Slashdot Mirror
Would Linux Survive if Solaris Was Free?
"If you look at what the Linux community is doing now, it has already been done by Sun. Solaris can do everything Linux can do, but better." This article at OsOpinion asks: "Would Linux survive if Solaris was free?" I wonder if Scott McNealy has ever asked himself that question - or if he will after reading this. An interesting thought, eh?
IIRC you can get an educational/home-use version of Solaris for merely the cost of shipping, at Sun's webpage. Granted, that's doesn't do any good for people looking for Enterprise Servers for their company, but it is free.
If you take the Amiga as an example: It had a huge following, and still does even though it is no longer being produced (Ok, it is, albeit in very small quantities by clone manufacturers). If Solaris was to suddenly be made free, I think something similar would happen - people would still flock around Linux because of it's almost cultish following. And the advantage here over the Amiga is that Linux would still be produced, and would be updated.
Linux is a very very strong contender in the Unix marketplace, and I really don't see it slipping if Solaris was free.
Just my 2p.
Micheal Whitmore doesn't say whether, by free, he means no-cost binaries or open source. Linux's key strength is not it's lack of cost to download, but rather the availablility of its source code. Without this key distinction the article is meaningless.
If we're talking about free as in open source, I guess Solaris' success would to a great extent depend on how maintainable the code was. After all, one of Linux' strengths is it's well documented source code that more or less anyone can hack into. If Solaris is less documented, or simply not as well laid-out codewise as Linux, I think that the majority of Linux hackers would stay with Linux.
But what the market would do is probably another question. Perhaps the "backed by a major corporation" bit is enough to make companies choose Solaris, but I'm not sure. I believe that, after all, many corporations switch because they've heard so much about this "Linux"-thingy, not because they believe Linux to be the best UNIX flavour.
So Linux would probably have quite a head-start, and I don't think the outcome is as clear as the article's author implies.
But it _would_ be great if Solaris was truly Open Source, with documentation and all. If not else, there would probably be a whole heap of security holes that would quickly be patched (and exploited). After all, there are still plenty of simple buffer-overruns in Solaris programs.
-- Soon we'll be sliding down the razorblade of life...
...but now we dont need it.
To echo some other comments: If solaris had been free (on x86), we wouldnt have needed Linux. But, now that there is Linux, there is no niche for a free Solaris to fill. Linux rocks on comparativly low-end hardware. I dont see anyone, even the biggest Linux-advocate, advocating running linux on the big iron (>4 processors, etc).
Sun makes great boxes to run your enterprise on. The hardware/software combo is great. But on a workstation? Solaris on a workstation may not be overkill, but it is certainly not that much different than linux.
These articles ("Linux would die if foo happened") all seem to miss an important point: Linux is not a finished product. It never will be. Linux is continuing to improve, and it will continually improve. There's no single critical company that can drop their support of the product when they want it to die. Linux will remain FS/OSS and it's hardly going to disappear overnight because a closed-source, closed-development OS from Sun is made freely availalbe.
Gates' Law: Every 18 months, the speed of software halves.
One of the strengths of Linux is that there is a huge developer base - it can evolve in hundreds of different ways as seen fit by the whim of a single developer. This is a strength and one that cannot be matched by a solitary vendor.
However, this amazing diversity also has a downside and that is that no-one can exert a unifying pressure on the development. If you compare Linux to FreeBSD, you will find BSD tends to have very good help files, it's packages are conveniently located at a central location and well categorised and listed, and the OS as a whole is consistent with it's layout, help files, and program defaults.
In my experience, Linux tends to be less well organised, less documented and less consistent than most other versions of *nix.
This is not to say one is better than the other, all I'm pointing out is that each has strengths that are opposites and that the money factor, while a component, is not by any means the key factor in determining which will survive.
Solaris is a pretty good OS and we use solaris i386 to develop stuff that targets actual Sun boxes. However, I don't rate the i386 version that highly, the cool thing about Linux is that it runs well on commodity hardware, and has large amounts of support for esoteric bits & pieces - ie the perfect hobbyists machine & good for a cheap server. Now solaris on the other hand is designed for and works best on Sun's own hardware, and is rock solid in this guise. Unfortunately the hardware is more expensive than commodity pc stuff, (it is built a lot better) - which makes it a lot less useful for hobbyists / people saving cash.
stty erase ^H
Except that:
A) Linux driver hackers aren't just going to give up and jump ship.
B) The odds of Solaris making a serious commitment to an open source development model are even smaller than the odds of them giving away Solaris.
Gates' Law: Every 18 months, the speed of software halves.
Solaris has its place. I still wouldn't run mission critical ECAD software on Linux but I would consider arming engineers with Linux boxes to log into the N processor Sun server running Solaris.
If Sun were to release Solaris under the GPL or BSD license tomorrow I think for the most part it would generate a big yawn in the community. Consider it this way: right now Solaris more or less is made for workstations running on SPARC processors. Intel processor support, at least the last time I looked, was just a best effort basis. A lot of the interesting features aren't even supported on Intel. The community would have to port these features into Solaris X86. Not everybody runs on Intel like processors though, some of us use DEC Alpha's, or PowerPC and so on.
The most economical thing to do, and the thing that would be most accepted in this community, would be to pillage the Solaris code base for its industrial strength features and roll them into Linux.
I think there was a golden opportunity to totally dominate the market about 5 years ago or so if all of the commercial UNIX vendors would've been willing to bury their collective hatchets in Microsoft's back. That opportunity was to improve Linux to support their best large system features and concentrating on designing hardware that best exploits those features. Of course any time I mentioned this to anybody from Sun at the time they basically laughed. Linux was and always will be a toy OS that hackers occasionaly boot into.
I think this move would've totally killed Windows NT and a lot of Microsofts credibility as well. SGI is realizing this now and so they're trying to go down this path now. Sun isn't in as precarious a position as SGI is and so they don't need to go down that path (yet)
It's shallow as it stand.
There is much more to the equation after all.
Is he also preaching the pros of close-source development associated with Solaris? (many still believe in this model)
The pros of old school many flavor of unix reflected by Sun and other unix vendors? (I assume safely there are a few pros left in this, though not likely)
The pros of having a mature unix that supports more high-end hardware perhaps?
And then he has to worry about the many pros associated with the polar opposite of unix diversity, or high-end hardware compatibility, or close-sourced development.
It's not an easy evaluation.
I would personally not try to answer all these questions myself. Since GNU/Linux is truely a moving target in many senses. If you have a raid card that works only with proprietary unixes, nt, and novell today--it could be accompanied by a GPLed device driver or great specification documentations tomorrow. And if this raid card is popular enough. Over-night it would see to those who use this card Linux is equivalent to a Sun box using the same raid card. Over-night. For many many diverse hardware--truely a moving target no one can track. If one even dares to claim it one should take their words with a grain of salt.
The best one can do is to ask a GNU/Linux vet (one who attempted to run production linux boxes since 1995), and ask her very specific questions (say specific to your computational needs or business problems) about what you are trying to do, where you want to go tomorrow, and the day after that. And try to catch up yourself to their level of expertises--which means patience and dedication.
Until then, my feeling is Solaris holds its own ground for certain customers. But it also holds back certain computer users (who use Solaris) at the same time. It depends on your circumstances.
For instance look at Java. When Sun came out with Java they had a simple threading model that they wanted people to use. You want to wait for some IO? Spawn a thread to make a blocking call for the IO. In some ways good, for instance this architecture removes the possibility of writing a lot of possible race conditions. However was it coincidence that it also uses lots of threads, and all of the other forms of Unix out there at the time could not handle large numbers of threads efficiently? How convenient to have a cross-platform language that coincidentally cannot be made to run as well on your main competitor's platforms without major modifications to the OS!
Sun has a history of these games. The current one is Java3D. They have a pretty nice spec for 3D graphics and vector math. There are two possible implementations - one is native (using the video card, etc for extremely good performance) and the other is in pure Java (for the molasses effect). Of course to get permission to even try and implement the native version for a platform you need Sun's permission - and they refuse to give it for Linux.
So if Solaris was made free, here is what I would open up that gift-horse's mouth and look for:
So yes, if Sun released Solaris free, I would almost certainly just stick with my Debian system. Yes, they do quality work. But Sun doesn't do anything that Sun is not the main beneficiary of, which is not unreasonable in and of itself but is unlikely to match my long-term road map. Linux (by a pleasant contrast) has no such hidden agenda to watch out for.
Sincerely,
Ben Tilly
My usual seat in the cluetrain is at A HREF="http://pub4.ezboard.com/biwethey.ht
I think this question is sort of funny as we recently received a Sun Solaris (SPARC) box and found the (factory pre-installed) software setup to basially blow chunks. Yes, Solaris may have better top-of-the-line performance and scalability for huge database servers, but the default software installation they deliver is (at least when compared to Linux) is incomplete and butt ugly. Let's see:
- no compilers shipped. This in my book is a cardinal sin for a UNIX environment
- default graphical environment is CDE. Yes, it's a standard but it's butt ugly and feels very slow.
- default graphical setup is very 80s looking (then again, plain X and Motif never were very pretty). Comprared to KDE or GNOME it looks pretty pathetic. Maybe it can be made to look better, but the default configuration is boring/ugly. For a desktop environment this will make or break your distribution/system.
- limited tool set. You really start to appreciate GNU/Linux once you're used to having nice little things like locate, perl, apache, PHP and other stuff installed by default.
- try running Intel Solaris on the same box you run Linux on. It is sssllloooowwww.
Solaris has it's place in the high-end server space. In terms of the desktop though, I don't think there's much of a contest anymore. After a few years of endless tinkering by the Linux hordes, Linux shines in this respect while Solaris increasingly seems like an example of how NOT to build a desktop machine. Sure, you could download and compile all the GNOME stuff, perl, the GNU utilities and make your solaris box a bit nicer to work with. But why bother when you can get a $2 Linux CD (or a free download) that outshines Solaris by far in a desktop environment. Comments like this make me wonder if McNealy actually ever sat behind a properly configured (modern) Linux distribution such as Red Hat, Mandrake or Suse (those being the ones I tried over the past year). I would choose a modern Linux distro over Solaris any day (for the desktop); not for ideological reasons (although those also come into play) but simply because Linux is such a nicer desktop environment and comes with a complete set of software.
Warning: Those used to my style of commenting on nearly an entire article in small quoted sections should find this to be nearly the same. What's worse, it's also chalk full of my strong opinions and is quite unedited (this post is way too long to edit.. I don't care if I look like a fool because of it). That said, read at your own risk. ;)
heh. You'd think that there hasn't been a GUI for Linux all this time after reading this article if you weren't previously in the know. I'm not sure how porting to the Merced can be something considered to have been done previously by every OS or what have you. Besides, if everything that had already been done on another OS (which basically means, if you balled up every feature from every OS, Linux would be just a little bloated, no? Talk about poor wording. What do they pay writers for these days, anyhow?
Sad to say, I don't consider myself an "Anti-Microsoft warrior". That's paramount to saying, "Once Microsoft is gone, Linux will have served its purpose and we can junk it in light of something that's actually good. We only need it for media hype to slay Microsoft and allow for a real OS to rise up." Being a proponent of Linux doesn't mean that your sole goal is wiping Microsoft off the face of the earth (it might not be a goal at all for many). It just means you like Linux, and enjoy using it. Perhaps others should check it out? If they don't like it, it's their loss. And whoever thinks it's easy to convince management to drop whatever they've got and use Linux is living in a lush living in a fantasy world where free beer flows quite freely.
Soo.. how many people here who use Linux do so because it's reliable and suits their needs, or because they want to be "cool"? Besides, why the hell would I want to stick my box in the basement ? (well, besides the simple fact that most houses in Texas don't even have a basement.. the ground isn't exactly all that.. soft.. around these parts)
Next comes those Solaris highlights..
Do I really need 64 processors? I mean, honestly? :) Sure, there are people who do, but I'm sure they could afford to pay for an expensive OS (I'm thinking they'd pay a lot more just for the hardware involved)
Um, I hope Intel never has a 64-bit SPARC. It would be rather unseemly to steal the trademark and architecture from another company. That seems to be more of a SPARC vs. Intel thing than a Solaris vs. Linux thing. Besides, aren't there already ports of Linux for SPARC? (and a wide variety of other architectures? do they just think we're stuck with Intel, or what?)
Linux, proud babysitter of the phone lines in two whole U.S. states. What, that kind of thing doesn't count?
Linux: has the support and backing of several major companies, and not all of them hype not yet mature technologies like Java when they first come out in order to make a buck based on media exposure alone.
Wow. I'm switching right now.
Um, and Linux doesn't have any software for it yet, right? heh.
I'm not sure, but didn't I read something about a port for Linux as well? Not that I keep up on office software.. That ends our Solaris highlights section..
Sun is about the last company I'd trust. Just because they want to carve up Microsoft's market share doesn't make them cool. I don't "like" that, I just think it's nifty that the vultures will continue to peck at one another while the real competition steams right on ahead. I can't get over how short-sighted that comment is. "Linux people should like that". Let me elaborate how much I "like" that: F@#$ Sun. Grr. ;)
NetBSD has been fully operational for quite some damn time (even when the Linux kernel was just an "infant"). And it's free. And it's still around. And Linux is still the one grabbing all of the media attention. By the way, someone care to remind me what Solaris is derived from? I seem to have forgotten.. =P
~ Kish
..so be gentle. It's too early to bother with being flamed. ;)
Wasn't the GNU Project started in like 1984 or some such? Even if Linus Torvalds had never written the Linux kernel, the Hurd would have been done by now (probably long before now since there would have been more of a point to developing the Hurd if we didn't already have Linux.. now the Hurd is pretty much just a pet project of the FSF that they started and since they started it, figure they might as well finish it).
And wasn't the original Linux kernel written around 1991? =P
~ Kish
Not to say that it's any different with Solaris, but the OS in question was not ScumOS, it was Solaris. ;)
~ Kish
Probably Linux would not see the widespread use it does, but I think some people will always want to tweak and twiddle with their own project just for the fun and learning experience. Writing a real OS kernel from scratch is Way Cool[tm]. :-P
On the other hand, maybe they would all have started working on the Hurd in a Solaris cross-compilation environment
Daniel
Hurry up and jump on the individualist bandwagon!
In what way..? Kernel merge? Doubtful. Real doubtful. Base tool set merge? Again, highly unlikely. Why don't we have a single distro, instead of several? There are many forks in the overall OS development of GNU/Linux (as opposed to forks in kernel development, since the "official" kernel implementation is overseen by Linus and co.). Complete OS merge? The most unlikely of all possible scenarios. I just don't see this happening.
Though perhaps you mean Solaris would try to embrace and extend Linux? That just seems weird. One thing I'll agree on, they wouldn't compete.. ;)
~ Kish
But that would never happen now, would it? But if Solaris were as completely open and unrestricted (not like Java or Mozilla, but like Linux itself), then I think yes, it would eventually kill off Linux. Remember, Solaris is already dual-platform (Sparc and x86), and further ports would certainly be done. And Solaris has perceptual advantages in the commercial market (it's Sun, it's "supported", it's well-established) that have kept Linux from growing even faster. I think enough people would "defect" from Linux to a truly free Solaris that the commercial focus would shift quickly.
Now, if you ask me if I'd switch or if people switching is a Good Thing, I'd say no. I'm happy on Linux, and I like what it represents in the computing world. Besides, Sun is in it for the money. Microsoft knows how to fight those kind of companies. That's why they don't know what to do about Linux.
- -Josh Turiel
-- Josh Turiel
"2. Do not eat iPod Shuffle."
NetBSD, FreeBSD, and OpenBSD are all free, fairly similar, mildly different, yet all thrive. And even though they've been "mature" longer than GNU/Linux, GNU/Linux thrives as well. I'm not so sure I'm all that scared of Solaris closing down the GNU/Linux market any more than I am of *BSD doing so, were it free beer or free speech.
~ Kish
Would Linux survive if Solaris was free? Of course it would. To suggest otherwise indicates a very poor understanding of what Linux is, and what it's good at.
Historically, Linux was the UNIX you could run on your PC - for free. It's ability to provide "serious computing" facilities on commodity hardware won it the hearts and minds battle a long time ago.
When I was at university we had rooms full of SPARCstations and similar kit. They opened up my eyes to what an open systems environment was capable of. Then there was X - for all it's clunkiness still based on a great architecture. The whole "it's more important to do it right than to do it quickly" philosophy which is found throughout the UNIX world - and which is still completely alien in the Windows world.
It was a revelation to me. And it came at a time when I was getting more and more frustrated with the limitations and costs of Windows 3.1 on my home PC. It crashed all the time. (Heh. We complain about NT crashing "all the time". Remember when "all the time" really was ALL the time?). You couldn't develop anything on it without spending a lot of money first. And I was a student - where would I get money?
So, when Linux hit us (in the form of Yggdrasil Linux 0.99pl13) almost every one of us CS students embraced it. Here was a free, cool, capable, stable (even then), platform that we could take home and do the same cool stuff on our home PCs that we had previously been doing on the X-tens-of-thousands-of-pounds SPARCstations. We could write C code for coursework. We could write little TCP servers and clients to our heart's content. We could write Xlib apps. And we could take them all back into university, put them on the Suns, and they would work!
It's difficult to express how significant that time was. The idea that you could run X at home now seems trivial, but back then it was a Big Thing. We're talking about students here - no money. Sure, UNIX for PCs was around in the form of things like SCO and Solaris 86, but they were expensive (VERY expensive). But Linux was free, and ran on my cheap 386sx20 with 2Mb just great.
It's no concidence, of course, that the people who discovered Linux at college back then are now graduated and starting to be in decision making positions inside companies just at the time that Linux is being taken more seriously by the commercial world.
The article's conclusion is based on some assumptions that don't seem to be right to me:
Most people would agree that the various BSDs are at least technically as good as Linux. But they are massively, hugely, enourmously less popular. So even if Solaris 86 was better than Linux, that wouldn't necessarily make a difference.
Not that Solaris 86 is better than Linux. Solaris SPARC is excellent and as robust a platform as you could hope for, but Solaris 86 I wouldn't touch with a bargepole. It simply isn't better than Linux. It has less hardware support, is less robust, has less software, and crashes more often. It is arguably more secure, in the sense that "broken" = "secure". Plus it eats resources like no other OS.
This all probably explains why people continue to choose Linux despite the fact that Solaris 86 *is* free to hobby users, as is Solaris SPARC. That's a good thing. But there's more to this issue than price.
There's the community for a start. There's the symbiosis that you get between developers and users. There's the complete lack of "us and them". There's the ever growing list of features that you can pick and choose at your own rate. There's even a healthy competition between distribution makers which is leading to improvements in installation and package support. There are thousands of applications, web pages, mailing lists, and people willing to help.
Partly, all of this is because Linux is popular. But partly, Linux is popular because of the community support. It works both ways - a nice positive feedback loop. One that just isn't there for Solaris 86.
So, nice try Michael, but try understanding what you're criticising next time.
While we're on the topic of spawning threads to do blocking read()s on sockets - this model is extremely inefficient for server processes. It is far more efficient to use non-blocking read()s and either select() or poll() when dealing with IO on a large number of file descriptors. When input is detected on one socket it can then be delegated to a worker thread from a (relatively small) thread pool. Because Java lacks any select() or poll()-like construct their server connections top out at around 1000 socket connections. Threads take up a lot of resources, and should be used sparingly if you want good performance and stability.
Solaris is already free, at least for personal use. You can download it from Sun. Sun did this to stem the tide of Linux -- and it failed miserably.
Also, on the same hardware Solaris is noticeably slower than Linux. In fact, I recently compared performance of my Ultra 5 (at work) and my K6/2-300 (at home). I did it in a simple minded way: I compiled GCC on both. My K6-2 started later and finished sooner -- I didn't actually measure the times, but it was around twice as fast.
It costs less than $500, the Ultra 5 costs around $3000. Bottom line is that in the low-end server/desktop market, Sun hardware just doesn't make any sense. Given that Solaris/x86 is not too hot (in my experience it is nowhere near as robust as Solaris/SPARX or Linux) why would we give up Linux?
Also, a lot of the advantage of Linux is that, instead of having to go out and get all the GNU tools to make a system useful after you load Solaris, it comes with them. Things like bash, GNU find, GNU grep are dramatically better than the equivalent bundled commands.
-- Slashdot sucks.
I own the free version of Solaris. I rushed out and bought it the first week it was released for free. I also own too many distributions of Linux. I've installed Solaris on about 8 machines at various jobs. Every time I've thought "What a poor imitation of Linux."
I have my own Web server. It runs Linux or FreeBSD (Which I also bought) depending on my mood. When I installed BSD I thought: "Wow! This is very usable, it reminds me of Linux!"
Solaris, on low end hardware (any Intel) is very slow compared to Linux. I haven't run Solaris on many Sun Workstations, so I can only hope that it is exponentially faster, but for me, what makes Linux so great is:
- Open kernel sources
- The ability to modify the kernel
- Hardware support (Especially Video, sadly lacking in Solaris)
- The community spirit
- The great software:
- Apache
- Gnome
- KDE
- Vi
- Emacs
- Gcc (Gotta have a compiler)
- Enlightenment
- The Gimp
- XFree86 in general
- PHP/MySQL
- Telneting
- A completely customizable OS
- Text files for modifying EVERYTHING
- The speed
- The great multitasking
- Samba
Now, it's true that practically everything on that list is doable under Solaris, in fact all of that software will easily compile and install under Solaris (Heck, I've done it!)But NONE of it is as nice or as integrated as it is in Linux. To me, Solaris is the NT of Unix, and Sun the Microsoft of Unix.
I like that Linux is developed by the community for the community. Same as the BSD's. For that reason, I am a total convert who will never give up my cherished platform.
I have deployed Linux as web servers into two environments, my own server, and one that was previously running IIS. In both cases, we fell under the category of being allowed to run the "Free" Solaris. In both cases we had access to NT, Linux and Solaris. In both cases we chose Linux. It had NOTHING to do with price.
If Solaris was OpenSource, MAYBE it would be a contender, but I doubt it.
Ben
http://moses.penguinpowered.com
In addition to all the other excellent rebuttals and points that others have made better than me, there is a big thing that the original author missed. Sun couldn't make Solaris Free if they wanted to. Does the author think that Sun actually wrote Solaris?
Sun licensed AT&T's System V Unix code, and incorporated it into their existing SunOS codebase (based on BSD). They then tweaked it a bit for better performance and features on their Sparc systems and called it Solaris. Since a good portion of Solaris code is licensed from AT&T, Sun couldn't Free it without AT&T's permission. Anyone who was working on BSD in the 4.3 days will realize how futile hoping for AT&T's permission to Free their source code is.
----
----
Open mind, insert foot.
A BMW is to a hand-built swamp buggy as Solaris is to Linux.
But you can drive a swamp buggy on roads, try driving a BMW through a swamp...
To put it rather bluntly, this article didn't have enough thought behind it to merit mention here. The author fails to address the fact that ideology was (and continues to be) more a driving factor in the development of and for Linux than simple economics. From the GNU toolset, whose developers take issue with existing ideas about intellectual property, to the assorted GUIs developed by those who feel that currently available user interfaces are fundamentally flawed, most Linux projects and components have more to do with doing things one's own way, unbeholden to anyone else's, and little or nothing to do with saving the odd dollar or two.
When Sun first started in '82, they aimed for the high-end workstation market. Now they still do. And they also make enterprise-class software and hardware solutions. I betcha the average Joe has no idea what Sun does.
Linux was started as a Unix clone for students who couldn't afford the high price tag that came with the Unix solutions offered by Sun/NExT/etc. Now those original students have grown up on Linux and are now working, they have no reason to move to another Unix platform for home/SOHO use even if it was free. The x86 platform IMO still offers the most bang for the buck. Not only that, a layperson can still have the multimedia capabilities of Win/Mac under Linux that Solaris still lacks.
All of this because Sun never intended to shoot for the home consumer/small-time developer.
"Microsoft is the epitome of innovation and product quality."
Sanity.html - Error 404 not found
The main factor is ease of maintenance of the software we use. When we bought a Solaris system it came with no compiler. No problem, install gcc. Of course emacs was missing. Install it too. We also need perl, pine, elm, etc. You get the idea. And then there is TeX which is really the reason we have this deparmental network.
With RedHat Linux once you do an install all these things are just there. These days we tend to buy systems with Linux pre-installed so we don't even have to do that. When we got a Solaris box we actually went out and got a consultant to install all the things above (oh, did I mention the latest updates to bind and sendmail). Keep in mind there are no rpm's here. We're talking compile and and install -- including a rational plan on where everything should go. This is a big job with 30 or 40 packages. Then there is the question of monitoring and installing security updates -- easy with rpms, but a horrible task if you have to track every package at its source.
The packaging, organization, and integration is what RedHat supplies us. That's why we pay full price for at least one of each of their releases.
The title of this article brings out the distinction between Linux and GNU/Linux. If Linux was just the kernel then Solaris might replace it. But Solaris doesn't come close to GNU/Linux. Did I mention Gnome and KDE?
If Sun were smart they would adopt the RedHat Package Manager and port all the standard things mentioned above to rpms for Solaris. Then they would be at least competitive with GNU/Linux.
August 1991 0.01 Linux (first release, not bootable)
December 1993 0.99pl14 Linux (usable)
December 1993 FreeBSD 1.0 (patches to encumbered "Net/2" 4.3BSDLite)
November 1994 FreeBSD 2.0 First truly open source
(no legal challenges) version
This according to http://www.freebsd.org/handbook/histor y.html
and http://www.redhat.com/mirrors/L DP/LDP/gs/node3.html
So, no, an open source version of BSD was being developed contemperaneously with Linux but not released in unemcumbered form until a little later. The fear of legal challenges probably kept some developers away in the early days, and there was certainly not even a gratis version of BSD for x86 when Linus started developing Linux.
--
"L'IT c'est moi!"
Good looking fonts. Ideally, those nice expensive postscript fonts that commercial unices have.
I HATE the way XFree86 fonts look - it's ugly. It hurts my eyes. It just doesn't look right. In fact, I dual boot BeOS and Linux, and I am running BeOS 90% of the time. The main reason is fonts. PLEASE somebody do something to make X look decent.
--
grappler
Vidi, Vici, Veni
Or how about, "Now that Windows is so stable, can Linux survive?" Or the old classic, "Have you stopped beating your wife?" This guy is supposed to have been in the computer industry for 12 years, yet he seems to be incapable of posing a simple question in a logically valid form. Anyway, what's the word "survive" doing in the same sentence with an OS that's been growing like yeast these past few years?
Other posts have already provided excellent technical rebuttals. I'd just like to point out some of the absurdities in this pathetic piece...
"I myself joined the Linux bandwagon in 1997." And what HAVE you been doing on the bandwagon since then? Sleeping like a baby? I don't believe this guy has ever used Linux for any serious work. "Solaris can do everything Linux can do, but better". Indeed. If you define "better" as "slower and with more pain to the poor SOB saddled with administering the pig", that would be a lot closer to the truth. Sure, there are some specific, mostly high-end, areas where Linux can't touch Solaris (yet!). But since when has that been "everything?" There's as much truth to this claim as "NT a better Unix than Unix"...
"If you look at what the Linux community is doing now, it has already been done by Sun." Really? Well if the geniuses at Sun had bothered to cover this ground properly in the first place, NT would be a real corpse today, instead of just smelling, tasing and feeling like one [joke courtesy of fortune(1)].
Some time ago, I had the bad luck to administer Solaris on a few SPARC boxes at a small research institution (astronomers). It was a pretty bleak three years. Quite fortunately, some burglars stole the SPARC 10, just as the other boxes became hopelessly obsolete. It was the best thing that ever happened to my career. I moved all the network services to Linux, and we have never looked back. Suddenly, I was a fraction-time admin, rather than a mostly full-time one, and I could finally write that PhD thesis I'd been putting off. If they hadn't stolen that SPARC, I'd still be locked in an eternal struggle with Slowaris instead of doing science.
From my involvement with Linux (since the days of 1.0.x) and commercial Unixen (besides Solaris, I have risked prolonged exposure to AIX, HP-UX, and SCO), I have this image in my head of various Unixen as dinosaurs. They're still big, strong and deadly. They also happen to be scaly, ugly (SCO's the ugliest of all!), clumsy, and totally unable to adapt. And there's this bunch of small, quick mammals (mammalian penguins?) scrambling around underfoot, and they seem to be beating the ugly idiots to all the choicy bits of food. And at the rate the penguings have been evolving lately, the dinosaurs may find themselves mounted at the Smithsonian a lot sooner than they ever expected.
In my other job (nobody makes a living doing science in Russia these days. There's always a second job), I've been doing some serious software development, mostly Air Traffic Control applications. We started with Russian airports, and have recently moved out to Europe. Initially we decided to gamble on Linux. And in the 1.0.x days, it was quite a gamble. ATC meant _very serious_ high-availability. So we set up dual boxes (one as hot-standby), and did the fallback/fallover stuff in the application software. (It worked beautifully. Somewhere out in Siberia, one of our systems is still cheerfully running 1.2.13. It's still the most stable system their airport has got. And the main reason they could afford it in the first place was the "cost" of GNU and Linux.) My boss kept rumbling about "time to move our stuff to a real Unix", but I managed to keep that idea sidetracked until it sort of died on its own somewhere around the time of the Oracle/Informix announcements. During our most recent installation this summer, I had great fun working side by side with some guys from Sweden who were delivering another system at this airport. Theirs was based on AIX. Bizzarely enough, they developed in Visual C++ under NT, then built under AIX. Talk about perversions... They were serfing on this well into the night. We were in and out of the place in two weeks, with all acceptance testing complete, which was a sort of a local record (the testing is very exhaustive and time-consuming); for all I know, the AIX/NT guys are still delivering theirs. Every time I looked over their shoulder, I could see the word "dinosaur" flash in my mind. When they looked over mine and saw DDD, they just went away, shoulders slumped. Compared to them, our Linux development environment (nothing more than a bunch of free software working _real well_ together), was like flying to crawling. But the best part was the look on one guy's face. He was the local engineer placed in charge of the Swedish system, the guy who would be responsible for running it once the developers went home. Here's how the look came about. I had a couple of hours to waste, so I slapped together a nice little "monitoring console" for the sysadmin's workstation. It was not in the customer's requirements, I just did it for fun. Nothing more complex than a few xosviews and xloads swallowed in a button bar. It turned out quite well, in that it looked cool, and was actually useful for keeping track of whether each machine (there were five) was running as intended (i.e., not running out of memory, or burning up CPU when it shouldn't). So just when I was demonstrating this new feature to our contact (the engineer assigned to maintain our system), the sysadmin for the Swedes' system wandered by. He spent some time drooling at the flashing xosview windows, and just then (perfect timing!), someone accidentally pulled the output signal cable in the back of the rack. The system initiated a voice notification (it monitors the signal), in a pleasant female voice. At this point the guy got this amazed/dreamy look on his face, then turned to our contact, and said, "You lucky bastard!"
Anyway, didn't mean to run on so. Original point was, I've done more than enough work in both environments, and there's few things that Solaris does better by any definition. It's rock solid and sophisticated, but it's also unwieldy and full of cruft. Did somebody compare it to a BMW here? BMWs are a joy to drive. This thing is more like an 18-wheeler. It won't die, but I believe that it will eventually be forced out to habitate exclusively where it really belongs: on high-end SMP SPARCcenters and the like, where Larry Ellison can generate more benchmarks to humiliate micros~1 (the mutant cockroach of my ecosystem concept). And the Whitmore piece is a sorry excuse for an article.
What would be really cool is if java had a callback mechanism that allowed for a function to be called when data became ready. This is one of the (few) cool things about NT.
-AP