually run companies. Perhaps even by Microsoft, who knows :)

Market cap is speculative bullshit

[blind biker]

Anyone with a brain (ok, it takes perhaps a bit more than just half) knows that the stock market, as an entity, is an idiot.

There, I said it. I expect to be modded down by Linux fanboys (which is NOT the same as intelligent Linux users, mind you. I like to think I belong to the latter).

Re:frist psot

[Anonymous Coward]

Did You Know? After maintaining a vow of silence for almost 7 years, Red Hat Linux founder Marc Ewing now freely admits that he named Red Hat Linux after Limp Bizkit frontman Fred Durst's trademark red New York Yankees baseball cap.
Durst and Ewing met in Ewing's hometown of Raleigh, North Carolina (Durst was raised in Gastonia, NC), where they became fast friends, sharing the same passion for low-level system programming.
Durst collaborated with Ewing on the first preview beta of Red Hat Linux before the demands of his rocketing stardom forced him to abandon his hobby and tour with his band.
Durst's position on the development team was filled by Damien Neil, and not many know of his contribution to the popular Linux distribution; however, a google search through the source code on Redhat.com (http://www.google.com/search?q=wfd+site:redhat.com) reveals many snippets of code authored by 'wfd', Durst's initials (William Frederick Durst).
Durst asked Ewing to keep his 'geeky' roots a secret as it would not lend itself to Durst's bad boy image, but as Ewing points out, it was "only a matter of time" before the origins of his NASDAQ-100 company's name were uncovered.
--mfh

Finally

[pha3r0]

2009, the year of the linux desktop!

End-to-End solutions are Sun's real selling points

[thtrgremlin]

No product can make money if you can't sell it. Doesn't matter how good it is. Call it marketing hype, but how is anyone going to know that your systems are superior if you don't say as much. That seems like such a screw up that the sales rep couldn't explain the whole "end-to-end" solution issue. Personally, I don't think that wasn't a gotcha question at all, that should have been prompted as a great opportunity for the rep to explain what kind of business Sun is really in.

But if Sun doesn't even know what kind of business they are in... one really needs to wonder how much longer they are going to last. I am sure any sales rep at Red Hat would drool at the opportunity to answer a question like that from a real customer in front of a crowd that would have influence over possible huge sales. Hell, even I live to hopefully answer questions like that! Shortly after Vista came out, I was at a Fry's Electronics, and a sales associate was telling this guy looking to buy a bunch of machines that Vista was the hot new thing and that he needed it for his business because everyone was going to be using it in the next 6 months when XP died. I bust up laughing and warned the customer that was likely the worst advice you could get to drive a business, let alone a speculation I agreed with. From what he talked about needing for his business, I did my little Linux/FlOSS speech, but really recommended Red Hat, and explained a bit about their free product with full end-to-end support, and that they could likely best advice him on what would work best for HIS business needs. I am fairly sure I sold the guy on it. The Fry's sales rep was upset and just kept saying I was wrong and that everyone was going to use Vista. I humored him and asked "So what's Vista got that XP doesn't?". He started explaining the higher system requirements (as if it was a selling point), and went on to show me Aero. I gave a long *sigh*, and walked away. I should hang out there more often :)

Bad marketing

[Jim Hall]

I for one hope that Sun not only survives, but prospers. Sun has greatly contributed over the years to the development community, particularly FOSS developers.

Sun has certainly contributed many highly-visible projects that we just take for granted these days: NFS, OpenOffice, Java, GNOME, etc. And ZFS is very powerful, but hasn't really made it to other places yet. However, it just seems Sun doesn't know what to do with it, or how to market it.

A few years back, I got to visit Sun for an executive briefing. We met with a lot of higher-ups at Sun (including Scott McNealy.) I repeated to whoever would listen that Sun needed to get their act together: Figure out an (easily-understood) strategy for Sun and FOSS, and move with it. Separate the hardware and software marketing; and at the same time, let me choose systems "menu-style" just like buying a Dell. Simplify your product lines and marketing. Release a consumer-based UNIX distro for commodity PC systems that has the polish of Linux (the apps are there - Firefox, OpenOffice, etc. - so for 99% of the population that's the "compatibility" they need.)

Yes, Sun has done some of these things, but not in a coherent way, and certainly not in a simple way. Things are just too hard to go through Sun.

Sun needs to get organized if they want to remain competitive.

Re:Bad marketing

[alexborges]

Thing is, they shouldve killed solaris and become a full fledged Linux or BSD shop with enormous expertiese on Unix and great services built on open source java. I think they probably could still pull it of, albeit with an enormous ammount of pain.

They misread the game and didnt wanna get their feet cold: this is what happened to them.

Kind of whats going to happen to microsoft and what is happening to music and movie distributors.

Re:Bad marketing

[ToasterMonkey]

I think you're seriously undervaluing Solaris. Killing it wouldn't be doing anyone any favors.

Re:Bad marketing

[An dochasac]

And ZFS is very powerful, but hasn't really made it to other places yet. However, it just seems Sun doesn't know what to do with it, or how to market it.

Read only ZFS has been in OSX since 10.5, full blown ZFS is in OSX Snow Leopard as well as various flavors of FreeBSD. It will go into Linux if/when the kernel licensing issues can be overcome, it's already available via FUSE. ZFS is also used in several storage appliances by Sun and others.

A few years back, I got to visit Sun for an executive briefing. We met with a lot of higher-ups at Sun (including Scott McNealy.) I repeated to whoever would listen that Sun needed to get their act together: Figure out an (easily-understood) strategy for Sun and FOSS, and move with it.

What is so difficult to understand? Sun divisions have contributed more than 1/2 of the OpenSource code out there (OpenOffice, Java, VirtualBox, MySQL, OpenSolaris...) Sun is a systems company. That code is out there so the world doesn't end up locked into systems built around Microsoft and Wintel fat clients. Where I think Sun could improve is in selling support and integration services for RedHat, OSX other *nixes. Sun definitely has the expertise and some Sun employees have expertise in enough *nix variants that they won't paint you into a corner. In fact, anyone who has ever been involved with porting from Linux to Solaris will tell you that step #1 is improve the quality of the code, Solaris/Forte doesn't let coders get away with the kind of sloppiness that gcc/Linux does.

Separate the hardware and software marketing; and at the same time, let me choose systems "menu-style" just like buying a Dell. Simplify your product lines and marketing. Release a consumer-based UNIX distro for commodity PC systems that has the polish of Linux (the apps are there - Firefox, OpenOffice, etc. - so for 99% of the population that's the "compatibility" they need.)

Have you been watching Nexenta or OpenSolaris 2008.11? This X86/desktop "usability" gap has closed significantly and if you're using Linux for a server, you really need to look at OpenSolaris, ZFS admin is far easier than anything you'll find on RHEL, svcadm takes the randomness out of system services and dtrace is awesome for diagnosing issues in production systems.

Sun needs to get organized if they want to remain competitive.

I agree wholeheartedly!

Re:Good enough?

[rbanffy]

It really depends on your workloads. I see SPARC boxes running rings around x86 boxes the same price range on highly parallel workloads.

Re:frist psot

[Z34107]

A thousand and one posts saying that it's illegal/immoral/impossible to make money from open source software will be along soon. They'll be followed shortly after by sevaral thousand more complaining that all corporations are evil and should be banned. In turn those will be followed by several million arguing that google are/aren't evil, or disputing the subtle nuances between doing evil and being evil. In other words: normal service will be resumed as soon as possible. The tuna salad is off, by the way.

Wow! The Readers Digest version of Slashdot!

Your opinion is worth squat.

[jotaeleemeese]

Can any MS based servers be configured to divide resources on "virtual" machines before the operating system is booted?

And then each one of those virtual machines can run a couple of dozen instances of a operating system?

And then can you assign on the fly CPUs, memory or I/O cards to any of your initial virtual machines?

Can actually any Linux machines do this?

You guys talk about desktops like if getting the latest version of Gnome working was the coolest thing regarding technology. That is not Sun's beef, Sun's realm is completely different, but clearly its main clients (specially banks) are being hit specially badly, but the stuff Sun does is tremendously cool, if you are technically skilled to understand what it is.

Re:Riiiight . . .

[Sun.Jedi]

First, in a friendly way ... nobody gives a rat about workstation class machines. Any fool can dump Ubuntu, Fedora, Windows, BSD, OpenSolaris, or whatever, and google up enough support/fun tools to get the job done and post on /. on most any cheap-ass 2 year disposable wintel capable machine.

So, on server class machines... number 1 reason: Support contracts.

Sun is one stop shopping; hardware and software by the folks that make it. Dell also supports RH linux through their customer care center, and I'll assume that HP/IBM do as well, but they are not RedHat, they are $vendor with RH knowledge and expertise. Thats a separate subscription to get RedHat support and you then get the RHNetwork portals in addition to standard phone support. So you pay twice (1 for $vendor and 1 for RH), and it ain't cheap as Sun's.

Number 2 reason: Reliability.

SPARCs just don't die. When they do, its very pretty of course, but it just doesn't happen as often as Intel/AMD architectures do.

Also, Suns do not often have the compatibility problems that Intel/AMD arch's have. By compatibility, I mean the mobo + raid + firmware + kernel version + PCIx firmware + BMC version = "unsupported" type compatibility.

Fact is, I've been admin on Sun's for nearly 15 years, been through the really bad 5/7 releases and lots of other SUN 'badtimes'. hey are nothing like the hassles I have to go through daily with AMD/Intel arch's. I'm in a 4:1 Sun:Intel/AMD shop, and have a documented (ticketing system) 5:1 Intel/AMD:SUN hardware problem ratio.

Yes, my alias is 'sun.jedi', I've worked on Sun's a long time. This was not intended as a 'fanboy' post. I'm a beer/vacation fanboy before I'm a SUN fanboy.

Relative to what?

[DesScorp]

Sun has been a great innovator, but when they were the only game in town they charged obscene prices for their products and services. It helped open the door for Linux and Sun has only itself to blame.

When you walked into a data center ten years ago all you saw were Sun servers. Where I work now I'm hard pressed to find a single Sun box anywhere.

Sun was expensive compared to what? Windows boxes? Linux boxes that came later? Sun became the huge company it was because they were far more affordable than what IBM and Digital was charging in the 80s, and everyone ran to them. It's kind of hard to blame Sun because some guy in Finland came up with an alternative that ran on El Cheapo X86 hardware, and then gave it away to the whole world.

Re:Relative to what?

[Anonymous Coward]

When Linux became viable, Sun was among the first to become aware that big apps like Oracle were being ported. It was a golden opportunity to come out with a Sun Linux and support it along with Solaris. It would have let them tap into smaller markets and crush Red Hat.

Instead they tried like hell to prop up their massively profitable Sparc platform. For smaller projects everybody went straight to Linux, or even Windows. Now... who in their right mind would buy Sparc hardware over HP or Dell servers running Red Hat?

There are of course cases where Sun is the clear favorite and worth the expense, but before too long Linux will catch up.

Re:Relative to what?

[Anonymous Coward]

Now... who in their right mind would buy Sparc hardware over HP or Dell servers running Red Hat?

I work with telco gear where the cost of an hour's outage can be measured in millions. Much of the newer software for them is written in Java, which is nice and stable when running on a Sun JVM under Sun Solaris on Sun's Sparc/x86 platform. However, the same software suffers from mysterious freezes and crashes when running on Red Hat Linux, and an insanely complicated yearlong four-way fight between Red Hat (OS), Sun (JVM), IBM (hardwave) and our company still hasn't figured out who is to blame -- and has already cost much more than the small premium for going with Sun would originally have.

Well, ya

[zogger]

Umm,. I know Dell and Redhat aren't in the same business. The article was more about Sun and Redhat. Sun sells software, services and hardware. Redhat only sells two of those things so far. That's why I thought it might be interesting if they took a crack at opening up the last leg of that tripod. I *do* know I would be more inclined to get hardware from Redhat knowing it just worked with Linux than getting a similar situation from Dell or Sun. Any place that advertises that they recommend Vista on the top of the "linux pre installed" hardware pages (that would be Dell) in their online catalog isn't really all that serious about it. They offer a few models, but that's about it, sort of a generic minimal effort sop to that market. Whereas if Redhat did it, they would most likely take it *very* seriously. Would it be worth say a 50 buck premium over a similar specced Dell offering (and just about anyone could beat sun on hardware prices)? I think so, especially if they shipped with the real long term supported redhat and not perpetual betaware Fedora. I use Fedora now but it is always a crapshoot every release, it is a fine line between brilliant and a steaming pile, because that is what it is, experimental, it's never stable. If I could get a decent specced machine from Redhat with a solid multi year supported distro on it and it all "just worked" out of the box..that would be quite tempting.

Great Joke

[afabbro]

Given enough time for its open-source strategy to play out, Sun's market capitalization will likely recover and outpace Red Hat's.

Bwwwwwwahahahahahahaha. "Likely" if you are a Jonathan Schwartz sock puppet account. Unlikely if you've followed Sun's dismal performance for any length of time.

Re:Riiiight . . .

[BitZtream]

Then you're blind on many levels, companies like Sun have enough technology that you've never even heard of (because they aren't selling it for various reasons) that they would get picked apart by various organizations taking the pieces they want. Then there is all the known technology they have produced and value in their products. If you think their products don't have any value, you A) Haven't been in the business of dealing with servers very long and also B) are blind, but I already said that.

Layoffs now days are hardly an indicator of economic conditions. You hire 10k employees, keep them around long enough to figure out which ones are worth keeping, and layoff the worst 9,550 that you have, rinse, repeat next year, you keep getting fresh good employees at the cost of paying a bunch of useless ones. But heres the kicker, you were going to pay them anyway, there were only so many good ones to start with, the mass of your employees are generally inept overall and can only do very specific tasks.

Several companies have announced layoffs in the last few days, its great to blame it on the economy rather than saying 'we had this planned all along', don't you think?

No financing? Are you serious? Stop watching the scary news and believing all the hype you hear. People are still getting financing JUST FINE, I recently refinanced my house to get a better rate, and my old rate wasn't bad! The company I work for just recently was going to aquire another business that was going under just for the hardware, and out of no where a venture capital firm jumped in and dumped them a good years worth of capital.

The economy isn't nearly as bad as you think it is, its just a great excuse to tell people rather than the truth. Those layoffs were likely to have happened regardless of the economic conditions, but its WAY easier to say 'we can't afford you' than to say 'your a twit and waste of resources on our planet' don't you think? Sure good people get caught in the collateral damage, but thats just life.

yep

[zogger]

Yes, I think they could do the assembly and shipping in house, at least for desktops and servers, notebooks and netbooks would come just premade, nothing much there to it, they will of necessity all be coming intact from Asia someplace. They just need guys onsite to insure quality and to have that quality warranted with whatever manufacturer they pick.

    The economy in the US is awash in reasonably technical people who need a job now. They need a few people to really make intelligent decisions on the hardware, and to coordinate that with the software devs they already have to get to the "just freeking works, guaranteed" stage. Building computers is just a factory job after all, it's not that hard,and if you studiously avoid the typical east coast and west coast uber and oh so trendy high rent districts and have your assembly plant in the rust belt some place, you get cheap rent and reasonable labor and you can be picky on the labor quality as well. You aren't making any of the components, just assembling them. As to shipping, Fed Ex and UPS go everywhere, that is a non issue really.

As to sending people- joe six pack or joe business- out to do their own research on hardware, still too many horror stories about what allegedly works and doesn't with linux, its a moving target all the time. Check *any* distro's forums there. You basically *have* to follow the Apple-type model to insure the best quality control and to guarantee it "just works" with the software you install.

The whole idea of a big linux company like Redhat selling computers is they *would* be a better over all experience than getting some random production run from acme computer and trying to shoehorn something in. And again, Apple has proven that people will pay a reasonable premium over the lowest common denominator for that experience.

Re:So, how about GPL OpenSolaris now?

[metamatic]

OpenSolaris is already Free Software. I see no reason to believe that such a license change would attract more developers.

The CDDL is incompatible with the GPL, which means that lots of free software code can't be used to improve OpenSolaris.

Sun's missteps over the year

[br00tus]

First - Sun had a UNIX for x86 in 1992-1993 which was superior to Linux at the time. This is not hindsight being 20-20, my manager complained to me in 1997 (we had Solaris x86 dekstops) how Sun was screwing up Solaris x86. Red Hat only got things like a decent kernel crash dump put in recently - Sun really messed this up.

Secondly - too slow to embrace "open source". Red Hat did and now their market cap is about to surpass the company that did not (soon enough anyhow).

Thirdly - how necessary was dumping the Berkeley-like SunOS for the System V-like Solaris? I personally think they put too much of an effort into this, although opinions may vary.

I watched SGI get killed in the mid-1990s. People began doing low-end graphics stuff on Macs or even Windows, and suddenly SGI only became a company for the high-end. It was easy to see that this was the future for Sun. Now Wall Street has collapsed, and the big market Sun had has dried up. And Wall Street has gone from an environment where in 2001 Linux was just a test project, to where some companies are now almost all Linux on the UNIX front, and are looking to dump their "legacy" Sun stuff. It didn't have to be this way.

I first encountered Sun in the late 1980s and until recently I still had a lot of love for them. Red Hat's lack of things like a decent kernel crash dump bugged me. But now Red Hat really does have almost all of the stuff that a critical production server needs. Windows-heavy shops like Suse a lot. I know a lot of UNIX admins and shops that develop for UNIX, including in the traditional financial companies - everywhere the new machines coming in are Linux, and a lot of places are trying to phase their Suns out. I think metaphors of a Sun set are becoming appropriate. Sun screwed up x86 and they screwed up "open source" and now Solaris is going to be relegated to the dustbin that Ultrix and HP-UX are in. If you search for admin jobs on Craigslist, Solaris doesn't even have much of a lead on AIX. With Red Hat now having journaling filing systems, virtualization, decent kernel crash dumps, production Oracle instances that run as well (or better) than on Solaris, high availability and so on and so forth, I can think of very little that Sparc's running Solaris have that a cheaper x86-64 running Red Hat doesn't have.