A lot of hooplah has been distributed by Sun Microsystems on the advantages of their OpenSolaris/Solaris 10 release. Martin C. Brown has been using said software for the past few months and files his review that helps answer the question: has someone finally found a Linux killer?
what a way to star flame wars before people even read the article..
has someone finally found a Linux killer?
brilliant!… *rolls eyes*
On this corner, we have the heavy weight champion, weighing in at 250 lbs, Solaris 10. With his heavy frame, he can take hits and give hits back. On the other corner, we have the new comer, weighing in at 175 lbs, Linux. Don’t let his light frame fool you. He has already taken out former heavyweights HPUX, AIX, IRIX, OpenServer, and Unixware. He might not be big but he sure can hit.
.
.
.
.
By unanimous decision, we have a winner. The winner is Microsoft.
you sure? the companies that owned those OS’s all started pushing linux cuz they thought it would kill windows. i dunno about the technology, but the corporate movement to push it was what did it.
I’ve never had a chance to play around with Solaris, but it sounds pretty good on the whole. I guess it has been for a while, like the author says, fairly restricted to certain markets.
This new open stance has to be a good thing, and I really hope we start to see more drivers being written for slightly lower end hardware soon.
The “zones’ functionality also sounds pretty good to me. Particularly in critical environments where a solid and secure server is required.
On a side note though, why is everyone always determined to find a “insert-product-here killer” ? I thought a lot of the open source movement was supposed to be about diversity?
gbye linux.. welcome Solaris!
It is, however, a Red Hat killer. Why on earth would you pay money to Red Hat when you could simply use Solaris? More innovation, better support, cheaper price, better scalability? Hm.
Well, several tests in danish magazines shows RHEL to have a much better performance than Solaris. But again, you can always argue about the test environment.
As I see it, use Solaris if that’s what you like. If you like RHEL use RHEL. If you like neither, don’t use any of them
dylansmrjones
kristian AT herkild DOT dk
Well said, that man.
Why on earth would anyone run closed/proprietary operating systems like HP-UX or AIX? When you can get open source choice from Solaris or Linux?
I thikn HP and IBM have a real problem on their hands.
Because if you need to run a highly secured and integrated system (say you’re bank for example), it can be better to go with something where the vendor has total control of how the system operates and have to ensure it meets your requirements as in the contract.
Open source isn’t always the best option. Just usually
There is still a market for big UNIX, but it is intrinsically tied to the hardware on which it is meant to run. There are four main server architectures in play today, here listed with their OS of choice:
SPARC – Solaris
Itanium – HP-UX
POWER – AIX
x86-64 – Linux
Linux can’t keep up with Solaris on SPARC, especially with the direction Sun is taking the platform. Linux isn’t ready for Niagara, for instance.
HP was pretty much blindsided by the practical failure of the IA-64 platform, and they might as well fold their hand. They were holding kings and flopped an ace, it happens.
I’m biased because I currently work for AIX Development (in the sense that I think it sucks). But there is no question that no other OS can support pSeries like AIX, and there is no question that pSeries is the finest UNIX platform money can buy.
The real area of contention is on x86-64. This is a high-volume, commodity UNIX space. Sun could not compete without a commodity Solaris distribution. OpenSolaris was inevitable from the moment Sun set it’s sights on Opteron. If Linux is to continue its push from web-facing IT infrastructure to the datacenter, it needs to offer a better value proposition than Solaris on x86-64.
To sum up, if your former college roomate offers you a truckload of HP Proliants from his chapter 11 restructuring, then you can either run HP-UX or donate them to a nonprofit for a tax writeoff. If you can afford a couple of Squadrons, then you better run AIX 5.3 with HACMP. There is simply no substitute for proprietary UNIX in these cases.
Until Opteron systems can support hotpluggable CPUs and dynamic memory reconfiguration, and until the Linux kernel is pageable with system-wide checkpoint restart and variable virtual page sizes, then there will be no substitute for proprietary UNIX. These are low-volume, but important segments of the UNIX space.
I agree with most of your comments. I believe what Solaris actually did is to bring a formerly proprietary but technically amazing OS to commodity hardware. AIX and HP-UX unfortunately chose to remain as niche products as Solaris was until recently. And AIX is actually pushed into this state by IBM’s policy of linux promotion at the low end. And IMHO every passing day they sound like they’re phasing AIX out in favor of Linux. I wonder why IBM do not try to extend AIX 6 to lower end instead.
Consider from a computer enthusiast’s point of view who is interested in running a solid UNIX based workstation at home for a reasonable price (<$3000). I don’t think Linux achieves this. Last time I checked its documentation was outclassed by Solaris and it had big compatibility issues if you decide to upgrade gcc/glibc/kernel. Anyone who has once worked on a proprietary UNIX workstation can tell Linux computers are no match for those beasts. UNIX running on commodity hardware has been nothing more than a poor guy’s consolation until Solaris came along.
And if you’re somewhat familiar with Linux development cycle and Linus’ philosophy, you’d know that Linux is the result of years of erratic hacking rather than a academically driven, well designed effort (check Linus’ comments on “Linux Not Designed; It Never Was” and Linus vs. Tanenbaum discussions). This is what sets Solaris apart for those of us who are computer enthusiasts: A great achievement in terms of design, engineering and involved computer science combined — provided for free.
I agree about AIX; why not do what HP have; an OpenVMS enthuiast kit – allow people to purchase a motherboard and processor from IBM (PPC970), direct, and provide a cheap, unsupported, non-commercial version of AIX for people to stuff around with.
But hey, this is IBM – they bitch that they don’t have the same depth and bredth as the SPARC community but they’re unwilling to FULLY open up their POWER ISA specifications; they scream they want the Power architecture to become more mainstream and yet they tell anyone who sells less units than 10,000 per month to piss off. They scream that they love Linux and yet not one of their client end applications are available for Linux – hello, where is Notes Client/Designer and SmartSuite – along with the mirade of other missing in action client end applications.
The sad part, however, is how suckered into the whole smoke and mirrors Linux advocates are; if FreeBSD had the same faboyism’s and momentum as Linux, they would have gone with that, heck, had Solaris been opensource 8 years ago, you would have seen IBM port it to the POWER architecture and used that instead of AIX or adopting Linux.
IBM is a fanboy company and it shows; their products have as little R&D invested into them as humanly possibly can, their hardware is over priced for what it can actually do, then lets not get started with the bureaucratic juggle gym that is trying to order a system without getting global services rammed down ones neck.
Mind you, look at the Ultra 20; its cheap, reliable an effective; and heck, it is VERY fast – everything one requires and it is 100% Solaris compatible; what more can one ask for, out of a computer?
I question whether IBM’s not porting Notes to Linux has anything at all to do with IBMs committment to Linux. While I suppose it might ease migration from Windows, I hope Notes dies, painfully. It may be that the version I’m using at work is not the most recent, but it’s horrible. Just horrible.
You can’t run HP-UX on a ProLiant since it requires either PA-RISC (HP 900/Superdome) or Itanium CPU (Integrity). HP/Compaq ProLiant machines use Intel Pentium and Xeon class CPU’s (we have a bunch of them at work running Windows and Linux). Check HP’s web site.
And whether pSeries is the “finest UNIX platform money can buy”, I think the jury is still out on that! And I have administered AIX on pSeries hardware, I just love “crazy 8’s”!
> Why on earth would anyone run closed/proprietary operating systems like HP-UX or AIX? When you can get open source choice from Solaris or Linux?
For legacy applications (or just because it is better for certain things).
Except at the very end where the guy just matter of factly said that it’s not a linux killer.
But Solaris seems to be a nice alternative to Linux or BSD, but probably needs a lot more drivers for desktop usage.
If someone would whip up a 1 cd Gnome-included iso with possibly whatever efforts they’ve got with ndis I would give it a try.
The summary of the article was taken from here:
http://www.linuxplanet.com/linuxplanet/reviews/
Isn’t there a difference between open standards, open source, and freeware, anyway?
Personally, I think ZFS sounds very cool. That, and zones.
Isn’t there a difference between open standards, open source, and freeware, anyway?
Yep, but both contain the word open, and that is definately enough to confuse most people
a distribution of solaris/opensolaris with hardware detection upon install and a package management system would make it a very attractive proposal
I’ve been using Solaris 10 for a several weeks now and my experience mostly matches up with the author’s. I have, however, hit a few sore spots that were particular to my needs. Maybe somebody from Sun (or opensolaris.org) is reading and can recommend these for the express pipeline?
* The SFW (sunfreeware) packages need to be packaged better. There are no x86-64 versions of the libraries, the apps are often compiled without many of the options enabled (readline, openssl, etc…).
* DTrace, while being the coolest thing ever(!), could be even better. Simple and efficient string matching and manipulation is difficult if not impossible. The cryptic error messages make the beginning of the learning curve nearly vertical.
* USB cdrom and floppy support during installation/rescue. Why should I need an internal cdrom drive in my storage server when the extra space could go completely towards drive bays? Cabinet space == money. It’s 2005, wake up and smell the low-cost peripherals.
* There are no supported PCI/PCI-X battery backed RAID cards that can be managed outside the BIOS interface(read: reboot). This appears true even for the Sun-repackaged LSI Logic 320-2X using the new lsimega driver.
* Installs and updates still suck. Sorry, but I use Debian and I know for a fact that superior installation and package management is within easy reach.
You write:
* DTrace, while being the coolest thing ever(!), could be even better. Simple and efficient string matching and manipulation is difficult if not impossible. The cryptic error messages make the beginning of the learning curve nearly vertical.
Absolutely agreed on string routines — which is why I added a bunch of the usual favorites (strstr(), strrchr(), substr(), index(), rindex(), strtok(), etc.) in Build 15 of Solaris 11. To get those now, download Solaris Express — or wait until Solaris 10 Update 1, to which we patched back these routines.
With regard to the error messages: a common source of error messages is the compiler, and I’ve never met a compiler that produced sensible error messages in all cases. We’ve tried to make the DTrace error messages sensible, but in some situations the compiler becomes so overwhelmed by the symptoms of a compile-time problem that it can’t get beyond them. (To see this phenomenon with your C compiler, open up a C file and randomly delete one line that contains a close-brace. Most compilers generate a hurricane of strange messages when you do this.) Anyway, if you ran into especially cryptic messages, it would be helpful to have the actual example(s) for which you’d like to see better messages.
Finally, in terms of the learning curve being “nearly vertical”: yes, there’s a lot to DTrace — but we’ve tried at every juncture to lower the barrier to entry. In particular, the three of us on the DTrace engineering team spent an enormous amount of time writing the DTrace documentation; if there are specific parts of that documentation that you found difficult, we’ll get them fixed as soon as we can.
More generally, head over to the DTrace community at opensolaris.org:
http://opensolaris.org/os/community/dtrace/
There you’ll find a large community (~500 people) with many willing to assist those new to DTrace…
Absolutely agreed on string routines — which is why I added a bunch of the usual favorites (strstr(), strrchr(), substr(), index(), rindex(), strtok(), etc.) in Build 15 of Solaris 11. To get those now, download Solaris Express — or wait until Solaris 10 Update 1, to which we patched back these routines.
Excellent, these will a great deal of ugly script.
The error messages that have been most confusing are not related to syntax. The one that I found most confusing yesterday was a “not enough registers” error. Other than attempting to understand the dtrace source, I had very little information to help me understand the mechanics behind the error.
The DTrace documentation is fairly readable, though I did find the text on the aggregations to be a bit vague. The cvs browser at opensolaris.org has been an amazing resource.
I’m still attempting to understand dtrace’s ddi_pathname. It appears to be unreleased/undocumented, but I know from your blog that it exists 🙂
Absolutely agreed on string routines — which is why I added a bunch of the usual favorites (strstr(), strrchr(), substr(), index(), rindex(), strtok(), etc.) in Build 15 of Solaris 11. To get those now, download Solaris Express — or wait until Solaris 10 Update 1, to which we patched back these routines.
It would be great to have a standard deviation accumulator to complement avg().
It’s zones baby. You get the client to buy the biggest commodity PC that runs Solaris, preferably an AMD x64 based one, and divide it up into task based zones. To the user it’s like he has a server room, he just doesn’t have to pay for it.
Just pick your box from the HCL, and be happy. Your customer will :-).
Why pay Red Hat?
Why pay Sun?
WTF is an AMD x64? Do you perhapes mean amd64 or x86-64?
“It’s zones baby. You get the client to buy the biggest commodity PC that runs Solaris, preferably an AMD x64 based one, and divide it up into task based zones. To the user it’s like he has a server room, he just doesn’t have to pay for it.”
Zones are nice, I agree, but they’re nothing special. It’s simple OS-level virtualization. You can get the same thing for Linux with Open Virtuozzo, although it is still maturing. IBM will be announcing their OS-level virtualization technology for AIX and Linux at the end of September. BSD has jails, but I don’t think it has the same scope as the rest.
Personally, I think the whole concept is overrated. People seem to think you get a manageability or utilization gain from OS-level virtualization. This isn’t true. All you really get is isolation. OS-level virtualization is a security feature, remember that. Host-level virtualization (Xen and other hypervisors, Vanderpool, Pacifica) is where the manageability, utilization, and availability gains are possible.
> Zones are nice, I agree, but they’re nothing special. > It’s simple OS-level virtualization.
> Personally, I think the whole concept is overrated.
> People seem to think you get a manageability or
> utilization gain from OS-level virtualization. This
> isn’t true. All you really get is isolation. OS-level
> virtualization is a security feature, remember that.
Strictly speaking, you’re correct that zones provides
the OS-level isolation. What I believe you’re missing
though is that Zones and the resource management
features in OpenSolaris (such as resource pools, the
fair share scheduler, resource controls) can be
combined to provide both better isolation and
utilization. Plus, this sort of integration doesn’t
require specify hardware or chip-level virtualization
as it works on single cpu x86 systems or multi-cpu
sparc servers (you can easily subdivide a single cpu’s
resources)
Check out some of the USENIX papers linked off of
http://www.opensolaris.org/os/community/zones/
for more details.
Not only in OpenSolaris. Resource management is a basic feature in Solaris.
First off Sun is all about PR, they hyped up zfs when it was basically vap-o-ware. If you’ve been paying attention to this Linux thing, it has become it’s own thing now. Linux has support for lots of various architectures, also good embedded support. You have tons of software also supported in your standard Linux/GNU build environment and there’s plenty of distros developing their own style. Too top it off there’s tons of people and companies already enjoying and depending upon Linux. OpenSolaris has a ways to go before it becomes “a Linux killer”, whatever that is…
“First off Sun is all about PR, they hyped up zfs when it was basically vap-o-ware.”
For filesystem design, the devil is in the details. I’m sure that ZFS looked like 12 person-years on paper, and that’s what management based their marketing on. However, in reality it is 12 py to design and implement the filesystem and 18 py to figure out why it isn’t performing as well as predicted. You gotta get the on-disk format nailed down, the kernelspace can be released as reference and tuned through beta testing. But once you get the reference kernelspace, there’s inevitably a point at which you realize the shortcomings of the on-disk format. It takes two iterations to get it right, and then you get to worry about the userspace commands…
Sun just realized the hard way that filesystems are tough. When you look back on the ZFS hype, there was a lot of hype all of sudden (they were close to a working kernelspace), and then there was the admission that it would be late (back to the drawing board). The hype died down to a trickle of interview responses declaring that ZFS is not dead, it’s just taking longer than expected. Microsoft did the same thing with WinFS, and it took 10 years to get from Reiser3 to Reiser4. Welcome to the world of filesystems.
The term ‘Linux Killer’ suggests that there is a desire to see “linux” beaten down, like in a boxing match. It is kind of funny to see this term bandied about with Solaris, as if Solaris is some sort of underdog, to be rooted for. People have or discover their favorite distributions, and often stick with them, or use multiple ones. Solaris is a great member of the distro family, but it is one , and only one, member. It will not kill Linux; Sun will be fortunate if Solaris continues to be a money winner for them.
The breadth of support for Linux distros has become a world wide phenomenon and people who live and make money working with their favorite distros are not going to abandon them for Solaris, for many reasons, it does not matter how functional Solaris is.
I would be able to license the source code for free to build my own Solaris based appliance.
I would be able to modify and resell Solaris.
I would be able to download and run a complete Solaris system from a liveCD for free.
I would be able to get Solaris for X86, X86-64, PPC, Alpha, Sparc, MIPS, ARM, PA-RISC, IA-64, S/390 and 680×0.
OpenSolaris is not Solaris, just like BSD is neither OSX nor Windows.
I would be able to license the source code for free to build my own Solaris based appliance.
Done. http://www.opensolaris.org/os/about/faq/general_faq/#cost
I would be able to modify and resell Solaris.
Done. (See above link)
I would be able to download and run a complete Solaris system from a liveCD for free.
Done. http://schillix.berlios.de/
I would be able to get Solaris for X86, X86-64, PPC, Alpha, Sparc, MIPS, ARM, PA-RISC, IA-64, S/390 and 680×0.
Ummm…
X86…Done.
X86-64…Done.
PPC… http://www.blastware.org/
Alpha…Possible, probably not enough interest yet
Sparc…Done. (Ummm, are you retarded, this is Sun we’re talking about!)
MIPS, ARM, PA-RISC, IA-64, S/390, 680×0…(See Alpha)
OpenSolaris is not Solaris, just like BSD is neither OSX nor Windows.
Ummm, OpenSolaris and Solaris 10 are distros of Solaris, just like RedHat is a distro of Linux (or GNU/Linux if you’re into that)
Go back to your bridge troll, OpenSolaris has just ensured that Solaris is here to stay, too bad HP, IBM, and SCO couldn’t have done the same.
OpenSolaris is not Solaris.
At least RHAS uses the same Linux kernel that’s in Slackware, SuSE, Debian, Fedora, etc.
It took Sun months to remove all the proprietary stuff from their kernel so they could release OpenSolaris. It is not the same code as Solaris.
A liveCD of OpenSolaris is not Solaris.
Being able to modify and sell OpenSolaris is not Solaris.
But whatever, just don’t bitch if your OpenSolaris based appliance doesn’t have the stability or performance of a Sun Solaris based appliance.
the truth is I’ll use whatever OS customers like. I hope these companies go after the little guy–the consumer rather than just enterprises.
Truth is enterprises will buy what their customers enjoy.
Sadly, that is often Microsoft Windows–even though many companies hate it. But now since we have lots of linux obsessed people out there we may have some competetition. But.. most of the Java people like solaris for some reason.. It seems a lot of the developers for sun REALLY PUSH the company, atleast they are doing it now. They used to think sun was going out of business and started liking IBM
I do not everythink that it is a linux killer. Both will definetly benefit of each other support by the communities.
OpenSolaris is a great initiative.
— Bouh
The author’s approach is flawed if the conclusion to his testing involves a desktop comparison to linux. I think that if Sun wants to equal BSD on the desktop while providing the premiere OS for commodity servers, that would serve them and customers just fine. Nothing comes as close to some Linux distros integration and control of pretty looking desktops.
The response by the Sun SE proves these grown folks are not concerned with hobbyist revolutions but rather technical excellence. When opensolaris was introduced, one would expect a lot of movement in both the Linux and BSD camps to scramble in NFS improvements. I use NetBSD on the desktop, so maybe it will happen one day, but I also realize that you can sometimes go down the wrong path and not be able to double back.
If you go to many corporations that use UNIX, you will find the suits use Windows and that has nothing to do with a real admin’s choice of what is used in the data center.
Look, Linux is perfect. It has a perfect virtualization infrastructure, it’s just not ready yet. It has perfect file system innovation, it’s just not ready yet. It scales, just not perfectly yet. It has mass device adoption, every device in the world can Linux perfectly, it just takes about a hundred monkeys typing shakespeare. Stop harshing on Linux, it’s perfect. Solaris is just a product, there’s no way it can compete against perfection. That’s what my HP sales guy says all the time.
That’s funny!
Is it true you can get Solaris (the real commercial distro) for free? And deploy/redistribute for free? Really?
“Is it true you can get Solaris (the real commercial distro) for free? ”
Yes. I use it as a commercial webhosting server. I asked my local Sun representative, if I need any licences, and he said no.
Also, security patches are offered for free. You have to pay only for non-security related patches.
I did. I downloaded it for free and set it up on an old PII 700MHz with 512MB of RAM. It installed and ran just fine. I had some screen resolution problems with the Xorg implementation but I switch to the SunX (or whatever it’s called) and everything works just fine.
Sun really is giving away Solaris. The only thing you pay for is support and, as I understand it, Solaris support is cheaper than Red Hat support.
Wow, looks like Sun fanatics think they can sell and redistribute Solaris for free..
Reading this thread you would think OpenSolaris and Solaris are derived from the exact same source code.
The TRUTH is Solaris is not OpenSolaris. Try to sell Solaris, not OpenSolaris, or redistribute it. I dare you Sun fanatics to do it. Prove me wrong.
I can get ALL the source code for Linux in RedHat Advanced Server and compile my own distro from srpms. Its actually kinda easy.
> Wow, looks like Sun fanatics think they can sell and
> redistribute Solaris for free..
Actuaally, they can sell and redistribute OpenSolaris
for free and I think the fanactics you speak of
are well aware of the differences between the two.
> Reading this thread you would think OpenSolaris and
> Solaris are derived from the exact same source code.
Again, I think most users are more aware of the
differences than you think but for clarification,
this is spelled out in the FAQs on opensolaris.org
The source code that goes into the next release of
Solaris which is code named “Nevada” and is being
published monthly as Solaris Express
http://www.sun.com/software/solaris/solaris-express/get.jsp
is the same source code that is on opensolaris.org
http://www.opensolaris.org/os/downloads/
with the exception of certain pieces which either
Sun does not have the right to distribute (someone
else owns the IP) or which Sun has not completed
verifying that it can release the source code. And
some of the pieces such as the X11 software or GNOME
are available from their respective open source homes.
> I can get ALL the source code for Linux in RedHat
> Advanced Server and compile my own distro from srpms.
> Its actually kinda easy.
And it’s the intention that it will be just as easy
to do that for OpenSolaris. This will take some time
but the quantity of freely redistributable source
code that is already available from opensolaris.org
is very large (and includes the whole kernel, and
most of the libraries and command line utilities)
I have installed solaris 9 on the PCs I own, solaris 10 does not run on any of them.
So good luck everyone. It is a big download.
is done by those people in the community who act like dogs: peeing against everything not their flavor. Simple example: I spoke to a guy last week who was looking for a job as an Linux/Unix administrator. We spoke about our platform (OS X, Debian, SuSe) and he went off: it all sucked. The only good distro for him is Gentoo… blahblahblah. He never even asked why we our administrators chose the lineup we now use, but he knew right away that our platform sucked.
It’s people like him who sit in a birthday party and hear someone telling about his first experiences with Linux and how he installed SuSe 9.3 (or whatever …) with even manually fixing a soundcard problem. Those Linux dogs immediately jump to the gun to tell to that absolute beginner how bad SuSe 9.3 is and how brilliant Gentoo is. They never realise that the average person sees his machine as a tool to do something with it: write letters, draw something, edit a movie, work on pictures … the average person doesn’t want to mess around with an OS … it’s the Linux dogs that in the end will save the day for Microsoft and Mac OS X.
Sad, but true. Needless to say: I offered the guy a job as cleaner, since I thought him to be unsuitable as *nix administrator. He declined.
I’ve used Solaris 8 on servers for years, used it as a workstation for a few of those too.
I recently looked at Solaris 10, and boy if you think Linux has 5 years until it hits the desktop, then Solaris has about 20!
Solaris 10 was a big disappointment – Gnome (aka Sun ‘Java’ Desktop) isn’t even the default, we still first boot into CDE.
There’s still no decent applications with the distribution – fancy coding in vi or dtpad, cos there ain’t no nedit/gedit/kate etc.
Moaning about no MP3 support in Fedora, Solaris doesn’t even have a decent audio player!
Unless you fancy spending a week surfing sunfreeware.com or compiling stuff off sourceforge, I’d give Solaris 10 a miss, it’s way to minimalist (even when you install everything on the DVD including the ‘extra applications’).
I still like the Sparc version’s stability for servers though, plus the sysadmins like it as they’re not Linux ‘certified’, and not using x86 hardware seems to be favoured by the money-holders.
“There’s still no decent applications with the distribution – fancy coding in vi or dtpad, cos there ain’t no nedit/gedit/kate etc. ”
uh, yes, there is. gedit is supplied in Solaris 10. So is a special Source Editor I forget the name of.
Just why is every editor after the Linux killer? Has suddenly Linux become bad or something? Doesn’t make sense… especially if the guys in here were looking for a Windows killer for years… suddenly they switch to Linux hunters… makes little sense to me. Could somebody explain the change and write some sensible reasons (apart from journalist psychology, that is)?
“has someone finally found a Linux killer”
What the fuck does this unconstructive bullshit have to do with anything? It’s not like developers are competing which kernel will take over the world. I say hail to anybody who contributes in making technology free to the people and ready to host nice open source application on nice open standars based platforms. I don’t wanna fucking kill anybody, get it?
Ad space: FSF, IBM, BSD, Genesi, Debian.
Probably not. But I don’t quite understand sun’s economic model.
Sunw is suffering as it is. Moving AMD-64 will mean even less money for hardware.
“Moaning about no MP3 support in Fedora, Solaris doesn’t even have a decent audio player”
Pardon me while I have a good laugh! Solaris is really oriented towards server applications. If you were playing MP3’s in my company I wuld fire your ass right quick.
I wuld fire your ass right quick.
Sorry, I only offer my services to people who can spell.
What people seem to be missing in all the OOOOHHHH BETTER THAN LINUX….is they author clearly states that on X86 hardware its got a long way to go..The situations where Solaris was better than linux, (for example his Gentoo box)..were situatiosn where he was comparing Solaris VS Linux on Sun hardware….thats about liek comparing Linux VS Mac OSX on a PPC machine…apples and oranges…NO big surprise to me that solaris works better on Sparc and Linux works better on x86…???
is there mailing list like LKML (in gmane for example) where I can read about Solaris developing .. ?
Solaris is UNIX based, GNU/Linux is nice for thinkers, students and programmers, but it is far away from quality standards as we can find in other commercial UNIX based operating systems.
That’s a really good idea.
I’ve just logged an RFE for this (CR 6325485) to deal with this.
In the next 24 hours you should be able to see this at http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6325485
Alan.