Erstwhile bitter rivals Dell and Sun Microsystems are set to announce that Sun’s Solaris and OpenSolaris operating systems will be supported in all of Dell’s servers. Dell founder and CEO Michael Dell and Sun Microsystems CEO Jonathan Schwartz plan to make the announcement during a joint appearance at the Oracle OpenWorld 2007 conference today.
I’d go for a Sun hardware and have Solaris instead nowadays. I don’t know how much impact this can have both in the software side (Windows licenses getting out) and in the hardware one (Dell purchased to run OpenSolaris instead of Sun to run Solaris).
Is Sun planning to abandon the hardware battlefield ? If it was the case, this could be a great thing: Dell offering a Solaris (well, OpenSolaris currently) server and a Linux desktop 🙂 This will of course mean also the end of Solaris in favor of OpenSolaris … as well as the end of Windows in favor of Linux ??? This last is just a wish, but who knows.
No, Sun has been expanding their OEM program for the last couple of years. The sparc hardware isn’t going anywhere, this is just another option for those considering Solaris.
You mean considering OpenSolaris, don’ you ? They are not the same right now to my knowledge.
I imagine he means both. The article says that Dell is offering both Solaris and OpenSolaris.
what do you mean sparc i going nowhere?! Have you used a T1? They are incredibly powerful! Or do you mean sparc is going nowhere in the direction you happen to use it (desktop x86 based systems?)
I hope sparc and the power chips keep getting developed. In my eyes there is no point fighting out of a software lockin to just to enter a hardware one…
“what do you mean sparc i going nowhere?! Have you used a T1? They are incredibly powerful! Or do you mean sparc is going nowhere in the direction you happen to use it (desktop x86 based systems?) ”
I think he means that SPARC is not being phased out, that it is going to still be around, and that you can still purchase it from Sun.
isn’t going anywhere he means that they are here to stay.
“I think he means that SPARC is not being phased out, that it is going to still be around, and that you can still purchase it from Sun.”
Why just sun? I prefer the Fuji SPARC64 personally, but that T1, a beast!
With going nowhere he means that they are here to stay.
gotcha, if that is indeed what he meant I apologise for any confusion 🙂
Sun faces some serious challenges on the SPARC side of its business, especially in terms of performance per thread. I don’t think it’s feasible in the long-run for Sun to compete with AMD and Intel for serialized workloads. Sun’s greatest hardware strength from its up-market pedigree is in system architecture, whereas it never moved the volume required to excel in the modern processor market.
I agree that Sun looks very strong at the moment in thread-dense boxes for transaction computing. However, both Intel and AMD have mini-cores coming in 2009-10 that will scale down to ultra-mobile and tile up to massively-multithreaded rack servers that will compete with Sun’s Niagara/Rock descendants. On the other hand, I could definitely see Sun as the leading OEM for these Intel/AMD-based thread monsters.
I don’t understand. There are very few operating systems and applications that don’t run on x86, and plenty of them run on SPARC and POWER as well. There’s no lock-in. The proprietary software vendors that ignore SPARC and POWER do so because the niche markets aren’t worth their time and resources, not because of any especially onerous barriers to cross-architecture compatibility.
Regarding SPARC and POWER; it depends on what you mean by success or failure? people could look at PPC and claim, because Bob isn’t running it on his desktop or that it doesn’t dominate small to medium business server sales.
Niagara/Rock have an advantage over x86; SPARC64 IV for example is very competitive with the high end x86 chips put out by Intel and AMD; if you were talking about UltraSPARC, you would have a good case, but the SPARC64 is a fpu/int monster who has no problem holding its own – its just too bad it took this long for Sun to finally move to a superior SPARC implementation given that in the limit benchmarks of SPARC64 such as TPC, it comes out close to the top – especially unclustered.
The article says that Dell will be offering both Solaris and OpenSolaris. Now the Big Four (IBM, HP, Sun and Dell) all offer at least Windows, Solaris, RHEL, and SLES on their x86 servers. This move doesn’t seem nearly as strange when you consider that Dell, the only member of the Big Four with no in-house UNIX flagship, was previously the odd man out in terms of Solaris support.
Nephelim… have you *ever* managed sun servers versus the competition like IBM eseries, HP DL3xx, or Dell servers?
Sun’s ILOMs (Integrated Lights Out Management) beats anything on the market hands down. Is the server dead? So what, ssh into the ILOM, restart the server and then connect to the Serial console… done.
Sun servers are more efficient, better designed, and easier to manage than equivalent Dell servers. How do I know? Because we manage thousands of them and have replaced all IBM servers in production with Sun x4xx series. Nothing compares to Sun servers for remote management.
If anyone disagrees, please explain.
SEJeff, my comments talked about a distant future (at least, medium term one), not any short term issue. I’ve managed some Sun servers, though I suppose than not as many as you do. And yes, I like them a lot, and yes, I agree with you in their quality and integration. May be that my poor English skills make it more difficult for me to express the things I really want to state … sorry about that, I suppose it’ll improve with the time.
Oh ok sorry about that. It just pisses me off when people talk about things they don’t actually know about.
Sometimes I come off as harse. It has to do with being very opinionated.
You are right.
Yeah ILO on the HP Proliant beats that. Not only can I do SSH but I can web into the box ad bring up a full Graphical Console. I can remote mount media from my workstation across the country and boot the dead powered off box with that media with no touch what so ever. The latest inegrity boxes also have ILO functionality. This is all integrated into the motherboards now. I know many customer that have zero touch data centers. Even new installs get done by HP they plug the ILO in and the customer remote build everything from there.
Blades become even easier because I can configure a ton of stuff for all the blades in the enclosure so adding blades is literaly just plug and play.
Sun ILOMS have the exact same thing… From the webgui in the UK, I can mount an iso image of a disk on my local workstation (in the US) and then use that to install some crazy OS that doesn’t support automation such as kickstart.
Blades are fragile and break too much. Leave a set of blades running with 2+ years uptime. Then reboot them. About 1 out of 10 won’t come back up. Blades have their purpose, but it isn’t anything I’ve worked on yet.
My last job (a delta owned Airline, Comair)… we almost exclusively used HP Proliant DL360 (G1-G4) and DL585 servers for Intel based hardware. They aren’t bad, but the Sun ILOM is still better.
pretty much how i see it – the main reason i have for using solaris is the sun hardware, not really the os.
i really can’t see the point of solaris on x86, linux is better at that.
I’d go for a Sun hardware and have Solaris instead nowadays
Yep, now look from this point of view. Nearest Sun service I could have is in next country. Could you imagine mission critical server like that?
The only way Sun could expand here is if people adopt Solaris to the measure where Sun hardware providers would start popping in our country. But as long as demand doesn’t exists both Sun products (OS and hardware) are not worth bother.
What do you need for server? Certified hardware and hardware support. Sun obviously lacks later here. Which they can expand with Dell.
Dell? We’ve got 3 god resellers with good (as far as Dell service goes) service. This could probably be the first sane choice to be worth to apply Solaris to mission critical services here.
Now play a little predicting (or guessing) game. Solaris sells with Dell fine. People start becoming interested about real Sun iron. There now is demand. Time to pop some Sun resellers here….
Is probably how most people will see it rather than a win for Open Source!
though you got modded down, i think there is some truth to your subject line. Most people that are into open source software are really just fans of certain open source projects, and prefer open source for the rest of their stuff. Every project has it’s “us versus the world” users, and the reason you come across a lot of linux zealots like this is largely because of it’s large user/developer base. Most people will see this as a win for open source, despite what some vocal fanatics say.
I’m not saying their aren’t people that are just purely into open source, and only use open source. It’s just that a majority of the open source world is not made up of these people
The only pragmatic argument against OpenSolaris from a Linux perspective is license fragmentation. Otherwise it’s great. However, this is a big problem, and it will become more annoying for both communities, not to mention ISV/IHVs, as time goes on.
To use a political analogy… Imagine a young democracy devoid of major political parties where politicians advance their own platforms independently. Then, someone comes up with the bright idea of forming a broad alliance to advance and protect certain ideals, convinced that, united, their coalition will one day dominate the national politics.
Nobody spoke up and asked, “What if, at some point in the future, somebody starts a second party that’s mostly like our’s, but which differs on some contentious wedge issues? Wouldn’t that be bad for everybody?”
So now we have a two-party system for free software platforms (along with some independent libertarians that advocate the BSD), and each party has a binding pact not to ever cooperate with the other. It’s a sad state of affairs, and pointing fingers isn’t going to help. Both sides are partially at fault, the general principle of copyleft is at fault, and a copyright system that wasn’t designed for organic, collaborative development is especially at fault.
We need a compromise. How about this: Let’s agree to relicense both Linux (painstaking as it would be) and OpenSolaris under the LGPLv3. The projects would also have to agree on whether or not to eliminate the anti-Tivoization language. Linux would be strictly LGPLv3-only, and while all of OpenSolaris to which Sun owns the copyright would be LGPLv3, third parties would have the option of linking arbitrarily-licensed source files.
To enforce this policy, the Linux kernel project would designate a very small portion of the kernel that’s vital to Linux but highly unlikely to be useful to other projects as a copyleft “seed”. They would license this seed under the GPLv3 with an explicit LGPLv3 linking exception. Then nobody would be able to distribute a Linux kernel linked with non-LGPLv3 code.
This empowers bidirectional code sharing between the projects while satisfying much of their respective licensing requirements. From Sun’s perspective, the LGPLv3 is very similar to the CDDL. Linux could preserve the incentive for third parties to contribute free software, and while mixed-source projects could merge LGPLv3 code from Linux, file-granular reciprocity would still be required.
Just an idea… I’m glad that OpenSolaris brings more competition and visibility to free software, but the whole idea behind free software is that competition and collaboration aren’t mutually exclusive. We can share to exploit our common interests without compromising our mutual ability to differentiate. Linux and OpenSolaris have different visions, and the ability to share code can’t change that.
We need a compromise. How about this: Let’s agree to relicense both Linux (painstaking as it would be) and OpenSolaris under the LGPLv3.
Well I would vote for it, but emotions run so high in this area that you almost qualify as a troll for suggesting it.
Honestly, if that was true, why isn’t everything BSD/public domain? The whole point of open source is that you are not a slave to the software maker’s will. You have the source code, you can fix bugs, add features, and fork if necessary. I mean, the GPL isn’t exactly friendly when it comes to competition and collaboration with other licenses such as BSD style licenses.
Your political analogy is flawed. Open source isn’t a democracy. What makes more sense is that there are three or four rouge countries run by dictators. These countries have similar ideas, and would all benefit from trade agreements with each other. Two of these rogue nations have clauses in their trade agreements that say if you use there products in anything, you must give all the information as to how to make it back to them, so they can trade it more. The beauty of this is that they will give you their products for free. The down side is that you have to give your products away for free. These are the nations of CDDL and GPL. Another nation is BSD, they just give you stuff, tell you to give credit where credit is due, and ask you kindly to give back.
the nations of CDDL and GPL are basically the same. The difference between them is that CDDL, do to past relations, a desire to keep it’s own culture/products, and the want to be able to give code back to it’s sister nation (solaris), does not want to merge with GPL to make a bigger nation. Both sides feel that they should merge, but they can’t agree on merging into CDDL or GPL. Mean while BSD thinks that they are both lame because their trade agreements make it impossible for them to ever get anything useful back from these nations.
In the end, these nations would be much more powerful if the combined into one nation, but in doing so it would mean that ideals would have to be compromised. This necessary corruption is seen as distasteful on all sides, because they already know they are right and the others are wrong. In the end they all settle on an alliance to try to fend off the attacks and trade blockades from the rest of the world.
What makes you think that GPL is better than CDDL?. Isn’t it a bit like America thinking that Democracy is the best for all nations?. Think about it – the Solaris community DOESN’T WANT GPL!
no, i’m just saying that they have similar goals. What i’m trying to convey is something like during the cold war, china and russia had similar goals/beliefs, so why didn’t they become one nation and become more powerful? Just because something is similar to something else doesn’t mean it is a good idea, or possible, for them to become one entity.
edit: i picked china and russia because of their easily identified similarities in terms of goals/ideals, not as some sort of social commentary
Edited 2007-11-15 21:57
I think most Linux supporters see almost success that is non-msft as a win for open source.
Msft is percieved as the monopolistic linux antangonist. At least for now.
That was exactly my thoughts. The past 18 months I’ve been recommending Solaris, rather than Linux for any server based operations to IT managers that I know. And for a variety of reasons – Linux is good, but Solaris is just as good, and better in many areas imho. Also, the cost of supporting Solaris is cheaper than going with an enterprise version of Linux.
Personally – I think Linux is half dead now – it’s promised so much, and not really delivered. True, part of that is due to the political machinations of Microsoft and others, who oppose the open source way of doing things (unless it’s BSD based code of course).
Now, before all your Linux fanatics mod me down, thnk about it:
Compared to 5 years ago, there are even less software ported to Linux.
Compared to 5 years ago, WINE is no better.
Compared to 5 years ago, 3rd party driver support for hardware is no better.
Compared to 5 years ago, OpenOffice has in all honesty, went backwards. Microsoft Office is EVEN more dominant now, than it was 5 years ago.
These are just my personal observations.
From a political point of view, we now have companies sidestepping the GPL, so that they can bastardise GPL code and get away with it, and not live with the spirit of the GPL. This is becoming more and more common. We have ex Windows users who don’t care about the principles of Free Software, and the GPL licence, all they care about is not having to pay any money for the software. There’s more to open source than free, as in money, software. All of this is driving to the commercialisation of Linux, with the big corporations being interested in it from a server side point of view. This leads to the ignorance of the majority of the users – the desktop users.
For Linux to succeed, imho, it needs to do several things:
1. Set up a body of corporations who donate money, for open source developers to work with 3rd party software developers, and at THEIR cost, port the software to Linux with no release of source code etc. If enough 3rd party applications get ported, people will shift across.
2. Narrow it down to one package manager.
3. Narrow it down to one desktop environment.
4. Get games ported to Linux (see point 1).
5. Get much better 3rd party hardware driver support, rather than ugly, open source hacks that aren’t reliable, or only partly work (or both).
Points 3 and 4 are critical – with limited open source developers, it makes NO sense to waste their efforts by working on splintered efforts and many different projects that have the same end goals. Better to work on one project, and combine the developers.
Just my honest thoughts. Don’t get me wrong – I like Linux, I like it a lot. It just has lost its appeal to me, it no longer suits my needs, and has to really improve in order to entice me back.
Dave
I don’t agree with everything you said, but you have made some good points. I don’t know why people here see it as their ‘job’ to suppress peoples opinions via the moderation system. I’ve done by attempt to even things up and added a point to your post.
That’s the problem with modding systems – you can make sane points, but because you don’t ‘follow’ the rest, you get modded down as a form of censorship.
Thanks for the mod point, at least some of us have some decency.
Dave
there are even less software ported to Linux……WINE is no better……3rd party driver support for hardware is no better…….OpenOffice has in all honesty, went backwards.
You do realise that this is just stupid, don’t you? Open Office is developed mostly by Sun and is the same on Solaris as it is on Linux, as is WINE, and driver support is far worse in Solaris than Linux – third-party or not.
Edited 2007-11-15 14:33
I think its more that it has been the “Year of the Linux Desktop” for the last decade or so. Linux has been hyped like nothing else, and what i think he is trying to say is that it has pretty much hit the peak of what it will become, if not started to decline.
Linux has been hyped like nothing else, and what i think he is trying to say is that it has pretty much hit the peak of what it will become, if not started to decline.
What he’s listed there has nothing whatsoever to do with Linux, or Solaris. If he’s arguing that Linux ‘isn’t doing it’, then I’m afraid Solaris isn’t either.
At least some have understood the gist of my argument.
Dave
OpenOffice is one part of the equation. And yes, Driver support is ‘worse’ for Solaris than Linux. However – you have to look at it from Dell’s point of view – they’re selling Solaris on Dell computers that will be used as servers. Driver needs for servers have always been more limited than a desktop system.
From a normal desktop point of view, Linux is better than Solaris. The problem is that Linux’s big inroads into the market have been via servers and corporate environments. It has made very little, if no inroads into the desktop market. If Linux can’t make it in the server environment, it doesn’t hold much help for the desktop market imho.
Linux has been the ‘best thing since sliced bread’ for a long, long while now, and in reality, this hype has not eventuated into real market share. True, not all of this is due to Linux itself, a fair share is due to anti competitive behaviour from Microsoft, and with the collusion of other large software vendors as well. Yes, I say collusion. There’s no real benefit for large software vendors to port to Linux – it means another port to support, which costs a lot of money to maintain. They already make nice profits with their Windows based software, why you those profits to subsidise a Linux version?
The average person doesn’t give a rats ass about Linux, sure, more people have heard of it now, but it still has a reputation for being a bitch to install, a bitch to maintain, and more importantly, their software doesn’t work on it, or work well in many cases. This is a huge turn off for most ordinary computer users, and their not going to be prepared to switch to Linux because of it.
Whilst I don’t have any hard figures to back me up, I think you’ll find that OS X users have avalanched in the past 18 months, and that has taken a lot of potentials away from Linux. Your average person is prepared to pay their hard cash for software, if they feel the software is worth it. That removes the one real advantage (to most people) that Linux has – free as in cost. Your average person couldn’t give a rats ass about ‘freedoms’, they look at that as a idealistic rainbow dream which is unattainable in todays age.
Dave
OpenOffice is one part of the equation.
Open Office is what Solaris uses, so trying to say that Solaris will do any better is a flawed argument.
However – you have to look at it from Dell’s point of view – they’re selling Solaris on Dell computers
They will get an option of installing Solaris, and then being palmed off to Sun for support. Like such previous deals, I’m not sure why people are going to specify Solaris to be installed unless they have some legacy need for some specific software. Most, universities in particular, left Solaris behind years ago because of poor open source software support, like Python for example, so it’s funny that you mention application support elsewhere.
The problem is that Linux’s big inroads into the market have been via servers and corporate environments. It has made very little, if no inroads into the desktop market.
So what is Solaris doing differently to be any better, and why?
This is a huge turn off for most ordinary computer users, and their not going to be prepared to switch to Linux because of it.
So what will make them switch to Solaris, which was your original point?
Whilst I don’t have any hard figures to back me up, I think you’ll find that OS X users have avalanched in the past 18 months, and that has taken a lot of potentials away from Linux.
OS X, mostly due to limited hardware supply, is pretty insignificant really. Also, how many Linux desktop users do you think there might be out there? :-
http://members.forbes.com/global/2007/1112/024a.html
“Kirk Yang, who heads Asia technology hardware research for Citigroup in Hong Kong, predicts that the company will sell at least 3 million Eee PCs next year but could easily tally 6 million. By comparison, Apple has sold 4.3 million laptops in the last four quarters.”
Your average person is prepared to pay their hard cash for software, if they feel the software is worth it. That removes the one real advantage (to most people) that Linux has – free as in cost.
No actually. Why do you think that Microsoft is big on pre-installation and keeping OEMs on the broken end of a bottle rather than itemising Windows as a separate cost? Because they want to make people believe that Windows is free and comes with the PC.
Free software bypasses that OEM channel completely, which is what terrifies Microsoft so much.
“””
“””
I’m glad you added that last about it being personal opinion. At least you admit you can’t point to facts to support those assertions.
1. I doubt that the amount of software ported to Linux from some other OS is less now. Corel dropped out after MS invested in them and the CEO left. But that’s the only significant dropper-out that comes to mind. But there is surely less need for ported, closed source packages today than 5 years ago, since more OSS packages are available to fill more niches.
2. I would disagree with this. In my business, which is supporting Linux servers and business desktops, I do use Wine in targeted areas. And Wine, while it has never blazed very rapidly down the trail, has indeed gotten better at the tasks I require of it. Though, thankfully, my need for it today is much lower than what it was 5 years ago. Today, as long as it runs IE6 well, I’m set.
3. Compared to 5 years ago, I find the driver support to be *far* better. Especially for that problematic area of *printers*. USB has gone a long way to help the situation, and more vendors are Linux aware. Of course, wireless network chipsets are the current bugaboo; But the stage looks set for that situation to improve. We, the Linux community, really have to take some of the blame for that, since our infrastructure did not make things as easy for hardware vendors or OSS driver writers as it could have.
4. I’m unclear if you are addressing quality or market dominance regarding OpenOffice going backwards. But I’ve had many users using it since 1.0.x, and I don’t see how anyone could argue that it has not gotten faster, more capable, and better at importing MS formats in the last 5 years. And though OOXML is not exactly an *optimal* standard. It *is* better for 3rd parties like OO than the previous MS formats. So I only see things getting better, there.
These are, of course, my observations. But they are informed ones. Selling and supporting Linux, and in particular Linux desktops, has not always been an easy thing in this MS dominated world. So I am faced with these issues every day. And it is exceedingly obvious to me that I am having a much easier time of it today than 5 years ago.
At any rate, since *Solaris has even less supported closed source software, needs Wine for the same things Linux does, is *way* behind on driver support, and uses OO in the same ways as Linux does… I’m mystified as to why you would find it preferable for the reasons you give here. Which is not to say that it doesn’t shine over Linux in other ways; I’m not knocking *Solaris. I have always commended Sun and the *Solaris community om their OSS work, and welcomed them.
While I *do* believe that a bit of consolidation might be beneficial to Linux in certain areas, I would disagree with your goal of narrowing everything down to one player. I feel that you are underestimating the *crucial* importance of ongoing competition. No player achieves top performance without someone or something to compete against. And that is before you even get to the arguments involving cross-pollination, and Linus’ concept of massive parallel development and software natural selection.
That last might be controversial, but you don’t have to accept that bit to accept that competition is vitally important.
Edited 2007-11-15 19:01
Good counter-arguments as usual 🙂
The problem is, facts are hard to find. Statistics can be easily manipulated, as we both know. I’m only going on my own personal experiences with Linux, and with people that I know etc. I don’t know anyone who has moved to Linux in the past 3 years. I do know a few that have been using Linux for 5+ years, but they are what I’d consider social outcasts, using Linux to voice their social displeasure with the world at large. I know of many people who have switched to Macs and OS X.
The gist of my argument was that if Linux can’t make it in a server environment, then it’s screwed on the desktop imho. Windows Server 2003 has done serious damage to the Linux market, and for good reason – it’s reasonably priced, performs well, is reliable and pretty secure. The free cost of Linux is killed by the cost of having a support agreement with one of the big vendors, and Sun beats Linux from this point of view when compared to the 2 big players in the market. I’m sorry, but very few corporations will run Debian on a server – I’m not knocking Debian, it’s a superb server operating system, but management WANT support, paid support. Having employers who are experts in the area is not the same as paid support from my experience.
1. Can you list me well know software products that have ported to Linux? A few games, that’s about it. Please don’t count software that originated from the Linux environment, like Mozilla FireFox and Thunderbird, or OpenOffice. Let’s talk Windows based applications that have been ported across.
2. Wine isn’t bad, don’t get me wrong, but it’s still very dodgy in use. Let’s take into account a major software application – Adobe Photoshop CS2. Up until the very most recent point release of WINE, it didn’t work. Capture one Pro doesn’t work. Neither does a host of Canon based software for their digital SLRs. You might argue that that is only a small percentage of the userbase, and that’d be a reasonable argument, but let’s consider that digital photography has really taken off in the past five years. Sure, open source has native applications, but in all honesty, they are pale compared to the native versions for Windows. Very pale.
3. Driver support – a difficult area, some areas, Linux is pretty good, like with printers as you pointed out. But, for a lot of stuff, Linux drivers are still MIA. Let’s take my Logitech Wingman II steering wheel – 7 years old now, no Linux support.
4. OpenOffice has gotten a bit faster, but it’s still a massive disappointment. Sure, it’s good with writer, but that’s about it. Excel support is still what I’d consider dodgy, Powerpoint support is ajoke, and Access, nada. There’s more to an Office suite than just a word processor.
Just my honest viewpoints.
Dave
The gist of my argument was that if Linux can’t make it in a server environment…
What makes you believe this?
Because it seems that sales of Linux from a server aspect are on the decline, and by a fair amount. IIs is catching Apache at a very fast rate. Windows server based operating systems are starting to get much better uptimes. Server 2003 has been very stable, reliable and secure for Microsoft. For the vast array of IT shops out there in the corporate environment, most of them have switched from Windows 2000 server to Win2k3 server and have been delighted with it. 3 out of the 4 employers that I’ve worked for in the past 7 years have all been Windows shops, with NO thought of moving to Linux at all – period. They are:
Toshiba
Esselte
Newell Rubbermaid
The 4th one was Apple Computers. If I take it a bit further back, the previous 2 employers to this were Swiss Air and Centrelink (a government welfare agency in Australia). Both are Windows based. There was some talk about Centrelink migrating to Linux, but that was delivered a pretty hefty death knell.
Telstra, our largest IT&T company in Australia, was going to switch to Linux, although it was ONLY a ploy to get cheaper prices from Microsoft. They never had any REAL intention of switching. Many, many, many companies are doing this – using Linux solely as a bargaining chip to get cheaper prices from Microsoft, and it’s working. Microsoft’s markups are high enough for them to drop their prices and offer good bargains to their long term customers. I don’t see this happening any time soon.
As a server operating system, I personally feel that Linux is far more powerful, and in the hands of a competent Linux trained IT technician, very stable, reilable and secure. Linux trained guys are more common these days, but they’re still a drop in the pond when compared to Windows trained guys. Until you see a serious change in the number of people being trained in each platform, I doubt you’ll see a serious change in the number of server deployments.
Linux will probably become more common in the embedded market, where it’s doing well, and probably in the high end super computer arena. It’ll probably hold its own for the next 5-10 years in the server environment, with slight to medium declines if my suspicions and reading of the current market are correct. The desktop side of things will see little inroads, UNLESS 3rd party software is ported to the Linux platform, and not just any particular software, it has to be key applications. And even then, Microsoft’s real monopoly is with MS Office, which nearly every single business uses. Excel and Access are really heavily tied to the Windows platform, and that makes it almost impossible to create Office suite equivalents on other platforms like Linux or BSD and have FULL compatibility. This is a killer area that goes against Linux imho – unless OpenOffice can do the custom macros that Excel can do, and do them accurately and well, people will not migrate. I know that this isn’t really OpenOffice’s fault, but your average user doesn’t give 2 hoots about who’s at fault – at the end of the day they want a working solution that suits their needs.
Dave
Thanks for threadjacking a discussion about opensolaris into a thread about linux. nice troll
the more major enterprise pc vendors sopport solaris the better. SUN’s future looks bright (no pun intended). good work guys
Just kidding
Cool to see that Sun seems to strike quite a few deals lately (IBM to, among others, if I recall correctly) and are going forward. I guess Schwartz is a major improvement over McNealy for them.
I hope this Linuxlikedistributionstuff works out for them aswell.
That is the one thing I do like about Schwartz – he is pragmatic; he isn’t hugging religiously to this duelistic idea of the IT world, instead his focus is “how can we maximise both long term and short term shareholder value”.
Yes, I do have some issues with the direction in some areas of Sun – I’m sure thats the same for every person out there, no one will be 100% agreement with management, but over all, they are stepping in the right direction. This Dell agreement is the first step.
They’ve now got their server side all tied up with big OEM vendors on the side of Solaris, now it’ll be a matter of getting more software vendors and hardware vendors. From that hopefully Indiana maybe in the future will be an option on Dell laptops next to Ubuntu (maybe that is what they mean by OpenSolaris) and thus the choice for consumers will increase.
With that being said, on the desktop they need more application, especially in the traditional markets which SPARC used to dominate; Matlab and Maple are poplar and available for Solaris SPARC but not Solaris x86, for example.
What they are hoping is name recognition.
The more people know about Solaris and Java the more they think they’ll buy stuff directly from sun and sun hardware.
That’s why they changed their stock ticker symbol to JAVA. They can boast and say how they invented java and opened solaris, etc etc, and how awsome they are… while IBM and HP’s OS’s are failing (you remember that IBM website that urged people to migrate to linux until that SCO suite happend?) Why buy IBM’s java software when you can just get it from the people who invented it.
I guess they hope for a huge major product like MS windows so they can branch out and sell things that can fit’s customers needs like MS has done.
This is pretty meaningless. Dell aren’t pre-installing Solaris in any form. It just means that if they ring up Dell, they’ll get palmed off to Sun for support.
As far as I’m concerned, Solaris has been on the wane for years. Many universities, who’ve long been big Sun hardware and Solaris users (one I knew was absolutely die-hard Sun), switched to Linux long ago because of better open source application support and cheaper hardware.
Edited 2007-11-15 10:12
Although I’m a Sun fan, I have to agree with your assessment. Who hasn’t heard of Sun Sites? Then again, I can’t remember how long has it been since I connected to a Sun Site FTP server.
Ever since Sun lost their UNIX workstation market to that Linux ‘upstart’, they have been focusing on server site solutions. Yet, all that money they spent on University handouts gave only a short term return on investment. Like you say, Universities prefer the cheaper hardware option coupled with open source software. Sure, Solaris is now open source but it’s a late comer compared to the BSDs and Linux.
All in all, Sun still dominates the corporate UNIX server and Java application server markets but as a workstation system, it’s been a long time since anyone got really interested in Sun’s offerings.
Hopefully OpenSolaris and project Indiana will help turn this around as it would be a shame to see this exceptional OS relegated to just the server market.
Although I’ll be flamed to a crisp, the problem with Sun is this; they want to achieve something without doing anything, without investing any money, resources or people power – the net result, everything that comes close to improving the desktop experience falls to pieces.
Take Blastwave for example, its a great service, heaps of people use it, and yet, when the maintainer who owns it, was struggling to pay the bills – where was Sun? Why didn’t they do anything to help a valuable community resource?
It goes further; there are some great employee’s in Sun who have a can-do attitude, and really want to make Solaris succeed; and yet, there are a huge number who seem to allow their toxic personality to infect all areas of Sun and as a result harm any possibly relationship between Sun and the community.
There are a large number of programmers within China; and again, where is management within Sun encouraging and boosting the engineers confidence to get involved with the Solaris community?
Its a lack of leadership within Sun, and like I said, there always seem to be a lack of someone wiling to stand up within Sun and say, “this is our vision, this is where we want to go in the next 5 years, and by hell or high water, I’ll drag this company kicking and screaming to that end target”.
Look at kaiwai and his bitter diatribes laced with MacOSX supremacy.
>Compared to 5 years ago, there are even less software
>ported to Linux.
Proof ? Tell me which software was being ported 5 years ago and has been discontinued. Things like VMware work great under GNU/Linux by the way, and are as updated as Windows versions … Java has been opensourced too.
>Compared to 5 years ago, WINE is no better.
WINE is far better than 5 years ago. There are nowadays other alternatives to achive things that WINE wanted to do then, and may be not an important goal today.
>Compared to 5 years ago, 3rd party driver support for
>hardware is no better.
I don’t buy this one. It may not be evolving as fast as we’d like, but it is improving anyway (see nVidia or AMD-ATI drivers and future plans).
>Compared to 5 years ago, OpenOffice has in all
>honesty, went backwards. Microsoft Office is EVEN more
> dominant now, than it was 5 years ago.
Office has an Exchange integration that a lot of companies do need and that OpenOffice does not provide. Anyway, OpenOffice is now much better than it was 5 years ago … weren’t we talking about GNU/Linux ? You can run Microsoft Office under CrossOver Office (base upon WINE) or with VMware for example … if you need to do it. No need to get tied to the underlaying OS anymore …
>These are just my personal observations.
I don’t agree. The previous ones were mine.
>We have ex Windows users who don’t care about the
>principles of Free Software, and the GPL licence, all
>they care about is not having to pay any money for the
> software.
So if GNU/Linux is dying, why are there ex-Windows users running it ? That seems to me a growing user base.
>This leads to the ignorance of the majority of the
>users – the desktop users.
Each Internet surfer may be “using” a lot of GNU/Linux machines without even knowing it, desktop usage is just the more visible, not always the main one.
Why is desktop usage the more important ? You can now surf the web or play games without a PC at all. Mobile devices are also there, and Java is good at that market.
>2. Narrow it down to one package manager.
>3. Narrow it down to one desktop environment.
Why ? I’ll never understand this self limiting approach to things … variety does not assure as quality, but neither do it the lack of choices. Personally, I like to have more than one option for almost everything in life, and package managers or desktop environments are not different in that matter.
>4. Get games ported to Linux (see point 1).
A console would do it far better.
>5. Get much better 3rd party hardware driver support,
>rather than ugly, open source hacks that aren’t
>reliable, or only partly work (or both).
We’re on the way to that. Open source hacks are reliable, though I agree that sometimes support is imcomplete for all the existent features.
>Just my honest thoughts. Don’t get me wrong – I like
>Linux, I like it a lot.
I do not believe you, that’s my honest opinion.
Sure enough, Jonathan just blogged about it:
http://blogs.sun.com/jonathan/entry/yes_it_s_true