Will Open Source Solaris Kill Linux?

Scott_Blayney writes "This guy contends that now that Sun is releasing Solaris 10 under an open source license, Linux will begin to wane in popularity. To quote, "Linux will probably not grow much beyond its current market share of about 10 % leaving Red Hat and especially Novell with a big problem." "

Comments

Uh...

[danielrm26]

"Will Open Source Solaris Kill Linux?"

No.

Re:Uh...

[nova20]

No.

Well put. Just making an OS open source doesn't make it any better. I find that solaris just isn't as "friendly" as linux, and I think coders that would do anything useful for solaris will prefer to spend their time (more wisely) trying to improve linux -- after all, it's what they've been doing for years.

Re:Uh...

[saintp]

Blow-by-blow:

The Solaris software is of proven quality and at least equal or better then Linux and the open source model will assure that it stays up there.

I'll grant everything before the "and." We don't know what Sun's OSS model will look like. It certainly won't be the GPL, and I'll be amazingly surprised if it's even as liberal as the BSD licence. People aren't fond of giving away their code so that a corporation can make money off of it, so if Sun's model is anything like Microsoft's Shared Source initiative, it will stifle development by the community, not encourage it.

By making it work with competing hardware platforms, there is no reason anymore to switch software to facilitate lower hardware costs.

Granted. Solaris on x86 was a very wise move. However, one big reason to run Solaris is the tight integration with hardware, which can't be said about Solaris on x86. Also, as much as Schwartz talks about running Solaris on Dell (HP, whatever) commodity boxes, it has a very short hardware compatibility list -- much shorter than that of Linux.

Sun with Solaris has already a large installed base and by becoming free and open source there is no reason for existing Solaris users to switch to Linux. Circular. He's arguing why people should stay with Solaris; "there's no reason to switch" is not a reason at all, but a question-begging.

Sun has a proven reputation in terms of quality of support. This should be at least as good or better then that of the Linux supporters. The help ticket I have currently open with Solaris will turn five months old on Wednesday. It was three weeks before any action was taken on it. It also had a four-hour response time. The issue has still not been resolved. Sun's support is far from stellar, despite what they'd have you believe. In fact, their support (or lack thereof) has been the trump card in my quest to get my boss to accept Linux.

Because Sun by default is the only designated party managing the open source software, there will be no risk of a version bonanza with multiple incompatible versions.

This may be nice to the PHB, but the ability to fork is something developers like. Plus, as was mentioned on /. just today, Linus is the final arbiter of the Linux kernel and, as long as he keeps doing a good job, will remain so. Multiple vendors (as opposed to multiple versions) ensures healthy competition, which is why Linux has gone from a nifty experiment and useful OS for unimportant things like web servers to OS of choice for everything from Wall Street to government desktops in just a few years.

I see no danger to Linux. And frankly, Solaris was already dirt cheap, but the support still costs big bucks, which means that Solaris still costs about as much as RHEL. Or, if you don't want support, it costs just as much as RHEL without support. That's the only front they gained on, and I don't see it as very critical.

Re:Uh...

[niiler]

Thank you. I've used Solaris for a number of years. And while it's not bad, these days, Linux is honestly a much more robust system in terms of software and hardware support. Additionally, (I have no figures to support this), but considering the previous cost of Solaris, I'd be willing to bet that there are many more Linux users. This doesn't necessarily make it better. I just interpret this as yet another premature prediction of Linux's demise.

Let me add, it's not hard to secure Linux especially since the Bastille scripts work with more and more distros and many distros I've encountered already have some hardening done (in addition to firewalls, etc...) right out of the box.

Much of the original poster's citation of benefits seem to be largely from his inexperience with Linux and acceptance of the usual corporate FUD.

Re:Uh...

[Pieroxy]

Much of the original poster's citation of benefits seem to be largely from his inexperience with Linux and acceptance of the usual corporate FUD

That is probably what makes his point valid, as much of the decision makers are corporate ignorants (at least technically ignorants)

Re:Uh...

[Zemran]

The contention that Solaris can kill Linux comes from the mono-os culture and that is what is wrong with the current M$ dominated situation. The contenion seems to be 'how can Linux achieve market dominance if there is another good player in town?'. It is not the Windows OS that is the problem (although I do not like it), it is the fact that it has such a dominant market share that causes problems. If Linux could get 20% of the market, Solaris 20%, Apple 20%, M$ 20% and the variety of others taking the other 20% we would have a good, healthy market. Try writing a virus that works in that market. I think if Linux had 90% of the market we would be back at square one. I welcome more choice and think that more choice and competition will lead to more innovation and even more choice in the future. The reason I hate M$ is because they think they know better than me what I want to do, I use Linux because it lets me do what I want. Some people want M$ to make those decisions for them and that is their choice. It is not wrong for someone to make a different choice, it is wrong to try and remove those choices (as I believe M$ tries to do). Solaris will give another choice which will suit a lot of people. I do not think this is as much of a threat to Linux but more of a threat to M$.

It all depends on the size of the group support

[eltoyoboyo]

If Sun continues to support it with their engineers, and IBM continues to support Linux with Their engineers, it will still be a battle. I do not look for either side to gain much ground, (netcraft aside.)

Re:It all depends on the size of the group support

[Sesostris III]

Its worse than that. IBM may support Linux, but Linus is not IBM. This allows Intel, HP, Novell and others to also support Linux.

Solaris, even if Open Source, is still Sun. Would Intel, HP and Novell, let alone IBM, be willing to support a platform that is still linked heavily with Sun, rather than one that is truly "free"?

I doubt it.

Erm...

[Punboy]

Doesn't Linux support a vastly large collection of hardware than Solaris? Also, Linux isn't just popular because of its open-source nature, but also the philosophy behind its design and development, as well as the number of professionals behind it. I don't see the NSA donating to Solaris, now do I see most of europe backing Linux.

I haven't RTFA yet so please correct me if I'm wrong.

A little reality check

[jmorris42]

Couple of false assumptions in the guys story. Typical when suits try to make tech decisions, especially when they are fool enough to believe their LACK of knowledge makes them more qualified, as this guy does.

A few of the obvious clues missed are:

1) Linux is already ahead of Solaris on Intel hardware, not behind as this guy believes from reading Sun press releases.

2) Solaris is not known to be portable beyond Sparc, Sparc64 and ia32. ia64 and AMD64/x86-64 might happen but as far as I know don't yet exist.

3) Sun has yet to announce a license for Solaris, it is very doubtful it will be actual Open Source and almost certainly not Free Software in the FSF sense of the term.

4) Sun is almost certain to keep parts totally closed due to licensing terms with third party suppliers.

5) Sun will rig things to retain ALL creative control from the Java experience. This will preclude any sort of community involvement on the scale needed to compete with Linux.

Re:A little reality check

[Anonymous Coward]

Sun has yet to announce a license for Solaris, it is very doubtful it will be actual Open Source and almost certainly not Free Software in the FSF sense of the term.

Actually, with respect to this point, yes it will 100% definitely be an Open Source license, as in one of "approved" ones that is generally accepted as one. Probably it won't be "free", as that would mean BSD (and relatives, MIT, other laissez-faire ones) or (L)GPL.

Why on earth would Sun announce Open Sourcing Solaris, and then, ummn, not do it? They haven't tried Microsoft way of "shared source", and have actually released a few products earlier using Open (and Free, OpenOffice is (L)GPL) licenses. So why wouldn't they manage to do the same with Solaris.

I agree with other points, though (with maybe exception of (5)... but I think it's too early to comment on that, everything's possible).

Re:A little reality check

[Waldmeister]

1) Linux is already ahead of Solaris on Intel hardware, not behind as this guy believes from reading Sun press releases.

Is it? My first filesystem tests showed almost par. ext3 is much slower on some tests than UFS, Reiser is a little bit faster.
The GUI (JDS3) feels much more responsive than a Gnome desktop on Linux. Maybe the reason for that is the "interactive" process class of solaris, which Linux lacks. BTW: Windows has such a class, too.

2) Solaris is not known to be portable beyond Sparc, Sparc64 and ia32. ia64 and AMD64/x86-64 might happen but as far as I know don't yet exist.

Solaris 10 does include support for AMD64. And there have been ports of Solaris to Power (which got killed by IBM) and Itanium (which suffered from several reasons).
And how much do other platforms that x86 matter today? If there should be really demand in the OpenSolaris community, they can hack support for PowerPC, ARM or whatever they want.

3) Sun has yet to announce a license for Solaris, it is very doubtful it will be actual Open Source and almost certainly not Free Software in the FSF sense of the term.

OpenSolaris is not the same as Solaris. Yes, there are still the gory details left. But there is already a closed "beta test", and the license will be OSI compliant.

4) Sun is almost certain to keep parts totally closed due to licensing terms with third party suppliers.

We'll see. From what I've heard from people which already have access to OpenSolaris, it looks pretty complete.

5) Sun will rig things to retain ALL creative control from the Java experience. This will preclude any sort of community involvement on the scale needed to compete with Linux.

We will see, how much community interest Sun will get. Interest in OpenOffice seems to be quite good. At least much better than other community efforts like Koffice. ;-)

Re:A little reality check

[plasticsquirrel]

Is it? My first filesystem tests showed almost par. ext3 is much slower on some tests than UFS, Reiser is a little bit faster.

I didn't know that a filesystem benchmark was a comprehensive test of hardware support. What about basic sound cards? Available accelerated video drivers? The fact of the matter is that the Linux kernel has more drivers available in it for x86 hardware than Solaris will likely ever have. It is not easy to get the same sort of hardware support for x86 that Linux has gotten over the past 13 years, and hardware autodetection is always getting better for it.

The GUI (JDS3) feels much more responsive than a Gnome desktop on Linux. Maybe the reason for that is the "interactive" process class of solaris, which Linux lacks. BTW: Windows has such a class, too.

JDS is also available for Linux and it runs Gnome. "A Gnome desktop on Linux" is very vague, but I will assume that you were running a workstation install of a very robust distribution such as one from Red Hat or SuSe. The extra efficiency you are likely noticing is probably due to the fact that it is a true desktop distribution, which is necessarily simple and manageable, and thus fewer running processes. Even if your opinion involved comparing Linux JDS to the Solaris JDS add-on, I would not be convinced that performance was anything more than a difference in distribution setups.

Re:I don't buy this at all.

[Ian.Waring]

Nor me. If Solaris really did go open source the way most people understand it, the useful bits of code would find it's way into all the various Linux distributions in pretty short order.

Seems unlikley

[SuperKendall]

For one thing, Linux is now as much a "brand" as Solaris. And there are too many people making difefrent distros for Linux to really wane for some time.

But also, consider what "Open Source" really means. I'm not familiar with what OS licence Sun is using, but if it's really "Open" then Linux can make use of the best bits to keep going. If it's not really that "Open" then the current forces that move Linux will continue the course unchanged.

In a world that already has BSD and Linux living together in harmony (well, as much harmony as a VI and Emacs user sharing the same house) there is also room, and a place, for Solaris.

Competition

[barcodez]

Why does one have to 'kill' the other? Why can't they stay in a state of equilibrium goading each other to improve?

Why do people always assume one product has to erradicate all the competition and become the only product of it's type.

Open Source Solaris = Linux with a direction

[stratjakt]

Solaris couldn't kick linux out of every possible niche (embedded wagoo 3sx-12 cpus or whatnot), it could cream it on the desktop.

Sun will throw all its muscle behind it's Java Desktop to deliver a polished, cohisive system. Linux will continue to be pulled in 100 directions at once.

Distros need to stop offering Gnome, KDE, fluxbox, and 9000 other window managers, and pick a path and stick to it.

There really isn't that much of a market for people who like to dick around with 10000 different ways to close a window, each with it's own myriad of quirks and bugs. They like to plug it in, turn it on, and have it work pretty much the same way as the one in the next cubicle, or the next building.

Linux' strength (versatility) is it's achilles heel when it comes to the desktop market.

Re:Open Source Solaris = Linux with a direction

[Stephen Samuel]

There really isn't that much of a market for people who like to dick around with 10000 different ways to close a window, each with it's own myriad of quirks and bugs.

No, there isn't -- and that's why most people don't. Most people will learn one window manager and stick with it. If I'm using KDE, the fact that Gnome is also loaded onto my box is entierely immaterial -- unless my friend who likes Gnome comes to visit, at which point he's free to use his favorite desktop.

The 'do exactly as I say and nobody will get hurt' attitude is why many people hate microsoft. Sometimes (OK: Often) what works for MS's business plan doesn't turn my crank.. The ability to do it my way rather than your way is what makes Linux such a good choice for so many people. We're not at the mercy of somebody elses business plans and/or persnal quirks.

Re:Open Source Solaris = Linux with a direction

[ErichTheRed]

Linux' strength (versatility) is it's achilles heel when it comes to the desktop market.

Yup...that's what still has a lot of IT people turned off of a Linux desktop. Luckily, in a corporate environment you can force your users to do things one way. However, who's to say that the "one way" will be available to you 4 years down the road? A commercial software company (like Sun) can provide direction, and a sane migration path from the old way to the new way. The old argument against Linux was that you couldn't get support from a bunch of hackers. Now the argument is that there are thousands of ways to do things. How many Linux programs are named "Yet Another..."?

Red Hat and SuSE have gotten better at this lately, but they still include way too many things out of the box. If I were building a desktop distribution, I would pick ONE window manager, bundle in enough hardware support to make things as plug-and-play as possible, put in ONE office suite, and a few other useful apps. Then I'd make the system easy enough to tweak later on, so anyone who wants to can grab components and install them as needed.

Linux would really benefit by having a few vendors back one set of standards. It would be easier to train sysadmins as well as end-users.

Too Late

[Doesn't_Comment_Code]

If Solaris had done this TEN YEARS AGO, then maybe. As things stand today, no. It comes down to a matter of trust. Do most Linux users trust Solaris enough to let go of Linux? No.

Take a choice between A and B. A is commonly considered a better product than B. Most will choose A. But B offers something that A has never dreamed of. In fact, A hates the idea of offering what B offers. B gains a lot of support. Time ticks by, and after seeing how much B has benefitted A changes it's mind and offers something similar to what B offers - but only because it has to.

Do all those people who would have originally chosen A, but chose B go back to A? NO. Because B is of a similar mindset to the choosers. B WANTS to offer things to the chooser. A offers because it has to remain competitive. That is HUGE. And that is why Solaris will not kill Linux.


P.S. A is not necessarily better than B.

Pundits suck

[kendoka]

Since when has a programmer enthusiastically dropped their own code to pick up someone elses? Not a technical person, yeah, obviously. =)

Arches?

[giantsfan89]

How many architechures does Solaris support? Sparc and x86? Is that all?

How many does Linux support?

As a Solaris Admin

[Chitlenz]

We've been using the beta of 10 for about 3 months now in testing with Oracle at our site. It's pretty, that's for sure, with it's nice Gnome facelift and all, but I think essentially that Linux and Solaris fill different roles, at least for us.

Linux feels a lot more like a general purpose OS than Solaris, and 10 while friendlier is still very much rooted in the proprietary Solaris tradition.

To sum it up, this is good for Solaris users who can throw away the whole CDE/Openwin experience and replace it with something refreshingly cleaner, however we were going to adopt 10 anyway. It seems to me that Sun is going to have to really dig and build new things rather than merely say 'hey we're OSS now too!' and expect Linux users to suddenly flock their way.

Looking Glass is an excellent example of software people could see as a reason to change platforms, and IMHO Sun should focus on this type of admittedly risky innovation instead of attempting to lure existing Linux users into a Solaris world. Someday, all these marketing guys are going to realize that there's more to an OS than just a name, and that actually creating something new is the best approach to picking up market share (OSX == good example).

All that said, ZFS is a really cool thing if it works as spec'd, however ZFS is NOT in open beta AFAIK, and it has not been released to us little guy partners as of yet (though I'm betting they've started testing at the larger shops), so all we have here to go on so far is a marketing claim of improved i/o that could be true, false, or in that grey area salespeople like to use where the whole thing ran great! .... once .... in a lab... and it isn't reproducable to the average joe...

I'm guessing we'll all know in about 2 months.

-chitlenz

Pfft!

[FyRE666]

As anyone who's tried out x86 Solaris is aware; it's no match for Linux on the platform - or any other platform other than Sun's own hardware for that matter. The hardware support is pretty dire. Sure, it's supposedly a superior kernel, but if it's unable to take advantage of the metal underneath it, then it's not going to win market share.

(yes I have tried it out, and it lasted a full week before I threw up my hands in disappointment and went back to the penguin ;-)

Re:The Desktop

[temojen]

But if Linus finally picks a GUI and starts up a desktop linux fork...

??? First of all, why should Linus pick a GUI? He's the head of the kernel development, not a distribution vendor.

And as for a desktop linux fork, that also makes no sense. There are lots of desktop distributions (SUSE, Knoppix, etc). There are no reasons to fork the kernel for desktop use. If the distribution vendor doesn't want RAID, SCSI, etc, they can just configure them out.

... innovate, simplify and give desktop users a reason to use it.

You mean like stability, security, efficiency, etc? Using a recent distribution is no more difficult for home and office users than using windows or MacOSX. Installing it may be annoying, but few home users do that with windows anyways. Most get it with their computer or use the restore disk that came with the computer.

Re:The Desktop

[mrchaotica]

The one GUI-related thing I think Linus ought to help with would be a UNIFIED, distro, toolkit, and desktop environment independent hardware control panel. It seems to me that Linus + Freedesktop.org + Linux Standard Base is what would be required for "the Linux Desktop" to be widely accepted.

Re:I don't buy this at all.

[killjoe]

"He's basically lost most of his credibility there, because a good decision maker needs to bring in technical people to explain this."

To me he lost his credibility by talking about an open source solaris before such a thing even exists. Why don't we wait till sun actually open sources solaris, examine the license, examine what had to be taken out due to licensing, examine what sun says about suing people who may be reading patented code and THEN start speculating about how it will all effect linux.

For example if SUN releases solaris under GPL no problem. Linux will take from it and improve itself. If it's under the BSD-like the same thing will occur. If it's released under something like the Java license I don't expect them to build a decent community or to present any real competition to linux.

Maybe, depending on how you define kill

[starseeker]

Most of the comments I've see so far are essentially "No." I disagree, but before I do so I'll make the following assumptions:

a) Sun uses a real open source license - meaning GPL compatible. This is unlikely, and if they don't Linux isn't going anywhere.

b) Sun doesn't reserve major high end components of the system. If they do, they are essentially another Linux with different and unfamiliar source code. Solaris can do some things well that Linux isn't good at, but if Sun cuts it down, game over. Open source Solaris is a non-starter.

NOW, let's assume they do it the Right Way. Several things immediately happen. Solaris has a good track record in high end servers that has been earned through experience. That gives it a lot of interest right there. Now, with real open source, drivers from Linux start arriving in droves. Intel performance is improved. GNU tools become standard parts of the system at almost all levels. Solaris becomes a more robust Linux, with a proven commercial track record.

Now, remember - to everyone above kernel level, KDE on Linux vs. KDE on Open Solaris is going to look VERY similar. Ditto for any other user (and even most developer) tools. Parts of Solaris that annoy users suddenly are fixable, and get fixed. Open Solaris can now go head to head with Linux, as a tier one platform for virtually all open source applications.

The parts of Linux that are not available in Solaris will be adsorbed in. Things like Dtrace are already available on Solaris, and not available on Linux. As people rapidly add in the goodies, Solaris adsorbs Linux's strengths (drivers, file systems, etc.) while leveraging Solaris's robust, tested, industry strength core. As far as I know, there is very little about Linux to fundamentally recommend it over Solaris, except for it's ability to function well on many platforms. If Solaris proves able to do this (impossible to say right now, but I'm assuming well done code will be able to do this, given enough eyes and hours) Linux will essentially dissolve into Solaris, and the end product will be Open Solaris with the best parts of Linux preserved. Does that constitute killing Linux? I wouldn't say so. If you mean will Linus stop being the prime mover, maybe, but the effort put into Linux will survive.

So I, for one, would welcome our GPL Solaris overlords, because the only thing that will happen is net gains all around, regardless of whether the end product is called Solaris or Linux.

Now, do I believe Sun will Do It Right? Nope. But I sure hope they do.

Re:I don't buy this at all.

[Anonymous Coward]

The author starts the article by declaring "Let me start by saying that I'm not a technical expert. Although this might be a handicap it could very well be a positive in looking at the issue of Sun versus Linux in the server business from a decision maker's point of view." He's basically lost most of his credibility there, because a good decision maker needs to bring in technical people to explain this.

I'd say all, because if he isn't a technical person he frankly doesn't understand what Solaris and Linux do. If he doesn't understand what they do he has no capacity to understand which one will do better at it.

Open source isn't some kind of magical fairy dust. It's a hell of a good development model and it can bring success to products that might not have gotten it otherwise but it does not equate to instant success.

This guy thinks of open source as magical fairy dust. Becuase he is not a "technical person" (in other words, because he is not informed) he thinks, hey, Solaris has these advantages, and Linux was open source, therefore Linux has been doing well becuase Linux is open source.

This is not accurate. Linux has been doing well for a great many reasons. Many of these are byproducts of the fact that Linux is open source. But open source is not the reason. Frankly, Solaris's customers don't care if it is open source. They want a good cost to performance ratio. They want their administrators to be easy to find and efficient. Linux provided both of these things. Solaris does not. This is why Solaris's customers have been leaving for Linux in droves. Not because Linux is "free"; becuase Linux is quality.

Perhaps this guy thinks that these "technical" matters don't matter because Solaris will succeed where Linux did not (the Desktop). There isn't much good reason to believe this. Sun has no history of creating end-user software. They have a bunch of money but other than that they're no more likely to mysteriously produce The Great American Desktop OS than, I don't know, Novell, or Nabisco. Creating a usable desktop system is hard. Creating a usable desktop system from a starting point made out of old, server-oriented software (such as the Linux or Solaris platforms) is almost even harder.

Re:I don't buy this at all.

[michael path]

I remember a similar problem with Microsoft CRM. The idea of Microsoft making a CRM product rocked hardcore to those in a Microsoft Office environment with Great Plains products abound. The execution was terrible, and now it's an embarassment.

People LOVED the idea of Microsoft CRM. Solaris people are going to LOVE the idea of Open Source Solaris. But does anyone really believe Sun is going to do this right? Moreover, what would be the right way to take an existing commercial server product, open it up, and find a way to make more money off of it?

Perhaps if this move is successful, Sun would start abandoning hardware, and move towards a more services-based company - like the transition IBM started 2 years ago?

There are many that still don't get it, 10yrs late

[postbigbang]

Sun doesn't get it. Their sense of open source only works when you write FOSS like this: F|=O$$.

As said before:
1) no Solaris on a mobile phone
2) no Solaris on a laptop/notebook/pda
3) no Solaris on a media centre
4) no Solaris in an automotive ECM, and so on.

I've talked to their PR people, and Sun engineers. They DON'T get it. Their idea of community is a country club. It's stockholder interests that they have at heart. That's ok.

What's not ok is to make believe that they're going to get Linux Love by putting on a blond wig and some lipstick.

The license is the key.

[miffo.swe]

What SUN and many others miss is that the most compelling thing about linux and linux distributions isnt the system in itself, its the license its released under. The GPL license is a protection against malicious corporations and what prevents anybody from gaining strangelhold. It forces corporations to work together with the users in a never before seen way. I and many with me will not migrate to a license that makes it possible for SUN to just rip the floor from under our feet in five years time. Even a GPL license will be very scrutinized so that all the bases are covered and not just some parts of the system.

I think they put all to much weight into peoples cheapness and think that GNU/linux is all about money. Well a big part of the money bit is to not let anyone lock you in like MS did. I find this a desperate move from SUN who cant decide what leg to stand on. MS is sure to be happy to have one single entity to crush if SUN should gain foothold with Solaris. With linux its just not possible to stomp it out in one blow.

This is perfect...

[Delgul]

- If they go really GPL-ish, SUN code will be used to improve Linux even further. Ergo: Linux will grow.
- If they don't, they will not get a decent developer base. They win nothing. Ergo: Linux will grow.

Sounds like a no-lose situation for Linux IMHO...