Slashdot Mirror
Sun's President Dreams of a Linux Future
Sara Chan writes "The Economist has a story analyzing the recent Sun-Microsoft deal. What's especially interesting is the ending. Sun recently promoted Jonathan Schwartz to President and Chief Operating Officer, recognizing the need for radical change if the company is to survive. According to the story, Schwartz's dream is 'to sell deep-discount desktop computers at Wal-Mart, carrying Sun's office applications on top of a Linux operating system'!"
When Mr Ballmer gives Mr McNealy a hug and says that "we do both believe in intellectual property", this is a not-so-veiled jab at the open-source Linux, which both men consider, in essence, communistic. Microsoft and Sun happen to be the only major backers (in the form of licence payments) of Linux's gadfly, a firm called SCO, which is trying to obtain money from Linux users with threats of litigation.
The article also points out that LINUX hurts Sun more than Microsoft:
Linux, however, is hurting Sun far more than Microsoft. Solaris is similar to Linux, which makes it very easy for customers to switch from one to the other. Migrating from Windows to Linux is a much more fiddly process.
I think Microsoft is particularly wiley here. They make nice with Sun knowing that Sun will probably become marginalized as a result of the growth of LINUX and not end up being much of a competitor at all. I am not faulting Microsoft for this, but, you gotta believe that they believe, in their heart-of-hearts (do they have those?) that they will eventually own the whole pie. This sure is fun to watch.....
Happy Trails!
Erick
http://www.busyweather.com/
Whoa!
There was a time when saying you had a Sun meant you weren't just 1337, but respectable, a power user. It may seem a cool thing to be mass marketing Linux boxen from Wally World, but that's a real comedown. Saying you have a Sun would be like saying you have a microwave oven. Is this what it takes to save Sun? Honestly, Linux boxen could easily become commodity hardware. You're not much of a player anymore when you're trying to keep your head above water by selling commodity PCs.
"Hi, my name is Bob and I still felt 1337 with my Walmart-bought Sun."
"Welcome Bob, to Sun-aholics."
A feeling of having made the same mistake before: Deja Foobar
Wait a minute, I didn't know executives could be Slashdot Trolls.
/^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}$/i
Just because it runs Linus doesn't mean the whole product's open source/free/whatever.
It seems like just yesterday (1996) I would have killed for a Sun workstation, but made due with linux. Now I have Linux boxen being used to replace Sun and SGI hardware for image analysis, and my Servers are running MacOS X.
--Tsiangkun
I'll be windows free for 10 years in June
with no direction. One moment they are advocating how big linux and OSS movement is, the next moment a backhand deal with MSFT. I wouldn't trust SUN too much.
Activists United
That's not a dream. Showing the Open Source Desktop as a 'deep discount' alternative is de-grading to the community, as if we are a lower-quality brand. Gnome and KDE both strive to be the best, and should be marketted in this light too. I don't mean expensive, just quality (like Tescos has managed)
puts ("Python r0cks\n");
they're fucked.
It wants its computer business strategy back.
Deep-discount computers S U C K. They *must* know this. A free office suite on top of a free OS isn't going to do anything to sell these things if people can't double click and install software, preferably the software they sell at WAL-MART.
"I bought this here Sun computer, but it won't run these deer huntin' and bass fishin' games I bought with it. I'd like my munny back, please"
Let's set up Linux so it can:
1. Find its fonts without having to edit the XF86Config file 189 times and install some half-working font server for the other three fonts.
2. Upgrade Gnome and KDE applications without having to install yet ANOTHER version of glibc. That or statically link everything and quit pursuing dynamically-linked utopia. I think there's enough disk space now.
3. Have a file manager that isn't linked to every single library on the system, so that if one library is upgraded/replaced, it doesn't make the file manager useless.
4. Make it so these problems can be fixed without changing distributions.
Business isn't willing to pay for products, innovation and careers, so we get brands, mortgage commercials and layoffs.
Anybody announcing a "partnership" with Microsoft gets screwed, hard, in the end. This is really an admission by Sun that they're losing.
Badly.
Watch Sun continue to wither on the vine. Watch it slowly shrink, more each year. They might have a "we'll sell Linux to lusers at Walmart!" strategy, but that's simply absurd.
Selling $199 computers at Walmart is not the road ahead for Sun Microsystems!
IBM has grabbed the Linux ball and run like hell with it, and they've done very well. Sun has pussy footed, flip-flopping more often than a spatula at a pancake shop on Linux.
They have no clear strategy. They have no real, effective, business case for using Linux in their organization. And, unless they come with something, and damn quick, the train will have passed them by.
As a post note, Sun made theirs by grabbing a commodity operating system, putting good hardware underneath it, and selling it for a fair price. Why can't they do that anymore?
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Things don't neccesarily look so rosy for MS either. Think about this, if Linux does totally marginalize Sun (like SCO is now) that means Linux has moved onto the big iron. How does MS move into a market where their OS is hardly supported on the machines required to do the job, especially when the OS is free? MS thinks their getting rid of one foe, only to find in it's place is something much more flexible, modern, and can't be outpriced.
Sun's business model needs to change. By building their own processors, systems, and operating software all at the same time, they are not going to do any of them very well and they will bleed out alot of cash. The only computer company to succeed at this sort of vertical integration - Commodore (they owned the company that made their processor) - succeeded because their product was aimed at one particular market and was extremely affordable. But that was the 80's. Today, there is just too much R&D that needs to go on... Sun is essentially making a profit on a single product when they sell a system while expending the cost of 3 products - a processor, a system, and an OS.
As an interning developer working at the NASA Jet Propulsion Lab, I believe that Sun developing a linux-based strategy will be a great thing.
We use many Sun boxen, along with various flavors of Linux, and it would be tremendous to see more integration. Their work on linux-based Java has already been an enabling factor in our work and I believe that Sun has many good ideas (and good engineers working hard on it).
This annoucement gives me hope that we can continue in our relationship with Sun for future missions, while taking advantage of many of the best features of Linux.
To be fair I should mention we also use Windows and OS X to great effect as well, however good news for Sun is good news for us, especially considering the tremendous quantities of legacy software we have for Solaris!
Cheers,
Justin Wick
Science Activity Planner Developer
Mars Exploration Rovers
It looks more like a ju-jitsu demostration.
More likely the hug by Ballmer was like the kiss of death. He probably whispered something into McNealy's ear like, "Sell Windows workstations and we'll let you live."
A feeling of having made the same mistake before: Deja Foobar
It's called 'keeping up to date':
.ttf's into ~/fonts/ and they were there.
:)
1. Find its fonts without having to edit the XF86Config file 189 times and install some half-working font server for the other three fonts.
Funny, I just dropped some
2. Upgrade Gnome and KDE applications without having to install yet ANOTHER version of glibc. That or statically link everything and quit pursuing dynamically-linked utopia. I think there's enough disk space now.
Windows dynamically links, and includes the dependancies in the download package. Systems like apt-get will get the deps, or if you get the packages from a physical media (I usually get stuff of a mags dvd), then the deps are usually there too.
3. Have a file manager that isn't linked to every single library on the system, so that if one library is upgraded/replaced, it doesn't make the file manager useless.
Opps, KDE and Gnome have very good file-managers with extra plugins for file previews. No problems there either
4. Make it so these problems can be fixed without changing distributions.
Done
Lets so if people can:
1. Stop whining and be more helpful.
puts ("Python r0cks\n");
That's what they tried to do, and it has resulted in them losing money for the last 12 quarters, as low end Linux servers moved up the food chain of what jobs Sun servers had always done.
"The high end" means a totally different thing today than it did 10 years ago. We used to buy $20K Sun machines to use around the network as everything from firewalls to mail servers to DNS servers. Now, all of those jobs are done by cheap Linux boxes.
The speed of cheap systems today is such that "high end" is only a small handful of corporate apps. This is simply not enough revenue to sustain a company the size of Sun.
Microsoft can only buy that which can be bought.
Intellectual Property
Intellectual: of the mind
Property: that over which one has control
Yes, wouldn't that be awful for Microsoft.. They wouldn't be able to extend their monopoly onto the server to take over the complete enterprise.
They would have to subsist on their dominance on the desktop, and they would be stuck at only making $35 Billion per year and a growing cash horde of $50B.
But, that won't really be the case anyway. MS currently has a pretty small share of the enterprise server space. As Sun declines, that opens up a lot of opportunities in that arena. Yes, Linux will win a lot of that business, but MS will get a fair chunk - much more than they have now.
Remember when Compaq bought DEC? Fired all the really good people, let the really good technology (64-bit Alpha) wither and die (not due to lack of innovation, but complete lack of marketing and executive support), and became just another brand of PC-clone?
Then Fiorina gets involved, HP gets sucked in, and bam, another really good technology company gone, now just a PC-clone seller?
Yeah, I have some grudges. I'm not the world's hugest fan of Sun... but I see all the really innovative stuff they've done (even though I'm not a Java nut!) going away. And the computer world will be worse off without it.
The high-end is way more than $20k. I've spend well over $1M on a single, fully-configured Sun machine (one of the original E10Ks, with all 64 processores, lots of RAM, and a massive disk array). I've seen rooms full of those machines.
If you want a single, big UNIX monster, its still basically monster Sun, HP or IBM. Clustering is bringing Linux up there, but I don't see any 64 - 256 processor Linux boxes around (that I know about, anyway).
I don't know if that is due to the Intel platform (I know Linux is portable, but its mostly used on Intel in my experience) or due to locking in ther kernel. I do know the "old kernel" (I started with Linux 0.96c+, when the whole system was on 4 floppy disks, and we didn't even joke about a graphical interface, and Linux was no more popular than 386BSD) was not terribly scalable in a multi-processor setting. I don't know much about the scalability of the newer kernels, honestly.
Sarcasm and hyperbole are the final refuges for weak minds
Sure, market Linux as a cheaper, "generic" alternative to the mainstream OS.
That'll do wonders for the server Linux market, not to mention the general public awareness of Linux.
Oh, and call it "Lindows", so it fits in with the whole industry of substandard equipment with brand names like "Toshipa", "Somy", etc.
Sun has started as a workstation company, so even if they have been very successfull in big accounts, they know that that is not enough to survive.
I think, the Opteron boxes are an good move to get more share in the low end server market.
The interesting part is the way they want so sell their software: from the cooperation with AOL on Netscapes server products (Iplanet) to the current Jave Enterprise System, they still seem to believe in selling software as a commercial, closed source product. Even if they they to license it on a yearly base (and give customers real value, different from Microsoft, which software assurance program mostly anoys customers), they still keep the development process in house.
Even their try to sell Linux for desktops, JDS, is something you have to pay per employee or per seat, although it's mostly based on open source software like mozilla, evolution and gnome.
I don't think that this is doomed from the beginning. They may be successfull, if they can convince customers, that it's not just the software they pay for, but also support, service and updates. This could work, both for companies used to a "classic" way of buying software once and paying extra for support and for companies disappointed by using "unsupported" open source.
But this is the software strategy, which is mostly independant from their formerly very successfull hardware business. And software was only a small part of their business up to now. The hardware part is much bigger (and responsible for most of the service revenue). Even if they have cheap x86 (both Intel and AMD) boxes now, UltraSPARC is still their choice for the big servers, and UltraSPARC is lagging behind more and more in terms of performance, so that even much better RAS features (Reliability, Availability, Serviceability) make it hard to sell those boxes and reason a hefty price tag.
So, even after almost three years with losses, Sun still heads interesting times. :-)
Workstations have not been Sun's core business for about ten years now. Intel based machines were already cutting into that market by the mid 90's (look at what happened to SGI at that time.) They switched to servers, particularly high end ones, and made loads more money. Faster and even cheaper Intel processors, Linux, and clustering plus the IT recession are killing this market, too.
The workstation market is a niche market with high margins, well suited to Apple. It is not a very significant market, though. Sun gave up on it a long time ago.
I know, it's a stretch, but what about this this scenario: Sun merges all Solaris code into the linux code and the GNU/etc tools that are used with it. Then they roll out a new breed of UltraSPARC processors, and contribute code to GNU/etc/Linux so that it interfaces very efficiently with the new processors. Suddenly, the best way to get Linux is to get it on Sun's expensive-ass hardware. Many people stick with their x86 machines at first, but soon when it comes time to upgrade hardware, Linux on Sun looks more tempting than ever.
Yeah, I know, ain't gonna happen... but I guy can dream, right?
I'd always envisaged Sun as a company with a rock solid quality OS, and faultless hardware for mission critical apps with top-notch support.
And then they have to go the Walmart route with a cheap box.
That's almost like Ferrari deciding that it's sick of making luxury cars, and opting to make bicycles instead.
There's a place for cheap linux boxes for those who want their internet/mail/office... but for Sun to do it???
This is going to take some getting used to....
READY.
PRINT ""+-0
THe OS may be free but MS is taking over the server market. They own half of it!
.NET.
.NET client/server apps to probably your active directory configuration, and perhaps be indexing all your incoming email on exchange server. Now if a new database was needed for an IT project which os would come to mind first 5 years from now? Oracle, mysql, or SQL-Server that is fully integrated with everything and supported by VB.NET?
As Windows takes over, Unix is fighting on another front called Linux.
Ever here of divide and conquor? Politicans and the Romans used this strategy quite well.
MS is estatic that Sun is going to go away since Sun is fighting 2 fronts it will not be able to have as much ammo agaisnt Microsoft. They are losing money while ms rakes in more and more.
The problem is since MS owns the desktops they can tie features into Windows2k3 via active directory, SQL server and
After awhile your workplace will have hundreds of MS_SQL-Server databases. They will be running on every copy of Windows(longhorn will use a lite version of it for the new filesystem), and from
MS SQL-Server will be the only one the CIO's would want due to desktop and Windows2k3 server tie-in.
PHB's love Microsoft for that reason. Its not just products but a whole architecture and platform across the enterprise. Java1 or whatever Sun planed with Iplanet and J2EE is too little and too late. They lost.
No wonder Eu is afraid of MS. They are the only ones seeing what they are doing.
The battle agaisnt Linux has only just begun.
http://saveie6.com/
5. Create at least one distribution in which in every single program, "copy" and "paste" are done in exactly the same way with exactly the same results 100% of the time.
6. Create at least one distribution in which every single scroll bar in the entire system looks the same.
7. No one ever has to think about the XF86Config file, ever.
8. There's a clear and obvious way to set and change your monitor resolution that works regardless of whether you know strange things about your monitor, or "scanlines", or the XF86Config file, and NO MATTER WHICH WM AND DESKTOP ENVIRONMENT YOU USE.
9. The way to set up a remote X session is clear and straightforward, and doesn't involve lots of poking at cryptic pages on google and headscratching trying to remember where you have to run Xauth or other such and whether you have forwarding enabled in your ssh_config , etc...
9a. No one ever gets the error message "Invalid MIT-MAGIC-COOKIE-1", for any reason, ever. That's just not descriptive as an error, and it doesn't give you any indication what to do to fix it.
10. If I am on a linux machine, and there's another linux or unix machine somewhere or hopefully even something more exotic (like windows), I can connect to that machine and open up a file browser window displaying the files there and edit them and copy them back and forth, without having to read the Midnight Commander web page, without having to set up cryptic emacs/vi plugins, without having to think about "does this remote computer have ftp, samba, afp, nfs, or some combination thereof?".
11. Make a GUI manpage browser with scrollbars, and hyperlinks, and tables of contents for individual manpages, and the ability to quickly expand/collapse individual sections within the individual manpages, and quickly sorted/filtered browsing of the man -k / apropos database; and put this program where people know it exists and know what it is.
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
Thats the problem.
No one is buying high end systems anymore. Of course customers still need them but what I am saying is as pc server hardware advances, it can do all the things only solaris boxes could and IBM mainframes a decade before that.
Also huge servers do not get upgraded as much as they do their job. If customers are not upgrading, Sun is not making money.
Java did make sun some money at first because it was a powerfull langauge at the time that was cross platform and had alot of libraries. Think of java servlets. MSDN was stealing their market share with proprietary win32 server apps. Java at the time was a great alternative even though its stagnated and it might be killed now thanks to the deal with MS.
But even to run java servlets Windows/Linux server provide a better value today to run them. Again this cuts sun right out.
I think Sun is testing the waters right now on Linux. IBM made money off it and will replace AIX with Linux for their blades and aix servers. I am sure it got Sun's attention.
TI also royally screwed them like Motorolla did to apple. The sparcV should be out already and be outcompeteing all the processors out there besides the power5. But they keep delaying and delaying and yet Sun is still waiting for the sparcIV??
Not only are they expensive but slloooww thanks to this.
If sun can kill all solaris development and use fast AMD Opterons then all teh power too them.
But the market is changing and they must adapt to survive.
http://saveie6.com/
Breakfast served all day!
Given that "Sun's Linux" is currently SuSE, and Sun gave up a previous attempt to create their own distro, I think you're a bit more worried than is really warranted. And who, besides Debian, distributes a completely free as in speech OS anyway? Not SuSE. I don't think RedHat. Who then?
7 November 2006: The day Americans realized corruption and incompetence weren't addressing 11 September 2001
Where have you been for the last 15 years? SPARC has always been an open, licensable processor architecture, which is why Fujitsu makes a competing SPARC implementation. Just because we don't want to give it away for free doesn't mean it's not licensable.
7 November 2006: The day Americans realized corruption and incompetence weren't addressing 11 September 2001
Red Hat does, actually. Everything they write is GPL'd, and they do not include non-free software with their distribution (IIRC, last thing they did that was problematic was Netscape, and that's been gone for years). In fact, one could argue that the inclusion of non-free software in the apt repositories for Debian means that Red Hat / Fedora is actually MORE free than Debian. I don't think that's true, but it's something to consider.
I think Mandrake also GPLs everything, for that matter. SuSE recently GPL'd YaST, too, so actually, they might be totally free, too.
I hope that educates you.
-Erwos
Plausible conjecture should not be misrepresented as proof positive.
You got one thing about that post right: "The battle against Linux has only just begun."
...
.NET is still struggling to gain market share on java, thats part of the reason microsoft did this deal, so they can hedge their bets. If .NET fails then they can fall back onto java and vice-versa.
.NET. MS doesn't hold ANY weight in the enterprise and sun's forays into the low end have been minor disasters
Everything else is crap, microsoft probably runs less than half of the sites on the net (apache runs 70% or the web servers, and I would venture most of those run BSD/UNIX/Linux). Microsoft can bundle the fuck out of whatever they want, it will HURT them in the long run because customers are already becoming weary of their crap with licensing and forced upgrades etc
Only one fortune 100 company uses windows 2k3. (source: netcraft). And MS-SQL is a piece of shit, everyone knows that. If they use it in their filesystem they will kill performance and negate any stability increases they have had in the past 5 years.
The EU went after MS for the same reason the American justice department did, they broke laws. The only difference is the bush administration let them off since they are big business friendly.
Then of course their is this POS DRM built in OS they want to release (whats the ETA now 2007 ?). That won't go over well. Linux has been gaining market share in the desktop arena over the past few years without major vendor support, not that companies like HP, Dell and Sun are backing it, gaining more share is a foregone conclusion, especially at its current price point.
The only market overlap that existed between sun and MS was the development arena. java vs
"Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe." --Albert Einstein
"ditching Sun's computer systems, the equivalent of Ferraris, for cheaper boxes from Dell, Hewlett-Packard or IBM that run Linux, the equivalent of Fiats."
As someone who works in an ISP that is almost entirely Sun I believe the correct analogy would be a Rolls-Royce. Sun boxes, in my experience, are not really that fast for the money, but the quality of them is undeniable. Once you go through the pain of setting them up (Solaris=least fun Unix IMO), they sit there running for a decade. Very nice, but not exactly Ferraris.
Linux on i386, depending on the admin's skill, I would put more along the lines of a nice VW Jetta or Toyota. Stable, quick, cheap, more than enough for most people.
Dream on. I've personally plugged in so many linux boxes in small business, installing them over Small Business Servers charging $2000 per Linux install, and they have all not only been running without incident for years at a time, but all have thanked me and entrusted us for all their desktops. I am talking about law firms with revinues exceeding $11 million, manufacturing companies, and real estate offices. Web file/print, email and backup within domain logins is all it takes. And Gentoo + Samba/CUPS + postfix/courier/spamd + Apache/MySQL/PHP has done it every time. And no reboots or worms either. Software upgrades for free. What a change. I can't tell you how easy a sell it has been. Taking over the server market --- please! The only takeovers I see are the endless variety of worms every month that take over Windows servers.
You, my friend, have given evidence by that statement that you do not have Clue 1
There are a lot of things that " huge, expensive Sun servers" can do that commodity Windows boxes couldn't dream about on the best day they ever had.
disk I/O, multi proc sclability, OS hardening (Trusted Solaris)
I could go on
There is a damn good reason why Sun boxes are still deployed, and will continue to be deployed, in critical environments.
They just work. All the time.
And I for one thank the Powers That Be that *my* bank is smart enough to realize this.
Time flies like an arrow, fruit flies like a banana.
FreeBSD?
"MS-SQL is a piece of shit, everyone knows that. If they use it in their filesystem they will kill performance and negate any stability increases they have had in the past 5 years."
I don't know that. I used to know that, until I spend some time working with MS-SQL2k, Oracle 8, 9, and 10, and PostgreSQL 7.3 and 7.4. I've done installation, admin, and same-hardware performance benchmarking on all of those platforms now, from a standing start in each case (I had a lot of networking and Unix experience, but no real DBA experience).
MS-SQL took a couple of days to install, patch and test, returning the best numbers of the entire set. PostgreSQL installed quickly, but it took a couple of weeks to learn how to tune it. After that two weeks of hard work, it was just as fast as MS-SQL in controlled conditions. However, it still has weird problems: sometimes it will refuse to use indexes on tables that have grown rapidly, and some nested condition queries can be created which completely choke its optimizer. One in particular took two and half minutes on MS, but was still looping after 14 hours on PG 7.4 when I gave up and killed the query.
All versions of Oracle took days to install, and I found tuning information to be very difficult to find and comprehend via free or paid-for resources (Google, O'Reilly and OTN in that order). 8i was unable to even complete my performance tests without dying due to fragmentation problems. 9i and 10g were able to complete the tests, but at half the speed of MS or PG. Perhaps if we'd hired a consultant they'd have been able to get better numbers, but no one was willing to pay to find out when we had two perfectly good platforms which cost much less.
Take a wild flier at which one of those three "supported platforms" gets recommended to customers who ask what to run the product on...
"Nothing was broken, and it's been fixed." -- Jon Carroll
Just yesterday a coworker and myself were trying to figure out how many desktops Sun has had, or is proposing. I use the term "desktop" loosely.
g Glass
I came up with:
DPS
NeWS
OpenWindows
CDE
Gnome
JDS
Lookin
But since my friend was an ex-Sun employee who worked on NeWS, he came up with a few more that I never heard of.
Don't blame me, I didn't vote for either of them!
All versions of Oracle took days to install, and I found tuning information to be very difficult to find and comprehend via free or paid-for resources
Eh? I have just installed Oracle 10g on a Linux box. Took 3 hours from start to finish. Detailed documentation about how to do this was available on-line at Oracle.
9i and 10g were able to complete the tests, but at half the speed of MS or PG. Perhaps if we'd hired a consultant they'd have been able to get better numbers, but no one was willing to pay to find out when we had two perfectly good platforms which cost much less.
Bizarre. After the 3-hour install, Oracle was up and running and giving at least a five-fold performance boost over Postgresql, with no fiddling or tuning.
I hired a DBA to do some similar tests and came to similar conclusions.
* MSSQL exceeded expectations, and if it only ran on a server we could easily ssh to, it would have been the best choice
* PostgreSQL good once the (at least well documented on the net) black magic of tuning shared memory, sort memory, 'free space' memory, and vacuum stuff was figured out
* Oracle - pain to set up (installation failed if you followed their docs to the letter - but DBA new the tricks) - Was dog-slow until tuned - but evntually after tuning slightly outperformed the others.
* MySQL's SQL syntax was a bit too nonstandard for us to port the test to.
Once tuned, they all performed similarly (not surprising, since they all can do merge joins, nested loops, etc; and they all can use as much memory as you tell them to).
But the surprise to me was the MSSQL was friendliest "out of the box".
MySQL is the Open source DB of choice for most for a reason. try it.
... and this is where the difference between a real database and some crufty piece of shit like ms-sql or ms-access comes in. A real DB will run much more effectivly on larger hardware that a crufty piece of shit. in other words: the performance increase once you get onto higher end machines is not equal, mySQL, postgreSQL and especially DB2 and Oracle experience massive gains in performance when compared to any MS database.
.... a large part of that is the platform it runs on IMHO.
Secondly although you can install Oracle on intel hardware it was not (and shouldnt be) desiegned for intel hardware
Not to be an ass (I am no DBA) but I have seen very large gains 15-20% in overall speed when PG or My are properly tweaked by a DBA with experience. I have never seen someone get comparable performance from MsSQL
"Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe." --Albert Einstein
Sun hasn't said "We're dropping Solaris" but embracing Linux without becoming a player in the Linux kernel team is a HUGE mistake.
Solaris does some things much better than Linux -- less and less, certainly, but, for example, Solaris does partitioning of machines, the IP stack is great, and Solaris boxes can be configured to run complicated apps with higher uptimes even than Linux -- it's close but Solaris still has a small edge in reliability.
So Sun embraces Linux, further marginalizes Solaris, and soon Solaris will only run on Sun's Big Iron -- E10K's and the like.
IBM will make Linux scream on their Big Iron, and some of us (more and more of us) will pick IBM's Iron over Sun's because it's the same across the board.
Sun really has two options. 1. Embrace Linux and be part of the process, cannibalizing Solaris for Linux's sake and becoming a major Linux player -- with the E10K running just a feature-rich on Linux as Solaris. 2. Push Solaris hard. Give it away for the small boxes, get it on the desktop, run Linux apps on it (they've already got a project to allow this), and keep a culture that's 100% Sun, stressing in their sales pitch the few, but legitimate ways where Linux is a liability on Big Iron.
Option 3, undermine Solaris, and remain apart from the Linux community, seems to be the chosen path, however. It's the same path SGI went down. You remember SGI, don't you? You know, the guys with the pretty colored plastic? Think back...