Sun Says OpenSolaris Will Challenge Linux

E5Rebel writes "Sun Microsystems has ambitious plans for the commercial and open source versions of its Solaris operating system. The company hopes to achieve for Solaris the kind of widespread uptake already enjoyed by Java. This means challenging Linux. 'There's an enormous momentum building behind Solaris,' according to Ian Murdock, chief operating platforms officer at Sun, who was chief technology officer of the Linux Foundation and creator of the Debian Linux distribution. Isn't it all a bit late?"

Comments

What is the platform?

[BadAnalogyGuy]

What's the point of an operating system when you've got Java running on top of whatever is there? The OS is just a bootloader for the Java VM.

Sun's interest in pushing two separate platforms is baffling.

Re:What is the platform?

[setagllib]

Sun has a lot more than just Java software, and has a lot to gain from having firster-than-first class support for Java in the operating system (e.g. kernel-level code caching, pushing code into kernel space, etc). Linux can technically have it all now too, with Java being GPLv2'd. But really, Sun has packages like StarOffice, which needs a lot more than just a JVM.

I encourage more competition for Linux. A free market is built on competition. Now that Microsoft is becoming a competitor rather than an oppressive regime, it'll be naturally selected out and increasingly powerful Unix systems will dominate the market. A Linux monopoly is not a good thing either, and whether BSDs or Solaris share the market, we all stand to benefit.

It'd be even better if we had some license consolidation, but hey, that's a pipe dream. I'd rather have license-incompatible code than no code at all because people refuse to use GPLvX.

Re:What is the platform?

[Anonymous Coward]

The concept of competition does not apply to free software because competition implies a winner and a loser. In OSS, there is no winner, nor is there a loser. OSS projects progress by the input and enthusiasm of the users. There is no reason that a single "monopoly" project would necessarily lead to lower innovation. Since the project itself is not in any competition to lure users away from a competitor, there is no incentive either way to innovate except for the "itch" to keep making the project better.

A monopoly-style OSS project would lead to more innovation, in fact, because with more users wanting more features, the project will have both a larger pool of ideas to choose from as well as a larger pool of developers to implement and grow the project. Growth encourages growth, at least as far as OSS is concerned.

Competition, OTOH, draws finite resources away from the developer pool. While ideas may be freely shared, developer time cannot be, so a project that gets X number of hours of work will have monopolized that time for that project. Sometimes this work can be easily shared among other projects, but most of the time it cannot be shared without significant porting and adaptation. Competition fragments the development effort of all OSS projects.

The only competition that truly exists in OSS is the competition of ideas. The actual implementation of code is where this is fought. If idea A has more support than idea B, it will be idea A that gets implemented. In this way, in democratic fashion, the best ideas (alternatively, the most popular ideas) get turned into reality. When the small group of idea B supporters break away from the main project to proceed with implementing their idea, only time will be able to tell whether idea A or idea B was the right way to go. But it is an unnecessary competition and draws resources away from the improvement of the platform.

Competition against Microsoft or Sun is not the reason Linux improves over time. Rather, it is because users who want to use Linux implement the features that they want so that the platform grows to fit them. As it grows to fit them, it also grows to fit everyone. The additive nature of OSS sees to it that the best ideas stick around and the lousy ones get tossed away. That's not to say that Linux isn't stuck in the Unix rut, because it is. It's that if there were no Linux, there would be something else.

Re:What is the platform?

[Starker_Kull]

I encourage more competition for Linux. A free market is built on competition. Now that Microsoft is becoming a competitor rather than an oppressive regime, it'll be naturally selected out and increasingly powerful Unix systems will dominate the market. A Linux monopoly is not a good thing either, and whether BSDs or Solaris share the market, we all stand to benefit.

Your faith in Microsoft being 'naturally selected out' is.... amusing. Considering, after years of barely adequate products, they still have 90% plus marketshare of desktops, and last I checked, they were still oppressing various standards bodies, hardware manufacturers, small software houses, etc., I think the corpse is still walking around, talking FUD, and otherwise making a nuisance of itself. The Linux Monopoly you fear is... a bit far-fetched just yet, IMHO. When I start seeing KDE desktops in some of the small offices I walk into, then I'll believe it.

Of course, this move by Sun is to try and make that happen; many non-computer people like 'simplicity', in the sense of getting everything from one computer vendor with minimum fuss on their part, assuming that things will work together more smoothly then. So, Sun offers a machine running OpenSolaris, with StarOffice preinstalled, as well as a really fast JVM. Worth a shot...

Re:What is the platform?

[setagllib]

Why do you think Microsoft is scrambling for OOXML standardization? Because the document format lockin is a huge, huge part of Microsoft's monopoly strategy. If they're forced to be an equal player in the office suite space, making Office largely replaceable, then Windows is largely replaceable too. When Linux + KDE + Firefox + OpenOffice.org can replace a Windows + Office + IE setup with lower costs, minimal training and solid vendor support (Canonical, Red Hat, ...), how much incentive is there to run Windows any more?

Gradually the government switches, corporations switch, and finally users switch. The numbers indicate it's happening anyway, and the format war is just going to nail the coffin on Microsoft's monopoly. They never even had a monopoly on servers, gaming technology, etc. so the office is their last stand, and in a matter of days it will be confirmed that they have lost that too.

And of course, as the demand for Linux installations grows, and more vendors sell pre-packaged Linux, then hardware contracts will also require useful drivers or even documentation, and the hardware situation will be largely solved too. Sit back and relax, freedom has won and the liberation continues as planned.

Re:What is the platform?

[trifish]

In OSS, there is no winner, nor is there a loser.

If 90% of people used a particular open-source program, I'd dare to call that program a winner. And if nobody used a particular open-source program, I'd dare to call that program a loser. The rest is idealistic crap.

Solaris

[Borongo]

The kind of Solaris penetration sun really wants is at the corporate
level. There are a lot of Sun Servers out there so they'd like to increase
that further in companies who want cheaper hardware than the sparcs.
From a TCO point of view, add Solaris X86 to your existing Sparcs isn't
that big of a deal and Sun has made pretty good progress in making Solaris
10 much more on equal footing with Sparc based Solaris so now you only
need admins who are expert at one OS, you've got easier compatibility
with your software etc. Then from there I see a push to companies who
don't use Sparc hardware.

What can Sun bring to the table?

[140Mandak262Jamuna]

What can Sun Micro Systems bring to the table that rest of the Linux could not? Its name, some kind of relationships with corporations and provide "not a bunch of amateurs in their spare time, this OS is backed by professionals" kind of sales talk. But that niche is already occupied by IBM. As for SUNW's vaunted professionalism, they fumbled the lead they had in unix and are struggling to keep up. As for their corporate vision, these guys never realized until it was too late, that Windows OS was the loss leader, in grocery store parlance, and the real deal is the vendor lock in office documents, email addresses and calender applicaions. MSFT might have fumbled many balls and lacked vision on the technical side of the market, but when it comes to business side, MSFT has been nothing less than visionary in gunning for monopoly and achieving it. Now SUNW is going to take on Linux? yawn. Nothing to see here, move along, folks.

Solaris has known stability...

[TooTechy]

Solaris has known stability in certain supportable configurations. Linux supposedly does too. I know that statement will get a lot of hackles raised but just hold on. I am a continuous Linux user since 0.99pl8 and I love it. But, as time moves on I see some instabilities creeping in as complexity rises and hardware moves on.

One of my boxes downstairs, a recent machine (less than 6 months old) running stock Debian (amd64) without a mod to the sources.lst has a slight instability (almost certainly in a driver) and crashes every week or so.

Now, one could say that I should replace the hardware which has the suspect driver (always seems to be on a disk access). Or I should get on the Debian lists and report it. If it was a Sun Solaris box I would know that the hardware I had was (or was not) supported. The word 'Supported' in the Linux world really (I am sorry) does not mean as much as it does to Sun.

Now I have other Linux boxen, (a little older) which have uptimes of over a year. No problems. But on odd occasions as this I would like to have stability and I can't find it. (Read, maybe don't have the time at the moment). And I need the box UP. I can't rebuild it AGAIN! I am on the 6th distro in an attempt to gain stability. That's an aside.

In Sun's world. You pay a little more for your hardware and 'Know' it is going to work.

Re:I remember when they opened the source

[E-Lad]

So what you're saying that you expected it to happen overnight?

I recall people saying similar things, only about Linux, back in the 90s. "Linux is the next big thing", Pundits and advocates trumpeted "Corporations will move to Linux as their preferred server/service platform", and so on. That pretty much did happen, but it took the better part of a decade to realize it. It took the one thing that a not even the most talented coders can't create during an all-night coding binge: Time.

OpenSolaris is a hair over 2 years old now. If you think about it, most decently sized shops change out comodity infrastructure every 3-4 years, a time frame pimarily goverened by hardware warranties. If an organization says "Let's try another OS the next time around... lets try Solaris" then the proper time to do that would be consumate with normal upgrade cycles. In other words, no one can reasonably expect one thing (Solaris in this case) to massively gain meaningful, measurable share instantly. It takes time. Just like it did with Linux.

Re:Sure it is fscking late !

[Eivind Eklund]

No, the GPL has the problem of co-existing in the same app as CDDL.

Place blame where it belongs - GPL is the one bringing the heavy restrictions creating license incompatibility with EVERYTHING that cannot be converted directly to the GPL (including all BSD style licenses, if you do an exact reading of the GPL and BSD licenses.)

Eivind.

Re:It's rarely ever too late

[Alioth]

ZFS is irrelevant to the desktop user, though. (How many desktop users care what filesystem they have?)

However, a stable kernel ABI - which Linux doesn't have - is FAR more important, as it means hardware manufacturers are far more likely to release drivers for your platform that can just be installed with the hardware. If Solaris on the desktop started outnumbering Linux on the desktop, my bets would be it would have everything to do with hardware manufacturers being able to ship a driver for $random_hardware, and little to do with ZFS.

Re:OpenSolaris

[hjf]

It is quite entertaining to see Murdock making such claims. He actually forgets that the greatest strength of Linux is that most of its codebase is understandable. While it may be missing some high end enterprise bells and whistles a relative newbe can sit down and understand most of the code straight away. Granted, his attempts at coding anything for it may end up being futile, but he will like it none the less.

I wonder how many Linux users are actually programmers? Like 1%, I guess? Sure, in 1991 when it was released, every user was a programmer. But now it's the opposite. Few users will do so much as recompiling a kernel, and even so, you don't need to be a programmer to do that.

On top of it he has the greatest possible documentation - the code and it is readable.

What? Are you on crack? Code is NOT documentation. You HAVE to add a manual somewhere, else it's "just a program". And that's the biggest problem with Linux. Documentation. There's a million things you can do and very few of them are documented. So you have to google everything. You'll have to end up at some obscure list server (which WILL be offline when you click on it, so pray that web.archive.org has a copy).

The other day I had this situation: A SCSI drive failed and md was degraded (raid-1). The drive was unaccessible, I didn't know that. So I went ahead and installed a new kernel. LILO was bitching about not being able to find /dev/sdb. So I go an run LILO again and forget to add the "-t" switch. WRONG - bootloader is fucked now.
I had to boot Debian Rescue, mount my drive (it's a LVM on MD). I figured, what I had to do was just very simple:

boot
mount the partition
lilo and read the config file from the partition... that didn't work, the files weren't there

ok, so I chroot into the directory. lilo. didn't work either, something about /proc
ls /proc. empty. what the hell? mount /proc. ls /proc. all there. lilo. bingo!

I would love to see a newbie doing all that guesswork just to recover a fucked MBR.

Regarding to the "high end enterprise bells and whistles": ZFS alone made me switch my Linux server to Solaris. I lost, completely lost, 320GB of data due to the piece of shit Truecrypt for Linux, supposedly "stable". Now I have a zpool with iscsi-exported zvols, that took like 2 minutes to make.

The great about solaris is that it WORKS. Right there and then: it just works. If it doesn't work, that's it. They don't pretend that it works only to have it hang at the worst moment (or worse: fuck 320GB of your data). I think that's another problem with Linux: version numbers. Serious programmers put 0.0.1-pre-alpha on their versions, so you kind of know what you can expect. Others just go and version 1.0 (and when you try to run that program, you realize that this isn't a 1.0 version). I don't think corporate folks like beta software, and that's what keeps Linux off the enterprise too.

Linux makes a great LAMP server, Asterisk server, etc. But that's because of the support behind those products. Asterisk, PHP, etc are backed by serious companies.

And don't let me get started on the stupid fights about the scheduler, while this isn't an issue on Solaris (http://blogs.sun.com/darren/entry/new_linux_sched uler_old_solaris), because that's what really makes me doubt about the Bazaar way of software development. Don't get me wrong, I think that's great, but when shit starts to fly around, I start looking for alternatives.

Re:OpenSolaris

[00lmz]

Regarding to the "high end enterprise bells and whistles": ZFS alone made me switch my Linux server to Solaris. I lost, completely lost, 320GB of data due to the piece of shit Truecrypt for Linux, supposedly "stable". Now I have a zpool with iscsi-exported zvols, that took like 2 minutes to make.

ZFS sounds great, but I don't think it's fair to compare TrueCrypt (which is not included with the kernel, and doesn't have too many users testing it) with ZFS (which is one of Solaris 10's most valuable features). Why would you put 320 GB of data at the mercy of TrueCrypt? A few hundred megabytes of sensitive files, sure... but 320 GB?

Re:OpenSolaris

[heelrod]

I'm gonna have to agree with this.

back in the day, the people using this stuff were programmers, so the code made for good docs ( kinda ) because it was new, untested, and really just a toy. Today, it is a big player in real world applications and systems. Most people just hang on to what we said over 10 years ago. It's starting to sound like a broken record. The Linux zealots keep yelling the same thing. I write drivers, port drivers and all that crap, and I'm sorry to say Linux needs to come up with a better documentation scheme just for the simple fact that it is becoming bloatware.

There I said it! Flame me if you must, but as for me, I think a new free OS would be great. Linux is getting boring anyway.