Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Unix Operating Systems Software

Sun Considers Releasing Solaris In Segments 87

Posted by emmett from the all-at-once dept.
VoidOfReality writes: "Check out this article on Infoworld about the open source release of Solaris 8. It seems Sun is running into some problems they forgot to think about when they initially announced the release." Hey, at least they're teaming up with Collab.Net to make it happen.
This discussion has been archived. No new comments can be posted.

Sun Considers Releasing Solaris In Chunks More

Sun Considers Releasing Solaris In Chunks

Comments Filter:
  • Sun is now busy removing all their previously GPLed source code that they stole from elsewhere. :) They're just dodging an GNU faux-paux. :)

  • Sun has released OSI approved code. (Score:1)

    by Anonymous Coward
    The Netbeans [netbeans.org] code that Sun just released was released under an OSI approved license. See the license [netbeans.org] if you want to see for yourself.

    This SPL is jsst like the MPL with only s/Mozilla/Sun/g and s/MPL/SPL/g. Take a look for yourself normal diff non-unified [netbeans.org]

    I think Sun is seeing their mistakes with SCSL. They seem to be doing a great job with the netbeans.org site.....maybe they are listening to collab net...or are listening to /.

  • I see quite a few complaints here about the fact that if you change the source code of Solaris you need to submit it for "evaluation". Every body seems to be forgetting that SUN still has to support, through traditional channels, even a free/open/whatever Solaris. Do you really think Linus would allow people to mess around with the source to Linux, and release the result as being Linux, if he had to personally organise the support? I don't think so. So while this may not be an OSS poster child, don't forget the reasoning as to why it would be getting done this way.

    Don't get me wrong, I'm all for the OSS model and the lines of code comment wasn't exactly a good thing to say (although I can appreciate where he is comming from... but that's a different story).

    -- David Smith
    C:\ is the root of all evil.
  • I can understand this. In my previous life as a network admin some of the comments in configuration files were nuts at best. Not the sorta stuff I would want John Q. Public reading.

    (Next on 'America's Funniest Code Comments', C.A. stays up until 5 in the morning, and...)
  • Ummm...
    Ultra II 360MHz SPECint95: 16.1 SPECfp95: 23.5
    Athlon 700MHz SPECint95: 31.7 SPECfp95: 24.0
    PIII 700MHz SPECint95: 33.0 SPECfp95: 30.4

    My numbers are from this doc [berkeley.edu] at the CPU Info Center [berkeley.edu]
  • The pitfall of large scale opensource development is that you get a product that has difficulty maintaining singular focus.

    Furthermore, there aren't a million eyes and hands working on the linux kernel, i'd say there are less than a thousand, but thats just a guess. Assuming everyone that uses linux codes is simply false. Kernel coding is even more false. Even so, you've already got Alan making his own patches that tend to be more anticipated than the "real" releases linus likes. Not to mention the gaping holes in linux where developers never had the inclination to implement those things they did not use/need.

    Linux is a good idea, and linux on sparc is even a good idea. For one thing, Sun loves to EOL cheap hardware. Linux will run on all sorts of handy sun equipment that Solaris no longer runs on.

    However, convincing sun to dump solaris would be 1) difficult 2) stupid. People that think linux can eliminate Solaris today either dont know much about solaris or are just plain zealots. (or both). A few of the people in the sparclinux community love to slander solaris all day long, but until dbri and Sunvideo are even _supported_, solaris is still better even as a _workstation_ OS, if you happen to have an SS10 or LX.

    Furthermore, for those machines having >1 CPU, solaris simply wins. The kernel granularity is better and the userland interface (solaris threads / pthreads) is better.

    Hopefully what will happen is the release of the solaris code will be sort of a cross-training event. There are some things in solaris that sun should definitely take care of, but doesn't because of their priorities. Likewise, there are some things solaris does really well that linux really out to take a look at. Hopefully both operating systems will improve. The zealots on both sides (there _are_ Sun zealots, beleive it or not :) will continue to FUD around no matter what happens.. meanwhile everyone else can benefit.

  • Wern't we promised the code to staroffice?
  • The commitment to offer the star office source code is still there in 5.2: See the faq at http://www.sun.com/pr oducts/staroffice/5.2/faqs-general.html#32 [sun.com].

    Partially quoted it reads

    1. Many believe that other Sun open initiatives, such as the Java platform, aren't really open, since Sun continues to control the intellectual property and evolution of the technology. Isn't this just another example of Sun disguising proprietary solutions as open technology?

      Sun is changing the rules and taking office software into the dot-com age. Significantly, Sun announced three initiatives:

      1. We will offer the StarOffice 5.2 binary code for free download to anyone

      2. We will publish the StarOffice 5.2 specifications

      3. We will offer the StarOffice source code

  • Re:BSD - SCSL (Score:3)

    by Kiwi ( 5214 ) on Thursday June 29, 2000 @09:02PM (#966851) Homepage Journal
    I find myself in the strange position of defending Solaris, being a long-time Linux user who is just starting to really learn Solaris at my current job, and still considers Linux superior in many ways.
    With Linux running on pretty much all of the commodity hardware these days (not as much as NetBSD, though, I think), I think it stands to become the standard Unix.
    I agree that Linux will eventually overtake solaris. However, there are a number of areas where Linux is still playing catch up compared to Solaris:
    • Solaris scales better. Mainly becuase its SMP performance is better than Linux's.
    • Solaris threads better. Last time I looked, Linux could not properly do core dumps with a multithreaded process.
    • Solaris has kernel crash dumps. Linux's kernel crash dumps [sgi.com] are not ready for prime time. In other words, you can find out exactly why Solaris crashed.
    • Excellent support. Just in case you are not a Solaris kernel guru, you can call up Sun and they can get someone who can tell you why your Solaris box crashed. Sort of like getting Linus Torvalds on the phone after your Linux kernel crashed.
    • Solaris has better NFS support. To put it mildly.
    • Solaris has a much bigger mindshare among the corporate suits. In fact, I had a hell of a time getting a job as a UNIX sys admin because Solaris people do not consider Linux sys admin experience real UNIX experience.
    • Solaris is well documented. [sun.com] With Linux, often times the source code is the only documentation you get.
    I notice that Solaris people often have had little or no exposure to Linux--a lot of people go to college see a lot of Solaris and very little Linux. I am amazed at the number of Solaris people who don't know their head form a hole in the ground when it comes to Linux.

    I believe that Linux will eventually overtake Solaris, especially with SGI, IBM, and soon SCO backing it, but Linux is not there. Yet.

    - Sam

  • Generally, we are better than everyone else because we can hold a coherent thought for more than a nanosecond.

    Heehee.

    Oops, looks like I just went over your comprehension limit again. Sorry.

    -Erik (with his tongue firmly planted in his cheek)

    I refuse to engage in a battle of wits with an unarmed opponent - Unknown

  • I don't think you are in the position to decree it "unacceptible" just because it doesn't meet your ideals. Sun has already said that Solaris is not "Open Source." They are simply opening the source. In better days, this would be considered open source, but in these days of fanatical OSS people, saying something is Open Source essentially means it is usable in other people's projects. They are not opening it for YOUR benifet. They are opening it for the benifet of Solaris users who can now work on making it better. This justifys them not opening it all at once because,
    A) They have legal problems which prevent them from opening the whole thing,
    B) They don't need to. They release the parts that they own and the ones that need working on.
    They are not opening this for the benifit of the OSS community, and they never made a claim they did. They are simply opening pieces to allow people who want to see the code (not steal it!) easier access to it.
  • first of all, if it's the size of it that's the issue, what's the big deal? anyone familiar with open source knows that teams will evolve around different pieces of it anyway. i don't see what difference staggering it like that makes. i think they are more afraid of their technology being embedded in other operating systems. which is the way it should be anyway!

    --
    J Perry Fecteau, 5-time Mr. Internet
    Ejercisio Perfecto [nai.net]: from Geek to GOD in WEEKS!
  • Just a warning, in case anyone actually read this; the corrections this guy says are in fact incorrect.
  • You know what I'd like to see from IRIX? The 3D pipe! Though DRI is as yet untested in major 3D applications, I'm pretty sure whatever SGI's got is 50X better. (Afterall, they've been working on it for a decade now.)
  • Thanks also for pointing out the fact that Linux + Mozilla + X + GNOME + Apache (which is comparable to Windows 2000 which includes IE and IIS) is just as bloated as Windows 2000. Do the math
    Kernel - 3 Million lines
    Mozilla - 3.5 Million lines.
    XF86 -1 12! million lines
    GNOME (including ORB, window manager, etc.) ~10 million.
    Add all this together and you get ~ 28 million lines. Not that far away from Win2K and its 33 million lines.
    In comparison BeOS has 1.5 million lines. True, it doesn't compare at all to the above two OSs in terms of features, but still. Notice something else. Linux is really selvete if you don't add in X (bloated) and GNOME (bloated.) I think Linux running with a simple windowing system/window manager (maybe embedded KDE?) would clock in at a managable 10 million lines or so.
  • You do realize that these days, Linux + X + GNOME is only slightly less bloated than Win2K? It is certainly more bloated than NT4. That's best case. The nominal case is Linux + X + Mozilla + KDE + GNOME. Why? Well, first, Mozilla because the 33M line count in Win2K includes IE. Second, GNOME because I use the GIMP. Third, KDE because I need KDevelop. Both these apps are critical, and unfortuneately, I often have to run them at the same time. With these loaded, Linux easily clocks in at 35+ million lines, being MORE bloated than W2K.
  • You didn't get kernel source to SunOS4.X, you got the source to a paramater file so you could tune things. You got to compile that. SunOS5.X tunes most things itself, and your interface to tunable paramaters are things like /etc/system and ndd instead of things like vi and cc.

    Any company running a 'Clean Room' must also prevent its employees from looking at GPL code - once you peek at it, you should be careful about "appropriating" the ideas of a GPL developer in your company's proprietary code.

    If you change the Solaris source, you give it back to Sun to do what they want with it. You're happy for them to be able to do this, because you read and accepted the conditions of access before you worked on the source code.

    Linux has become the lowest common denominator unix-like OS - it is very far from being the superior unix OS
  • People often read newest first, thus, when they got to yours, they've seen so much belly acheing about Sun (who, by the way is not claiming that Solaris is Open Source(TM), but simply that they are opening the source) that they decide to take out their agression on you.
  • You do realize that the Linux source tarball that contained every thing that the Win2K tarball contained (Kernel, Web server, file and print servers, GUI, window manager, desktop environment, advanced media APIs (a la ALSA)) would clock in at a LOT more than Window's 30 million lines?
  • Without that reason to buy their hardware, no one will. No software that only runs on their playform, no reason to buy Sun hardware.

    Redhat has a sparc port right? I think a few others do as well. All in all the Sun HW is some of the best in the business, and if Sun sees an opportunity to make more money by releasing source, and using it to sell more boxen, then they will.
  • If they are opening up the source for inspection via the SCSL it should still be viewed as a derivitive of "open source" even if it is not perfect compared with the BSD license or the GPL.

    -rt-

  • Interesting (Score:1)

    by Anonymous Coward
    This seems quite interesting. I'm going to be doing some summer work at Sun shortly and one of the things that I will be doing is advocating my beliefs concerning Free Software and compliant Operating Systems such as GNU/Linux. I personally like Sun very very much and I don't want to say too much here in case they misinterpret what I'm saying but basically:

    Sun make money through producing very nice hardware and "solutions". The Operating System side of things (from my viewpoint as a consumer) appears to be more of a "must have this to ship with our hardware" type thing. I'd like to convince some of the people that I meet that no matter how good your coders are, having potentially millions of people around the world contributing cannot be matched by any team and that releasing the source under a Free License would do them more good than harm - particularly in the "war" against certain other corporations.

    --
    You Know Who I am.
  • My biggest hope for this is that it will result in an x86 version of Solaris that will perform comparably to the sparc rev. I would dearly love to have a full-on Solaris box in my bedroom, but the cost of Sparc hardware is just out of this world and is justifiable only for corporate budgets.

    Solaris will never run as well on x86 hardware as SPARC because, well... it's x86 hardware! =) As a general rule, it's inherently worse than Sun's stuff.

    It's like those kit Fieros you see sometimes, all spiffed up to look like a Lamborghini or something. You can put a lot of time into making your Fiero look like a Lamborghini. You can put more time into making it go fast. But even with all that time, effort and money... it still won't be a Lamborghini. =)

    Besides... the lower end Sun workstations aren't at all that expensive these days. Pricier than your average econo-model, yes... but are they worth it? Hell ya.

    --

  • Why is this bad? (Score:3)

    by softsign ( 120322 ) on Thursday June 29, 2000 @08:10PM (#966866)
    So why is everyone so miffed anyways? If Linux is the better OS as everyone must know by now, why do we even need to see the Solaris source?

    Or is it conceivable, that maybe... just maybe, Sun has had one or two good ideas about how to implement Unix in the decade that they've been developing SunOS and Solaris.

    Why is it so fashionable to rip into Sun so much around here? Solaris has a kernel that's generations ahead of Linux, their name is synonymous with reliability and scalibility and it - along with DEC and HP-UX - pretty much held the fort for Unix while Linux was still gestating.

    Development doesn't occur in a vacuum. There isn't just a black and white wall dividing Linux users from the rest of the world. Isn't it possible that the Linux community benefits from the leagues of programmers/engineers who were first exposed to commercial Unix at work and decided to volunteer their expertise - developed using Solaris et al - to the Linux community?

    No, Sun isn't scared their marketshare will disappear the day after they release their source. I doubt if any Sun employee loses too much sleep thinking about whether or not Richard Stallman approves of the SCSL. Most importantly, I don't think anyone at Sun is going to break into tears at the hissy fits some people are throwing over this.

    Grow up and realize that Linux can still benefit greatly from understanding what makes Solaris. Whether or not Sun gives you the 10 million lines of code as a whole package or in logical segments really doesn't make a lick of difference. And you know it.

    --

  • I was going to moderate, but then... (Score:4)

    by trims ( 10010 ) on Thursday June 29, 2000 @08:19PM (#966867) Homepage

    ...I saw some of the massive ignorant egos strutting around, and couldn't just let it pass. So I post.

    :-)

    Look, people, Sun will never OpenSource Solaris. I do take issue with them (in particular, the marketing 'droids) initially toutting Solaris as being OpenSourced, but they have backed off this - you will note that everything in their current campaign talks about a "Free Source License" and similar terms. Yes, some people may confuse it with the Free Software movement, but face it people, 99% of people associate the word free with cost (as in beer), and not with libre (as in GNU).

    Also, look at what Sun is trying to target. Essentially, they make money on hardware and services for Solaris, and no money on the OS (even when they charge for it, it's insignificant in the scheme of things). By using the SCSL, there are the following benefits:

    • Sun gets a new revenue stream from people wanting to make use of customized Solaris (embeded OSes and such, primarily). By opening Solaris under the SCSL, Sun makes it alot easier for VARs to evaluate using Solaris as a base technology, and the VARs save lots of money on R&D costs.
    • Many people get to see how an industrial-grade OS is written (warts and niceties, both). Academics may very well use chunks of it for teaching purposes, and in general, you get a better-informed developer community. This helps both Sun and the Sun-oriented developer community.
    • Sun retains control over Solaris branding and standardization. All the problems Linux has over standardized distros and kernel levels is avoided. The *BSDs do the same thing - one organization has veto power over what goes into the OS.
    • You do get bug fixes. Yes, Sun doesn't have to incorporate them into the next release of Solaris, and redistribution is limited - but see below for some notes about redistribution of such fixes.
    • Sun retains it's Intellectual Rights. Yeah, I know we'd all love it if we could get our gubby mitts on some of the nice stuff in Solaris, but honestly, I can see why Sun isn't interested in it. There's no real payback for them that the SCSL can't provide. Remember, Sun's goals are different than the OpenSource movements.

    As a side note, please note a couple of things about distributing mods to Solaris 8 (and using Solaris code):

    1. You can redistribute diffs however you want. Context-diffs are a bit of a fuzzy area (are the context "included code"?), but certainly, output from normal diff solaris_file.h modsol_file.h is allowed. You don't have to go through Sun to distribute these changes. You own the copyright to the diff contexts, and can license it however you want (though it better be compatible with the end-user SCSL in order to be usable by others).
    2. Fixed software containing Solaris code must be redistributed by Sun. That is, if the Solaris grep has a bug, and you fix it, and want to distribute the whole thing, you have to submit it to Sun, who will then redistribute it on their "secure" web site. Sun doesn't own your mods - you still do. They can't use your mod in Solaris(tm) until they get permission (and possibly pay) you to do so. The way I read things, you could even conceivably charge for the fix without owing Sun royalties, though that may be wrong (I'd have to have a serious lawyer re-read the agreement). But it has to be redistributed by Sun, not you.

    Yes, the Free Solaris Source Code program isn't an OpenSource movement by Sun. It has it's uses, and for that I'm happy. It definitely is limited, but for those target markets, it's a Good Thing. Maybe someday they truly will Open Source Solaris, but only when Sun sees that the advantages for Sun to Open Source outweigh the benefits they get from SCSL.

    -Erik

  • I really don't get the attitude of the open source community. Don't take wrong I only run Linux at home and support it all the way. But the attitude of the community is what puzzles me.

    If open source software is so good then why are we bothered about solaris, Aix,HP-UX or Windows or whatever. Why do we care about the SCSL or what ever
    license. There are tonnes of GPL software out the and tonnes of developers. Why do we need closed sourced software to look at and improvre. Why not compete healthly instead of throwing mud at each other. Linux is this popular today because of companies like RedHat,Caldera and the such exist and make it usable by the average person. Companies need to make money as does the average open source developer. How many opensource developers are out there that are sitting at home and writting linux only software without making money elsewhere? People have to earn for a living.
    Companies that sell software cannot give awy all the control they have on thier product which they have maintained, developed and put together.

    Instead of all this useless quarrel why not make linux better than solaris with out looking at how Solaris is written and let Sun take Solaris where it wants. Let the OS win on its merit and not because it is open source or closed source.

    Ask Linus, ALan cox and all the open source developers to leave their jobs and develop linux full time, with out any stream of income from Linux,to make it better and that day all the
    companies will give you thier souce.
  • I seriously doubt a 100 meg tarball of source code would be any good right away. Its just too massive. Can you analyze the mozilla sources and tell me about it? Didn't think so.

    The reason is because bits of the code are copyrighted with other companies. Same reason you don't get open source tnt2 drivers, only binaries.

    Tell me, what would YOU do with the full source code to solaris? Be happy they are even considering it. Oh shit someone gave me a ferrari but the doors were locked! Guess I'll have to scrap it. Typical zealot, never happy with what you are given as a gift.
  • 100MB tarball of source.. that's a lot of source. Mozilla is 20MB, the kernel is 18MB. Linux has a LOT more drivers in the standard kernel that Solaris, and a LOT more supported architectures. 100MB of source is something along the lines of Windows's 30 million lines. The TNT2 drivers are proprietary because nVidia feels some childish need to protect the names of their hardware registers. It has nothing to do with other companies - free drivers existed previously, in fact. Although, they are facing a GPL violation for _not_ making them open.
  • One thing that needs to be addressed is providing common commands. Solaris and linux both have commands like 'ls', 'route', 'ifconfig', etc. , but the parameters for each are different enough to require looking up and re-learning the command.

    Perhaps this is due to all the GNU extensions that standard Linux tools have.
  • They don't have to be used SPARCs... for instance:

    Ultra 5 Model 360 Workstation with 4 memory slots, 3 PCI I/O slots, includes:

    • (1) 360MHz UltraSPARC-II CPU, 256K L2 cache
    • 64MB memory (2 x 32MB DIMMS)
    • (1) 8GB 5400RPM EIDE disk
    • Onboard PGX24 graphics
    • 32X CD-ROM
    • 1.44MB Floppy
    • Solaris 8 Installed
    • Warranty Upgrade to SunSpectrum Silver

    The price? Just over $2000. That's right... no missing zeroes there. =)

    or...

    Ultra 10 Model 440 Workstation with 4 memory slots, 4 PCI I/O slots, and 2 EIDE disk bays, includes:

    • (1) 440MHz UltraSPARC-II CPU, 2MB L2 cache
    • 256MB memory (2 x 128MB DIMMS)
    • (1) 9.1GB 7200RPM EIDE disk
    • Creator 3D graphics
    • 32X CD-ROM
    • 1.44MB Floppy
    • Solaris 7 Installed
    • 19-inch color monitor
    • Warranty Upgrade to SunSpectrum Silver

    $5000.

    Granted, you could buy "equivalent" PC hardware for much less. But this is Sun hardware. It works. Well. =) And you get their warranty and support.

    Check it out. Sun hardware [sun.com] doesn't have to cost a fortune.

    --

  • I'm a little curious as to how post number 21, which is chronologically the 4th post visible at +1, could be considered "Redundant".

    If the moderator who modded down my previous comment would kindly explain himself/herself in response to this message, I would be very appreciative. It is clear that I have much to learn about the moderation system.
  • They're starting next month by releasing all the '0's, then the following month all the '1's, then the '2's and so on. Should be ready by end 2025

    Rich

  • Why on earth...? If you're doing file operations across NFS, you don't want ANYTHING to be cached locally. You want changes committed in real-time so that you don't have race conditions between two systems doing I/O on the same file. If you don't commit a write operation across the network immediately, then you enter a race condition where the master copy and the copy the NFS client recognizes are not the same file. If the file is opened from the master or a different client before the cache is synced, then they'll have an old copy, even though the first client has already "written" a newer version. Bad ju-ju there - why on earth would you want that?
  • Ha! Solaris is bigger than Linus Torvalds' imagination!
  • Whether it's written by professionals or not, sometimes other companies do things so badly that you have to comment exactly WHY you are doing something the way you are.

    Our code is littered with comments like 'XXX product spec says it does blah, but is in fact full of crap. It appears to do foo, so we have to work around it like this'.

    Mentioning names and referenceing documents, often in a less than flattering way, is necassary to highlight where an impementation differs to the specification because the spec is wrong.

    I say, name and shame those companies that release useless spec.
  • A sparc10 has quite a bit of power actually.

    My "second" computer at work is a sparc10 (first is the G4) and I admit it has "zip" for the specs, but, really, the overall level of performance would drive me crazy if I used it as my home machine. The bottom line is that I'm looking to get a fourth machine (1. G3/350, 2. 9600/233, 3. PII/300 laptop) so, frankly, price is the biggest factore. The sad fact of life is that x86 land is where the cheap boxes lie and since I already have a bit of a mac habit I can't really afford to get embroiled with another hardware architecture that has a $2000 entry level...

    Oh, and as a Canadian I have to consider that killer exchange

  • Umm, let's start from the begining here.

    Once you peek at it, you're infected with Sun Intellectual Property and should be careful about "appropriating" their ideas in your own code. If you do change the Solaris source, you have to give it to Sun to lock in...

    Perhaps I read the article wrong, but what I got out of it was, you could modify the source but don't call it Solaris, not, you can modify the source but you have to give us those modifications."I am happy to give someone Solaris source code and let them do whatever they want with it, if they don't use the name 'Solaris' when they are done." Sun's vice president and general manager for Solaris Software.

    Also, it seems more like they are trying to open it up more so that people can write better programs for it, and allow for hardware manufactures to support Solaris better. This is a really good thing because one of the many reasons x86 Solaris isn't very good is its lack of hardware support.

    UltraSparc processors perform a little more poorly than current x86, PPC and Alpha processors

    True, the UltraSPARC II has poorer Integer and Floating point performance then the three processors that you mentioned. Of course I wouldn't buy a Dodge Ram if I really wanted a Viper. If I wanted a good 16 processor server, oh wait x86, PPC, and Alpha can't do that, whoops. FPU and INT performance isn't the end all to be all. Look at what SPARC stands for, Symetric Processor Architecture, it may not be MIPS but the E10K seems to be quite a hit. The UltraSPARC processor is not out there to be the sports car of processors, it is out there to build the infrastructure that is needed. Also, the UltraSPARC I/II processor architecture is 5 years old, and going to be replaced by the UltraSPARC III architecture, which on top of improving INT and FPU, will be able to scale up to 1024 processors, I want to see the standard P6 or even Wilmetta arch scale like that.

    Although Solaris may not be the best UNIX out there, one has to admire its ability to survive. Other, what some would call more fit to survive, UNICes like IRIX have folded. Right now Solaris is the most widely sold Commercial UNIX out there. With Solaris 8 being free, for 8 or less processor machines, and soon to be open, one should not expect Solaris or Sun to disappear any time soon. Probably the best feature of Solaris is its ablility to adapt.

  • Erp? They're sanitizing comments (Score:3)

    by Chairboy ( 88841 ) on Thursday June 29, 2000 @06:14PM (#966880) Homepage
    "open source is fine when you are talking about hundreds of lines of code"

    This is the rationale for not releasing it all at once.

    What about Linux, is it really so amazingly efficient that it only takes a few hundreds of lines of code? I don't think so!

    I imagine the real reason is that they want to check the code for comments that could get them in trouble. In our software, there are all sorts of comments about why he had to hack something or do something in a wacky way, and I don't imagine THOSE companies would be too wild about seeing their names in lights.

    Also, there are parts that describe exactly why we're doing something one way instead of another because of some customer, and there might be cracks about the specifics. If we have it in ours, it's almost certainly in theirs.

    Too bad, it would be interesting to read some of the more 'colorful' remarks in the source...
  • ...a Beowulf cluster of these! It would be a complete OS!

    --

  • This article says it will be released Open Source, but the Sun Community License is _not_ OSG compliant, and is especially not Free Software compliant.

    I wish the posters and article writers would be more careful with headlines.

    - Serge Wroclawski
  • Sun is more worried about the splintering of Solaris than anything else. I don't think Linus would like to see some one fork off the kernel do a piss poor job and call it linux.

    The key differences between linux and Solaris are :
    1. Linux is the kernel, Solaris is a complete distribution.(GNU tools make linux a complete OS).
    Therefore SOlris source is definitely larger and more complex than Linux(kernel). Even the Solaris kernel is huge.

    2. Solaris and its license are a money stream for some, not the only but some. Linux is not the only money stream for Linus, Alan Cox and many other developers. They make money and code Linux during the free time.

    3. Sun has customers, shareholders and an image to take care off. Linus is under no such obligation. If linux cause some damage because it crashed who do you blame. With Sun thay have to be careful that this doesn't happen.

    4. It is wrong for Sun to call Solaris open source. Similarly it is wrong for the open source community to expect people to give up thier income stream and property. If you have the resources(brain power and talent) build your product to be better.

    Techincally speaking.
    First to make it clear I am a linux advocate and use only linux primarily and will always. But there are a few things linux really needs to improve on. Being objective.

    Linux
    Great Multiplatform solution. Light and efficient.Free.designed for servers-targetted market everything(Laptops-Mainframes).Not very scalable to meet enterprise needs. Good for webservers, desktop and midrange servers.

    Linux is trying to be the jack of all trades and the master of none. It needs to heavily improve on usability(average user), scalability(Enterprise) and reliability. If it needs to be the best O.S.

    Solaris is amazingly reliable and scalable. It works great with the hardware it supports. It is a bit resource intensive. But so is Linux (Try running X on a pentium 75 Mhz with 32 mb ram).

    Linux is not going to eat into Suns market any time in the near future. So Sun is not scared.
    Linux has a long way to go and I hope it does on its own merits.

  • "One issue is getting it ready so that people can make sense of 5 [million] to 10 million lines of code. There are not many people who know what to do with 10 million lines of code. Freeware, open source is fine when you are talking about hundreds of lines of code," Anil Gadre, Sun's vice president and general manager for Solaris Software, told IDG News Service. "So one is an ease of use issue, and we have to try and make it friendly. The other thing we are finding out is that maybe people actually wanted certain parts and not the whole thing."

    You are wrong Anil, very wrong. There is a man who knows what to do with millions of lines of code, and has been doing so for the last 10 years. His name is Linus Torvalds.

    This whole paragraph outlines how Sun simply does not get the Open Source movement. Yes, they are trying. we have to give them that. But they really need to re-evaluate why they are doing it, and who they are catering to. This paragraph shows that, at this time, they just don't understand.

  • The implications of a free solaris are very interesting. For your reading pleasure,here [slashdot.org] is the link back to last years /. discussion on what a free solaris might mean.

  • Afraid of 10M lines of code? (Score:3)

    by bit ( 22507 ) on Thursday June 29, 2000 @07:03PM (#966886)
    "One issue is getting it ready so that people can make sense of 5 [million] to 10 million lines of code. There are not many people who know what to do with 10 million lines of code."

    I really don't think any participant of the obfuscated perl contest or anyone who can make sense of some particular pieces of free software can be afraid of that amount of code or more.

    This is just a lame excuse not to release it all. People will work with those pieces they need to work with anyway. Why not let them make the choice themselves?

  • BSD - SCSL (Score:4)

    by rothwell ( 204975 ) on Thursday June 29, 2000 @07:12PM (#966887) Homepage
    Funny how Sun's operating systems used to be open source. SunOS was BSD -- they even gave you a compiler and the kernel source. Bill Joy wrote the BSD license -- which later proved to his benefit when he founded Sun.

    Now the latest incarnation of their OS -- Solaris 8 or 9 or 2.9 or 3.14159256 or whatever -- is "opening" a little, ahtough SCSL is hardly an open source license. It's more of a way to contaminate other companies' "clean rooms" -- you can peek at it, but can't change it. Once you peek at it, you're infected with Sun Intellectual Property and should be careful about "appropriating" their ideas in your own code. If you do change the Solaris source, you have to give it to Sun to lock in a vault at the bottom of the atlantic ocean right next to the secret underwater illuminati bowling alley. Okay, so maybe they'll actually include changes in the next release or patch or whatever -- let's say they do -- you still won't own copyright or anything about your code. You have to get permission from Sun to distribute your changes. With GPL, you at least retain rights to what you do. With SCSL, you become unpaid employees of Sun Microsystems.

    I'm dogging on SCSL here, but there is at least one nice thing about it, if they ever actually release the source: people writing programs for Solaris can at least look in the code to see why the published APIs are acting all funny, or to see how to best interface with the Solaris VFS, etc.

    That's more than we can say for Microsoft. Not a whole lot more, but more.

    Solaris is a nice enough OS. I wish it had some Linux-type features, such as /etc/ld/so.conf. But it scales well, and these days the price is right (free for 8 processors). UltraSparc processors perform a little more poorly than current x86, PPC and Alpha processors, though, and Solaris x86 is kind of a joke.

    With Linux running on pretty much all of the commodity hardware these days (not as much as NetBSD, though, I think), I think it stands to become the standard Unix. Companies are probably more afraid to contribute to BSD systems than GPL ones, because competitors can snatch up their BSD-licensed code and use it against them via closed-source products. With GPL, companies cannot take the code private, so the original developers' IP is protected better.

    It will be nice to have Unix largely unified again. There will always be special versions -- that's part of the beauty of Unix. Cary's UniCOS and other variants for special hardware will probably exist for a long time, because they're designed to take advantage of specific hardware. But the alure of a single API -- write once, compile and run anywhere -- is very tempting. IBM's even making Linux available on its 390 machines. BSD includes linux emulation, as does Solaris.

    Good times!

  • For Reference (Score:1)

    by Anonymous Coward
    Linux Kernel (2.4.x) has 3 millions lines of code. Mozilla has 3.5 millions lines of code And XFree86 4.x has has around 12 millions line of code. That's "somewhat" more than the `hundred or so lines' quoted by Sun's spokesperson. Mabye they need to hire a more cluefull one.
  • ... this results in a better Solaris for x86.

    I admit it. I like Sun's stuff (boo, hiss and downmod now)... and let's not forget that they gave us a lot of cool and useful things (NFS anyone?).

    My biggest hope for this is that it will result in an x86 version of Solaris that will perform comparably to the sparc rev. I would dearly love to have a full-on Solaris box in my bedroom, but the cost of Sparc hardware is just out of this world and is justifiable only for corporate budgets.

    If sun gives the community some source and the community gives the world a Solaris that can run on my compaq Barbie/Hotwheels celeron box, I will dance in the streets (after rush hour of course)

  • Yes. The OS side of things is simply a tool for sun to use in providing solutions.

    Unix just happens to be the base they work from.

    People can rave all they want about how 'linux is better' or 'freebsd is better'.
    Well.. better for what?
    Better for learning/hacking on? sure.

    Can your little linux/x86 box that you are planning on buying come close to touching what an e10k can do? Hell no.. not even remotely close.
    My 350Mhz ultra-5 workstation probably crunches numbers better than a Ghz PIII.
  • Why should they be scared? Does linux or any open source OS scale to their 64 cpu E10k server? Nope. Last time I checked ultrapenguin was booting on a 16 cpu box, but thats a far far cry from being production ready. I have plenty of reasons to buy sun hardware.

    1. reliability

    How many IPX's and SS10's are being used as light mail and dns servers? How many 386's from the same era?

    2. scalability

    Can your Xeon box scale to 64 cpus? Redundant power supples? RAID support?

    3. support

    If you have a contract sun techs will be out that same day to diagnose problems. Now I know you are babbling about redhats support but its trivial at best. They even admitted its just for installation questions and not anything advanced.

    4. product lifespan

    Again, how many 386 linux boxes are being used as servers? My SS2 has been running fine since 1992. Can you say the same of your 486?

    I'll mention it again, parts of the source code are copyrighted from other companies. Thats why they must check everything carefully before its released.
  • You're absolutely right, I overstated myself with the "perform comparably" line. Let's change that to "perform passably"... we are talking about chips that take twice as many MHz and 3 times the power to do the same thing....

    Yes, there are used sparcs... but even a sparc10 is $400 and that's just got no power

  • I could care less if they open Solaris (although I don't argue that it would be a great move under a GPL not this "Gateway" crap), but my real concern and plea to Sun is - RELEASE JAVA!!! With C# peepin' it's head out of Microsoft's ugly ass, one really needs to ask themself. Do you want Microsoft to sponsor the next big Open Source Project? Sure if they drop there buisness practices, apologize for monoplizing the world, hindering technologies, and running people out of business.
  • A sparc10 has quite a bit of power actually. The 10 was originally built in I believe 92'? Lets see we had 2/386's? hmmm, you can put 2 dual hypersparc mbus modules in a 10 giving it 4 cpu's at if I recall somewhere around 150Mhz each... 4x150Mhz... That seems like an awful lot of power to me. I have a sun server with a dual hypersparc 125Mhz module in it and I can tell you it will blow any of my pc's away. So yeah older sun hardware even in the factory config still rocks. Go buy a ss10mp with dual sm51's and tell me that for two 50Mhz cpu's it isn't fast. 1MB cache per cpu is pretty damn good. Trust me!
  • Not to mention that Solaris threads are more efficient, due to the use of thread mapping vs. kernel scheduling for all threads.

    Solaris IS a better OS in many ways right now. It's less convenient to use, but runs better and is easier to debug, and has decent support, all like you said.

    However, it's not the better technology that wins, it's the more common technology.
  • The only point I was trying to make in the article submission was that Sun's reasons for initially deciding to open Solaris up may have been less than noble. Now that they actually have to follow through on their promise, they're finding out that there's a good deal of work involved in getting the source out there, and they might not make the deadline. Personally, I think Sun should never have put a deadline on the release of the code - releasing a code base that large is bound to have some problems or delays...

    Just my $.02
  • I didn't realize Sun would pay for and seek permission to distribute patches. Neat!
  • Yeah considering these benchmarks wouldn't mean anything in the real world. You can't compare sparc/x86 using some pathetic test like that. You have a lot bigger issues than your cpu, like bus, i/o throughput, etc...
  • Isn't there some funnyness with Solaris (it may be gone with the latest incarnations, I haven't kept up). I know sockets weren't the same, and I thought they didn't support pthreads or something... this is just some vague recollection, so it might all be different now.
  • 1. Sun boxes were always designed to be more I/O friendly than the x86 series. A 386 has really poor I/O processing, but I have several 386 boards that run just fine...

    2. Redundant N+1 power supplies can be found in a lot of mid-level x86 servers - not much to do with the CPU... Raid, either. The rest of the hardware, yes, and like I said before, Sun has architected their systems to perform well in the I/O sector, but have been further behind in processing power (less important for most blind servers).

    3. Excellent point. IBM, Sun, HP all offer far better support at the moment for their unix boxen than Redhat could for their OS box. Note the hardware/software (I hate to use this term) synergy, though. Everything from head to toe is by one company - easier troubleshooting. Less third party involvement and dirtying of the system.

    4. Commodity hardware with a short expected lifespan (1-3 years) versus server-grade hardware that is expected to have value over a longer time (5-8 years). No question about that, though...

    I don't think they are scared, I think they are trying to grab mindshare. IBM has been winning some big battles against Sun in both the marketplace (IBM passed Sun to lead in UNIX midrange sales this 1Q, and the much touted success of the RS/6000 S80 versus the E10k) and in mindshare ( A.Root, supercomputing, Linux support announcements, Transmeta announcements, etc...). Sun is trying to gain more in the mindshare of the open-source community and other tech-savvy people that might be making decisions. It's a tough marketplace, and Sun is trying to keep growing at a high rate.

    This is somewhat of a marketing move, but they also need to satisfy the other parties involved with the code they own... hence the restritive licensing and release in chunks bit.

    #include<std_disclaim.h> /*I may work for IBM, but I don't speak for them*/
  • Having worked at Microsoft, I can assure that Microsoft developers swear in their comments. My roommmate's group was releasing some example driver code. He was handed the task of removing any comments that said BUGBUG, XXX, TODO, or stuff like "HP sucks!"

  • This is at least better than what they had before, but it's still a pathetic attempt at truly open source, let alone free (speech) software. Woohoo Sun. Yay. We're all so proud of you.

    It's nice to see corporations like IBM heartily embrace open source/free software. It's pathetic to see Sun try to imitate.
  • All in all the Sun HW is some of the best in the business

    This was true for the Sparc line of machines, but from personal experience I can assure you that the Ultra workstation product line is pretty crappy:

    • the NIC is substandard (haven't seen a single Ultra 5 that can do 100MBit over a 90m CAT 5 cable)
    • the PGX24 video card isn't much better
    • the power supplies die easily when power spikes occur - HP Vectra PCs connected to the same grid continue to work fine. If the power supply isn't broken, you have to remove the power cable for 10 seconds and plug it back in before the machine wants to boot again

    Even the server product line is kinda weird:

    • The E450 has 3 power supplies but only 1 power chord (!). What do you mean, highly available?
    • The L280 6-tape DLT robot needs to have its drive replaced about every 2 months, not an impressive life span for a $10,000 server-class product
    • The A1000 hardware RAID box can be connected to 2 separate machines, but if the total length of the SCSI cables is over 1.2m the SCSI controller cuts the bus speed in half. This means you can't service one of the machines if it breaks without disconnecting the SCSI cables. This of course removes termination from the bus so the other machine is off-line too (sigh)

    Sun needs to get their act together, they are behaving more and more like box pushers.

    OTOH, the release of the source code would be a great step forward. Instead of waiting forever for Sun to fix a bug, we can send them the diff :).

  • From the Article: "I am happy to give someone Solaris source code and let them do whatever they want with it, if they don't use the name 'Solaris' when they are done," said Gadre. "

    This sounds a lot like they want to protect their brand name more than anything else. That's understandable, given the time and money spent to create the brand name. However, managing the brand name means controlling closely the product and those who use the product. If means placing the product into certain niches and portraying it in a certain light. That's fine for a closed-source, proprietary product.

    In order for open source Solaris to succeed, SUN has to be able to loosen the reins a bit and allow the community to take Solaris into places where the community feels it should go.

    Let me get this straight. Sun is opening up the source of Solaris, and saying to the world "do whatever you want to do with it, just don't label it Solaris", and you still whine?. Not even Bill Gates is that arrogant. Have you written Knuth already? He has the same condition for the source of TeX. And do contact Larry Wall to complain as well, as one of the options of the Artistic License is renaming the resulting binary if you take the source code and with it what you wish.

    SUN, if it really wants to release Solaris as open source, should require distributers to place their company names in front of it such that we'll see things as "SUN Solaris", "Red Hat Solaris" or "PPC Solaris" In that way we'll always be able to evaluate and distinguish between different releases.

    And that's to SUNs benefit how? It would be a severe disadvantage to SUN. They either have to give up the connection between Solaris and Sun (which from a marketing viewpoint would be incredibly stupid), or have to deal with all the problems introduced by outside coders (which would be incredibly stupid from both a marketing and technical viewpoint.)

    Solaris shines in some areas Linux hardly dares to dream about. SUN says "Here's the code. Study it. Learn from it. Use it. Just don't call the thing you use it in Solaris." And instead of grabbing the opportunity, you whine.

    Words fail to describe that attitude.

    -- Abigail

  • Learn to read before posting.

    No, I am not a zealot, despite your need to label me as such in order to prove your point.

    The Linux project is run by tens of thousands of programmers, hackers, and tinkerers (and spearheaded by a single man, whose name I have already mentioned). By not releasing the whole source code prevents this army of coders from accessing, modifying, auditing, or otherwise improving the Solaris code. That is the whole reason for releasing it, right?

    Being given scraps is not a gift. It is a joke, and a futile attempt by Sun to embrace open source as a viable business model.

    Like I said, read before you post.
  • Imagine the chaos that would be caused trying to release something as bloated as Windoze as Open Source. I wouldn't be surprised if M$ tried using SUNs arguement next time they're confronted by "set the code free" activism.

    Of course, Sun is only making excuses for the process taking so long, as opposed to using this explanation to completely scuttle the project. They, at the very least, deserve credit for that much.

  • Hopefully what will happen is the release of the solaris code will be sort of a cross-training event.

    Yes, indeed. I'm a bit shocked that so many people here diss SUN because SUNs plans for making the source available doesn't fit the tunnelvision people here have of open source. Even if SUN would say "here's the source code - but keep your hands in your pockets, you can only look", it's an incredible opportunity to be taken advantage off. Solaris blows Linux out of the water when it comes to scalability, reliability and handling of heavy load. Being able to look at the source of Solaris gives others (not just Linux coders, also BSD, and even Microsoft) the opportunity to study how SUN deals with scalability and reliability issues; the gained knowledge can then be applied in your favourite OS.

    -- Abigail

  • Can you read?

    "There are not many people who know what to do with 10 million lines of code"

    And you counter that by providing one (inappropriate) example of someone who can? If you are involved in a project from it's inception, you can make sense of it's '10 million lines of code'. But when it's dropped in your lap all at once, without any previous exposure?

    They do have a point...If they were in it simply for the buzz they'd make the source available and not give two damns about how useful you found it.

  • How open? (Score:3)

    by Erich ( 151 ) on Thursday June 29, 2000 @07:22PM (#966909) Homepage Journal
    Are we going to see NFS caching in Linux any time soon? It would make my life better...
  • I asked SGI this very same question, I was at a big linux conference and listened to some marketoid pipe SGI's linux support and how Linux is the future of SGI, blah blah blah. SGI decided to give a critical piece of IRIX to the community, XFS.. blah blah blah.

    During the Q/A session I asked if SGI is so bloody committed to Linux, instead of trying to pick and choose what you have the might benefit Linux, why not open up all of IRIX? "Well because we have secrets we don't want Sun to know."

    You have to look at this and see that no big company has made a major concession to Linux or the OSS/Free software community. IBM has put out a mailer and a filesystem and ported some stuff. SGI has released a filesystem and some clever hacks and a few other things. Nobody has stepped up to the plate and opened a full OS. No matter how much they say they support us they won't fully commit. Sun is just too stupid to know that so they say it, try to ride the Linux stock wave, and now they are pulling out. "geez, we know you've all developed a full OS with CORBA based graphical desktops and one of the most sophisticated UNIX kernels around but we don't think you can handle our 10million lines of code it's too big..."

    Mozilla, Linux, GNOME, KDE, all multimillion line projects. I imagine GIMP, emacs, and a bunch of others are in the multihundred thousand line range, maybe a million. LOC is a terribly metric, BTW. I also believe that it probably has a very complex build and a lot of the code is really ugly but it's still nothing that couldn't be dealt with. It's probably just the kernel itself that interests most of us and it's not 10million lines.

  • This might be an okay thing, but if not there is nothing Sun is doing that couldn't be solved by RMS or whoever confronting them and going "bock bock bock" and doing that hands-beside-the-body-like-a-chicken thing. Industry studies have shown that this strategy is almost always foolproof!

    Really though, I don't get this exactly. Are they just chickening out so that they can leave some key parts out or something, or are they really genuinely trying to make the code more usable for people who want it?

  • The source was officially released by Netscape on 31 March 1998.
    ---
  • Excellent support. Just in case you are not a Solaris kernel guru, you can call up Sun and they can get someone who can tell you why your Solaris box crashed. Sort of like getting Linus Torvalds on the phone after your Linux kernel crashed.

    Yes, but have you ever had them respond with anything other than "replace the CPU"?

    As for Linus; you can't get him on the phone, but you can damn sure send an email that will get response from the appropriate kernel developers, include Linus.

    --
  • 100 MB isnt that much for the whole operating system, in fact I'd say that is pretty light. Tru64 5.0 base source is about 270 MB.
    --
  • Bellandorf announced Collab.net at JavaOne...but hadn't SourceForge been around for a while then? Why splinter? Why not just concentrate on what's already there?

    So what's so great or different about Collab.net?
  • I really don't think any participant of the obfuscated perl contest or anyone who can make sense of some particular pieces of free software can be afraid of that amount of code or more.

    Um... I think you mean obfuscated C contest. I'm pretty sure Solaris was written in C, not Perl. :-)

  • I was up at Sun in Feb. for the platinum beta test for Solaris. At that point, here is what they said about the source code:

    • It is not "Open Source", they are just trying to make as much source available as possible as a service to Solaris users to aid in determining bugs.
    • The reason they are not immediately opening all of the source is that they have had 200+ subcontractors work on various pieces of the code. Some of these subcontractors are no longer in business so the legal issues pertaining to code release are hazy. Some of these subcontractors may not want source they developed published, and Sun is negoitating with those.

  • lets look at it in a little more detail... What do they have to worry about in opening the source to Solaris....

    • Loss of revenue Nope - Solaris is so close to free-as-in-beer already that in any situation but a huge corporate installation it cannot be a significant revenue stream when compared to their hardware.
    • Support issues Here there IS a valid concern. If you call Sun support and tell them you're running Solaris version whatever they need to know exactly what you mean by that. It also works the same way if you're doing your own level 1 & 2 support with a little help from sunsolve. If its all open source then they lose that baseline so its reasonable to require that all mods are redistributed by them only - if they dont do this they are going to take a much bigger hit in the "poor support" area than they ever will by not sticking to the letter of open source.
    • Protection of IP Releasing the source makes cleanrooming their features more difficult so if anything its easier to protect your IP in an open source environment.

    All in all I think they are getting as close to "true" open source as it is practical for them to get, and as such I applaud it. If open-source means organisations with a huge investment of time and effort cannot be reasonably practical and pragmatic about how they participate, then they just wont participate at all and I really dont think thats what we want at all.
    # human firmware exploit
    # Word will insert into your optic buffer
    # without bounds checking

  • Has anybody used Solaris x86 was a workstation machine? I was thinking about getting the Free Binary License, but was put of by the reports of Solaris 7 on intel being slow. However, I hear Solaris 8 is much faster on Intel, so would like to know how it performs. I'm not doing any server work, just playing around with it on the workstation.
  • sparc hardware sucks at number crunching. what youre looking for is I/O where sun can kick the pants off everyone. althons are roughly equivalent to sparcs at the same clock speed and since you cant get a 1Ghz sparc, my althon box will kick your ultras butt.
  • From the Article: "I am happy to give someone Solaris source code and let them do whatever they want with it, if they don't use the name 'Solaris' when they are done," said Gadre. "

    This sounds a lot like they want to protect their brand name more than anything else. That's understandable, given the time and money spent to create the brand name. However, managing the brand name means controlling closely the product and those who use the product. If means placing the product into certain niches and portraying it in a certain light. That's fine for a closed-source, proprietary product.

    In order for open source Solaris to succeed, SUN has to be able to loosen the reins a bit and allow the community to take Solaris into places where the community feels it should go. SUN's marketing people may feel that Solaris should be used in a certain environment - they may even have research that shows this. However, I don't think Linus Torvalds could have forseen Linux on IBM mainframes, Macs or PDAS. The public took Linux in directions no single person or mega-corporation could ever dream of or plan for. Sure there are problems with some ports to obscure platforms, but on the whole, no one would say Linux is a failure. Linux, the open source product, provides people with OPTIONS. If I don't like Win/MS-DOS, MacOS or even Solaris, I can opt to use Linux, instead. If Solaris is going to be released to the public, SUN should expect it in computers housed in coffee-cans and pizza boxes. It should expect it both in a Mac and a mainframe.

    Open source isn't a business - you can't make money off something you give away for free - it's a philosophy and for some people, a way of life. If SUN tries to understand open source from a business perspective, they'll only waste their time and come away more ignorant. SUN, if it really wants to release Solaris as open source, should require distributers to place their company names in front of it such that we'll see things as "SUN Solaris", "Red Hat Solaris" or "PPC Solaris" In that way we'll always be able to evaluate and distinguish between different releases.
  • Brian didn't announce Collab.net at JavaOne, he announced what Collab is doing with Sun vis-a-vis NetBeans at JavaOne.

    Collab.net, and what they are doing, has been public for a while now. See the Collab [collab.net] website for more info on what they do. Much different than SourceForge. In fact, from what I understand, Brian was putting this together much before SourceForge was a gleam in VA's eye.

  • Sorry, but no - I was out of town...

    Fawking Trolls! [slashdot.org]

  • Because they are scared (Score:3)

    by Mullen ( 14656 ) on Thursday June 29, 2000 @06:39PM (#966924)

    "One issue is getting it ready so that people can make sense of 5 [million] to 10 million lines of code. There are not many people who know what to do with 10 million lines of code. Freeware, open source is fine when you are talking about hundreds of lines of code,"


    Because they are scared.

    How much money do you think they have lost due to Open Source software? They put down qmail (Saw it on the mailing list), they "improve" sendmail and bind. However, they lose money everytime an admin decides to install Linux or a BSD.

    It is not only software they lose money on, but also hardware. Which is where the money is. Which is why they would slander Open Source software. Without that reason to buy their hardware, no one will. No software that only runs on their playform, no reason to buy Sun hardware.

    But I am wandering.

    We all know though, that is stupid of them to claim only a handful of people can read and understand the code in their OS (Which is true). But they forget that there are ALOT of us. Alot of people who are willing to take a piece and eat all the code they give us. Then we can take that code and improve on it.

    Not only improve their code, but also write better documentation.

    Plus, look at who is speaking for Sun. A person who makes alot of money off selling hardware. Not a programmer. Not hardware designer. Not an Open Source programmer.

  • Not talking about two R/W file handles - file locking should take care of that.

    I'm talking about changing a file on an NFS partition and having another client read the file. The changes might not be committed yet, depending on how tight your scheduling is.
  • Okay, you won't see this, but...

    NFS caching is mainly for reading. It's most helpful when I have large amounts of programs served by NFS, and want to cut down on network traffic.

    And, I mean, we currently have NFS caching, but only for memory. You can't assign 2G of hard disk space for caching stuff read off of NFS.

    The way read caching keeps consistency is using leases on the data... the server will contact the client if the data changes before the lease is up, and if the server can't reach the other client then the client is not assured correct data after the lease expires.

  • 1. Clever illustration. 2. Good style. 3. What the hell's up? Seriously.. WTF is up with that?
  • Linus himself has admitted he doesn't actually know every line of code in the Linux kernel. What he does have is a good sense of aesthetic architecture, a knack for delegating roles by carving up the project into little fiefdoms, and the earned respect of an entire community, among other qualities. Sun is having trouble acquiring these things, and it's little wonder they're taking the approach they've taken, though it should be noted that there's no reason to forestall a source release for want of people who understand it when releasing the source is the very event most necessary for such knowledge to arise.

Slashdot Top Deals

Two can Live as Cheaply as One for Half as Long. -- Howard Kandel

Close