I've been following the debate, and my take is entirely the opposite. Sometime in the near future, I'll have to go over the license and make sure my understanding is correct.
The SCSL is just a transparent attempt to steal PR from real open source efforts and put pressure on Microsoft.
In the beginning, this was entirely true. It was designed in a fashion that I don't believe this is true, but marketting and management are sometimes so evil and inept... (like, umm.. $300 million stolen from LLNL's NIF). But, Sun stopped doing that type of marketting and got its act together. The SCSL tries to merge one aspect of open source (not free software), in that developers can view the code, better understand the product/platform, and build on it. With Sun putting the SCSL on Solaris, this is the exact reason, developers can build ontop of Solaris far easier than on a closed platform, like Windows. That's a great benefit of Linux and free BSDs, correct?
For Solaris, the aspect that developers can take the code and add features isn't to useful, unless perhaps they wanted to take the OS and set it up for a special purpose application, like a computer in your car. For other things, such as Jini, java, and various hardware, being able to add features and such can be very benefical fr a product line. Sun gets royalties to make a profit, the developers make money on their product, and Sun's hold to the standard of certain things, like Java, means the market Sun deals with will grow, as will its platform. This license is for Sun's market to increase, but to make it easier for the developers.
That, IMO, is the essance. That is not to say Sun has some evils in the license, its grip is to strong. I think for a coprerate license, for someone like Sun, the goals above are very good, and do help its community. That doesn't do any good for the open source / free software community, but Bill Joy said they were different. Its using open source, in terms of the ability to see the code and hardware, and that's quite useful. I do think Sun could improve the license, but Sun has always liked to have control.
So, its not open source in how we view it. Its a better method than closed source, and if Sun had jumped on the open source scene, with the GPL or BSD, I wouldn't have believed it. Its not logical in their revenue for the most part. SGI seemed to do it only for a last shot... though IBM has been the best of the bunch.
Well, I already replied to one poster on the whole article bit, so I'm going to shoot a bit of a tangent off now. For a while now I've been trying to get support for a site that tries to educate people about licenses - no bias. When I say no bias, I mean that anything contriversal would have to be footnoted with research. Unfortunately, not to many people have been to eager to help, other than say.. sounds useful.
Obviously people disagree on things, from ignorance or real problems. Would it not be better to give more of a resource? Even here people are mixing up what FSF's and BSDL's license goals our and the SCSL's our.. which is why the emails went back and forth with RMS, ESR, and Bill Joy.
First off, when Bill Joy talks about "improving" Linux, it's not clear whether he is talking about a sold distribution or an internal one. The distinction DOES matter.
I believe it doesn't matter. The GPL does not allow modifactions to be made that would co-exist under a seperate license. A module is not part of the kernel, it is a peice of code that is called by the kernel, loaded into memory, etc. If you changed code in the kernel, that must be GPL'ed. The same is with other programs under the GPL. The GPL does not make a distinction between used externally and internally.
Secondly, he does not make it clear what kind of "improvements" he is referring to. Again, this matters.
Why should he limit it to a list of types? He might not make a complete list, etc. He is talking in a general sense, for a profit-driven change, that the SCSL is better for th companies because other than bug fixes, they can make capital off their new code. The SCSL requires compatability so that things are not forked. The Linux community fixes this by "ignoring" these changes, as ESR says. Sun has to deal with HP, MS, and others who would be quite happy to "break" Java or other Sun innovations.
Anything that can be compiled and inserted as modules (INCLUDING code which modifies the kernel, as it's running) is NOT covered by the Linux GPL. Thus, such improvements CAN be shipped as closed, proprietary binaries. There's NOTHING to stop you from doing that.
I'd be quite interested in your explanation of how this is true.
here's nothing in the GPL that is inherently "anti-sell".
Its only anti-sell in that if you GPL your product, why would some one pay thousands for it other than support, documentation, etc? They are not paying for R&D, or the code. That makes it a bit harder to gain capital, especially since others will happily distribute your code, and compete in support, documentation, etc. That is not saying you can't sell GPL'ed programs, but its not as profitable.
This guy says he worked on the BSD licence. So why not use that?
This guy? Where have you been? BSD is one of the founding figures of open source, as is FSF. Bill Joy made numerous improvements to BSD, which played a major role in UNIX in general (as AT&T UNIX later incorperated BSD innovations). BSD has always been an open source license, except in the eye's of FSF, which cannot dictate as OSI has approved the BSDL and the community approves OSI's definition of open source.
Sun doesn't use the BSDL because the SCSL has different goals, just like the GPL. Sun wants, IMO, this: 1. Developers to have access to the code for further understanding how to work ontop of Sun's products/innovations.
2. Developers can improve Sun's code and make a profit. This is only restricted in selling bug fixes.
3. Developers cannot create forks, in the essense of creating incompatable versions of the Java programming language, etc.
4. Sun can make royalties, a profit, while also bringing developers to working with Sun, and at times on Sun's platform.
The GPL does not do this. The BSDL does not do this. Sun's philosophy here is different than the GPL's or the BSDL's. Sun understands the open source movement, and is not jumping on for free development as many companies hope to do (ie, Netscape had hoped to increase market share by free development). Sun rather wants to target for free developement, but just developement based on its technologies. This would increase Sun's market, and thus the company's profits.
We DON'T need Yet Another Licence
We need just the number of usuable licenses that fit the goal. Look at BSD OSes. Free is targeted for Workstation/Servers on the i386 platform (targetted, not limmitted to). Net is for a BSD on all platforms, and Open is for a secure BSD on many platforms. BSDI is targetted for an i386 BSD server with corperate support. You could not have 1 BSD that could do al of these as well. Linux can do all of these adaquately, but that does not mean its the bet at security, as server performance, at desktop performance, etc. It does a good job, ok job, or lousy job in every catagory. So do BSD OSes. Each OS targets a different use, and thus is better than another at that area.
The BSD and GPL and SCSL target different things. Each are useful. SCSL is not open source in the OSI definition, but its goals are different than OSI's definition. Each are needed. Otherwise, they wouldn't be used.
How was the comment not refering to views of any community?
More anti-Linux dribble and crap from the FreeBSD zealots. Why dont you people stop with the jealousy, get off your knees and do something productive instead of the constant jabs at Linux ("bastard OS").
"anti-Linux dribbe," "crap from...," "constant jabs at Linux." Those lead me to believe that that AC things the FreeBSD, or more correctly the BSD community is attacking Linux. The article was refering to the black-and-white structure of the UNIX famility tree.
And of course, your right on the branches. I'm not to keen on the idea of calling Linux a "bastard OS," nor would I call it a new branch. Both titles could be accurate, but I'd refrain from using them. Quick question, was it System V or System VII that incorperated BSD features? I remember something about AT&T UNIX incorperating a good deal from BSD Unix...
hmm.. the BSD community wasn't quoted by saying that. Its the UNIX community, the article was saying consider Linux a "bastard OS." That's pretty true. Sun hasn't 'seen the light' and dumped its decrepit (hah!) OS in terms to Linux, neither has IBM. Those UNIXes are qutie good, and its a bit sad many people view UNIX dead, and only Linux (and BSD) as server OSs of choice. Sun, SCO, etc are only now starting to have some Linux support. How old and well creditted is FreeBSD Linux support? So, umm.. your wrong.
Its always a bit disturbing when almost every article on BSD gets Linux thrown in, whether just to say "its like linux" (which gets extremely annoying..), or comparisions. The reason for Linux's mentioning was alright, and that was to say Jordon and others want enough word out so BSD can follow, safely, behind the Linux trend. They don't want BSD to be overshadowed, or to become extreme.
I agree, though, its pretty bad few people recognize Linux and BSD as all open source. That's not a BSD side problem, its both. With the larger userbase for Linux, there's an idea that seems to float around that open source = GPL = 'free' software.. and the rest are just annoyances that 'ride' along with Linux. It should be a joint thrust, Linux people should advocate Linux, but not condem but even advocate BSD. Same with BSD people. All it takes is education, and the will to admit we're friends with simiar ideology. Friends can always disagree on a few points.. but still friends.
You have to be kidding. Why would this be a violation if MS is secretly, publicly, or not at all porting GPL code from a UNIX platform to the Windows CE platform? That doesn't mean they take the code and incorperate it in their work, it means they get more, and supposedly better development tools on one of their platforms. MS would do this secretly because they've accepted the role as anti-open source (which I think was forced on them, btw). If they 'embraced' open source in publicly by doing the porting publicly, wouldn't every slashdot zealot automaticly claim this proves open source (and likely I'd guess they'd mean GPL open source) is superior?
They're likely not doing the porting just for internal use, because they want others to develope for CE so its superior on embedded applications. MS already has development tools, but 3rd party groups may not want to pay for them, etc. Microsoft is merely using the tools at hand, and one of them is open source. The more choice for developers, and the more freedom, the faster and more successful a platform evolves. Thus, automaticly creating FUD (the propiganda that MS might be intentially breaking the GPL, though without any hint of proof, or an argument for it) is just useless.
Anyways, I assume MS will silently get the open source development tools ported and public, and do their best to make it look like it comes from a 3rd party. That, or they neglect the idea entirely.
With Ati-Vec.. ehh, "velocity engine" Dragon should port to the Mac. It would follow Apple's goals, and the CPU is perfect for it. Dragon now supports KMI SIMDs, which improve performance by a great deal (supposedly). The Alti-Vec instructions are far superior, and are not just marketting ploys. With them, and the far better quality of the PowerPC design, I'm sure a speech recognition program would be superb.
... course, a lot of us don't have Macs and want to use another OS anyways.
I found this old tidbit again, out of sheer luck looking for something else.. (today).. and forgot about it. Its a bit interesting, pretty old. Not sure what else to say.. just that anyone bothering to read the forum....
Yet, you must ask yourself, if it is wrong for MS to promote itself, then why is it right for Rick Moen (an obvious Linux Fan) to promote Linux. That's the same thing. Hell, this whole supposedly "journalistic" and "objective" web site is nothing more than Linux Community self promotion.
Well, I agree with the MS bashing. I view it as hypocritical. When I got a bit out of my mind, and asked why on SVLUG, along the attacks and such, I did get one (well, also Rick's first was good.. 2nd a bit brutal:^). I can't find the message, of course...:-) To sum it up in a far worse way than it was said, MS's web server never stops cranking out the FUD. The web page doesn't get tired, it just keeps going on and on. Because Microsoft is such a big entity, the Linux community (as of yet) can't just have a page do the same fashoin to counter MS's. So.. screaming and shouting and the rest of it are needed to counter. Until Microsoft stops, the Linux community can't... (of course, one can always say the Linux community started it.. but in return, MS started it by making poor software and doing various unethical business practices)
Now it's just a bunch of zealots trying to dominate the world and following some stupid dictator. I'd rather work with FreeBSD.
heh. well, I started thinking the same thing a while back. Which is basicly why I lost it a bit... BSD in general, seems calmer and more orionted towards coding and progress, while Linux/GPL seems bent on good code, but more importantly to get a real fat ego boost. The latter can over shadow the former.. I emphasized the seem because it may be judging the Linux community harshly. The BSD community doesn't generally scream and shout, while at least some of the Linux community does. That may be the wrong impression... Rick's good responce on that part...
But just remember... Rick's reply was not meant as an article, or some Slashdot post, or anything else. I was quite surprised ot see it on Slashdot... It was just a reply off the thread in SVLUG... nothing more. It wasn't an article, it wasn't meant to be anything I think that Rick expected so much responce from.
Learning by understanding their code (which is freely viewable), and then re-engineering it.. but.. that likely wouldn't work, because the developer is agreeing to certain conditions, and thus may make his code truly Sun's. That's why companies must inform developers that they cannot agree to such licenses or else the company may have to give Sun whatever code they create...
Also, Sun does better SMP both because Solaris is superior in that area, and Sparc is far better than x86 in MP configurations. It would surely help, in general. But, licenses are always a problem.
I would think so.. but Bill Joy seems to ignore open source, and has adopted the idea that you use open source when playing with the little guys, and when you need the big ones, you go to Sun. That's why they support Linux, Sun believes that if your done playing, you get Solaris. Same with PCs. From the few times I've read Bill Joy make remarks on open source in recent press, he seems to have forgotten that he pioneered it, along with FSF and others. He did much of the work on BSD.. and its sad Sun went System V, because it seemed entirely for advertising rather then users (as many clung to SunOS and disliked Solaris..)
I know. However, human nature makes that impossible. People will do something because they want to, or are forced to, or see something making it worthwhile. That's only logical, though not the best. Sure, I wouldn't want companies to use child labor (ie, Industrial Rev), but they saw money as the factor. Over time, enough workers protested to change that.. because it was in their interest not to have 16 or so hours a day, etc. A things can't come by saying.. I would hope man would be better.. because man only can when things lead up to it. The open source revolution is making open source a real option. In the 80s Sun never would have even used the SCSL, yet now they do. You keep on inching, and eventually you meet your goal.
In that way, I think RMS was before his time, but in his time he was to extreme.. yet he did change the way people viewed software. I'm almost afraid to compare RMS to Marx (and if I get flamed, you better have actually read Marx).. but both changed people's views, seemed to understand in better detail then fellow revolutionaries (I guess you might say BSD) around him, and outlined away towards an eventual utopia, if you will. Sheesh, I'm fearing for my life because I said that.:-) Better shutup now.
Re:How reliable are these larger drives?
on
IBMs 73Gig Drive
·
· Score: 1
I got the IBM UltraStar 9zx back when it was a pre-release.. The one problem was that it got way to hot, but it was IBM's first 10k rpm drive, and Seagate was definately in the lead in solving minor problems like that (they had worked the kinks out then, as they had unleashed their 2nd generation). The drive did die, sadly, but not horribly. It died because I was to slow finding a solution for heat, and even though I had, months later it gave up. It tried, failing to spin up. It was still in good enough shape that I saved all the data I needed... which is the good thing about networks/2nd hard drives.
IBM's tech support didn't even care about hearing my problem.. they just said, sure, well send you a new drive, it will be there in 3-7 days, and you have 2-3 weeks to get the defective drive to us. That was it... an having two 9gb UltraWide SCSI 10k rpm drives.. when both working.. you feel a bit bad about sending one back.:) Course, now with the 72/34 gbs.... (BTW.. people never noticed its Ultra3 / Fibre only. Hard to find a cheap (~$200) scsi card for that...)
Ok, when anyone does anything, its generally because they see an advantage in it. RMS created FSF due to his morals, Linus created Linux because he was interested in it, etc. Sun would embrace the Open Source movement if it saw an incentive for it. The article basicly says.. Sun is inept by openning its code, but doesn't give it out freely for the developers to tweak. I don't see the incentive, I generally see the same people saying "Linux rules.. we'll kill all other UNIXes (and OSes)." That's one way to solve the UNIX frgmented market, I guess.
Lets examine how SUN would make money (more money) off putting everything under the GPL or the BSDL. The Solaris/Sparc combination is highly used at ISPs because of stability and performance. Many of the same ISPs that use Sun boxes use FreeBSD on lower end PCs, because FreeBSD has great performance/stability, and can handle what a Sparc isn't needed for.
Sun made Solaris's code open, which IMO was to make it easier for developers to write on the Solaris/Sparc platform, rather than getting them to go find bugs and enhance the OS. I doubt that if Sun made the OS open source, that anything more than leeching of code would take place, just like the Slashdot forums were full of comments about trying to GPL open source BSDs (for good and down right sickening reasons) when the BSDL changed. This merely improves Linux (& BSD), but on a range of platforms. Linux gets stabler, and now Sun has to deal with competing with PCs running an OS closing in on Solaris's quality, rather then NT which many have gripes towards. That, IMO, would increase compitition, and as PCs are dominent, it would not create a bigger market for Sun hardware.
Lastly, if Sun did do that, its competetors (IBM, Compaq, Microsoft) could take the code too. That is, unless they use the GPL due to its restrictions, and thus taunt its competitors with access.. yet not without them opening up too. Instead, Sun is using the SCSL to help developers use its technologies, but Sun gets some revenue and doesn't see splintering. Sun never gave up its power over Java, and has bitterly fought Microsoft for trying to make it platform dependant. By opening its technologies under the SCSL, Sun also increases its market by not having increasesd compitition, but rather having more out there for customers.
It could easily be argued that SUN would help the community by opening up, but when you look at the bottom line, I have not yet seen a good argument on why Sun's revenue would sky rocket. If someone has an answer, I'd love to see it.
Seagate is part of Microsoft? I've never heard that one. lol.. unix friendly hard drives. Any IDE/SCSI drive that complies to the standard, with the controller supported, will work... And before I ask relavent questions, if you have problems theres always email support (ie freebsd-questions@freebsd.org) or newgroups.. even irc and perhaps a BSD UG in your area.
Since the drives new, my first responce (1024 cyulinder limit) is useless. I'm not quite sure what to ask.. just a dumb one, you did set an active partition, right?:^)
Why didn't you (or did you?) try the tricks on the Linux Laptop pages? There's a solution for the Inspiron 7000 on the LT chip. Of course.. I've got the new media-P (I think they ran out of stock, so mine was a roll over). I still need to play more with xfree to get the tricks listed to work... its not fun getting Caldera at Linux World and than seeing the nice, pretty KDE desktop mangled and unusable.
I've been tempted a few times to go download a demo of AccelX just because I followed the thread on the Inspiron / media-P off their site...
Now in defense of NT you say that NT is the old code base. Can't you MS Defenders at least make up your mind.
Well, I never would have called Linux 30 year old code, because that's not true.. Linux is UNIX-like, it doesn't contain any 30 year code in the kernel to my knowledge. Due to the model of Linux, which was my point, it changes a monthly, and thus is not old code, but a mixture of 90s code. (You should also note that BSD programmers like to say how stable and efficent their base code is, some which is quite old.. and Linus in Open Sources counters by saying old code isn't good.. code must always be dynamic. In many ways, both views are correct)
On the "you MS defenders" crack, I merely don't bias my OS because some guy on tv or where ever else said 'I hear its kinda cool.' I've used NT, Windows, DOS, Linux, FreeBSD, and Solaris (and others.. but some, like Domain/OS, were to painful to go into). I still like DOS's simplicity with adaquate power.. but that doesn't mean I dislike Solaris, BSD, etc.
Seems to me this is a perfect argument FOR Linux/Open Source, not against it.
When was I against? I never, ever, said that. If you looked in my postings (what.. 40 are on the current list), you'd see I'm not anti-Open Source. Rather, I merely thought, and still do, that the different models do give reason why you can't compare NT to Linux off the bat, and make obscure statements about how one is worse than the other. The both models have their benefits, and as everything is being networked, a large number of common people use computers, etc. the Open Source model will prevail. I do believe that the closed source model was useful before, simply because it was hard enough getting the system to work.. updating its base files monthly or such would be more painful.
I saw this too.. but I've heard of NetMax. They did FreeBSD way before linux, and only started linux fairly (meaning I believe near the beginning of this year) recently. Their setup is supposed to be quite nice. That's really all I know, but they've been in the game for a long time, and seem well respected.
I know this article is on Daily Daemon News, which many BSD readers go to. I wouldn't which board had the article first, but that doesn't mean OpenBSD people hunted it down for Slashdot.. they may have hunted it down for BSD people.
Also, I thought NT people where well-aware of how many friends they had. I think on slashdot, both NT and BSD have been given bad tones, with BSD getting some lenancy as its open source and one can't play the FUD games as easily. The post of the top 10 list showed a good number of NT users, and less bashing than one would think possible on slashdot. Maybe readers are growing up, or more likely, there's a whole mess of people linux posters (the FUDders & zealots) don't realise exist.
ahh, true. I forgot all about Fibre Channel. It would be nice if FW could take the middle ground between IDE and SCSI. SCSI is expensive and few make adaquate drives at a resonable price, and IDE is still to cpu intesive...
I also haven't seen any good IDE w/ scsi chip drives for a long time. Those were great for home systems, not much more and all/most of the benefits of scsi. Nowadays its to split.. its either IDE for cheap storage, or scsi for fast/reliable storage...
NT came before FAT32 and that is the reason it cant read it.
Exactly, I figured that was an obvious point I didn't have to make. However, the developement models are entirely different, which is why NT didn't adapt. SP are Service Packs.. they are bug fixes, not special enhancements or anything else. When Linux updates, they're patches, enhancements, etc. The two are different. For the model NT uses (in both design and development), updating the kernel every 6 months could cause conflicts, making the system weaker and unstable.
Both Linux and NT evolve at a given rate, the difference is that NT (like other products) releases its next version at one point, Linux is always released. Windows 2000, or NT5, has been in the works for years, back in '97 is was due out in the beginning of '98.. and now its been pushed back.
In your reference of comparing NT SP 4 to 2.0.36 Linux, you pretend that they are on the same cycle. If you compared NT5/w2k at that time period to linux, you would be showing two systems (both under development by programmers). NT4 is only bug fixed, and while things do change in SPs other than bugs, they are to minor to count. If I compared W2k to Linux, Linux would win from my various uses of both, and that is a real comparision of up to date code on both ends.
My point was that you cannot compare code bases from different time periods and say.. 'wow.. I'm surprised x is so horrible to y, where y just was released 15 days ago.. and x.. well, its 5 years old. But still.. x sucks because it doesn;t support this new z that came out 4 months ago.' But wait.. x v2 does support it, has been in the works.. your just neglecting a lot. When w2k comes out.. cry your heart out of how horrible it is to Linux, because that's really comparing codes. When people think of Linux today, 5 years ago, 5 years from now.. they think of it as one thing, because every week it improves by a small ammount, not every 2 years by a large one.
NT doesn't support Fat32, yet NT 5 beta1 did. In an extreme example.. your method of thinking is that the 8088 should be like the pentium (or athlon, or alpha, or...), and your bitching because Intel was so ignorant and lazy to implement all these things into the 8088, even though they either didn't exist, or weren't feasabe. But the 286, the 386, and so on came out and did make improvements. So the 8088 didn't have an MPU... later chips did. When Intel's working on a new chip, why should it go back to the old one and re-invent it to have all the aspects of the new one.. and then release both?
Ahh, good comment (well, cept for 2nd to last line.. *shrug*). Here's where I differ, which of course is my opinion.
I tend to include BSD work before the free Net/1 release because institutes were freely able to obtain the code, and work on it, others easily (though illegally) obtained BSD from UGs (as BSD was *very* popular), and companies were able to get the BSD work, while paying AT&T the license fees. Thus, I see it as inclusive, not exclusive.
As the average user could not obtain the code, the average user was not going to be writing the code. The people writing the code were from universities, as RMS left the MIT AI lab just so he didn't have any legal problems. These were the the programmers, and they could see the code and contribute code. Looking at the copyright, it begins 1979 (though I thought BSD began earlier), which predates FSF's founding.
Of course, your more in the right than I am. In the 80s, it was far different than it is today, so I tend to view BSD as open from the beginning, even though it was not.
Why don't you think about this a bit before you say it? Otherwise, it ignorant FUD.
First, NT can read Fat32 - a simply program for NT has existed for years. You do have to pay for write, but that's fine. Now, why does NT not support Fat32 natively? Well, lets see. NT came out in 1996, so it is 3 year old technology, plus bug fixes. Back in 1996, Linux didn't support NTFS, it didn't support Fat32, it didn't do a lot of things. I can't understand why people love to compare NT to Linux, while NT is an old code base (most of W2k, remember, is new code).
Yes, comparing the two is handy, because NT4 is the current NT - but NT's code evolves only in MS's labs for the next version, and new releases don't come out every three months, or the like. Compare the two where it counts, but don't compare Linux's new code to NT's old code. When w2k comes out, then compare on merits of the developers. If you want to compare the code and functionality of NT to Linux in such an obscure place, go load up a copy for Linux from 1996, and tell me which is superior, and who supports MS better.
Oh, and for NTFS. Aspects of NTFS are good for NT, and not for consumer windows. The reverse is true for Fat32. Supporting reading of NTFS on windows (the same guys with the fat32 reader make an NTFS dos reader) is only a security problem. If your at the machine locally.. you could grab a simple dos disket and read/write (with that could can read, slowly, and have to *pay* for write ability). Also, NT is still a networkable fs (it has permissions, shares, etc) and so reading it natively from windows 9x generally is worthless to users...
really, I can't go on and convince you because your giving me a headache. Basicly, if you say something, think about it. Don't make FUD and degrade something you have no clue about. You may sound smart to those who know nothing.. but you overly foolish to those with any sense at all.
Slashdot Mirror
I've been following the debate, and my take is entirely the opposite. Sometime in the near future, I'll have to go over the license and make sure my understanding is correct.
:)
The SCSL is just a transparent attempt to steal PR from real open source efforts and put pressure on Microsoft.
In the beginning, this was entirely true. It was designed in a fashion that I don't believe this is true, but marketting and management are sometimes so evil and inept... (like, umm.. $300 million stolen from LLNL's NIF). But, Sun stopped doing that type of marketting and got its act together. The SCSL tries to merge one aspect of open source (not free software), in that developers can view the code, better understand the product/platform, and build on it. With Sun putting the SCSL on Solaris, this is the exact reason, developers can build ontop of Solaris far easier than on a closed platform, like Windows. That's a great benefit of Linux and free BSDs, correct?
For Solaris, the aspect that developers can take the code and add features isn't to useful, unless perhaps they wanted to take the OS and set it up for a special purpose application, like a computer in your car. For other things, such as Jini, java, and various hardware, being able to add features and such can be very benefical fr a product line. Sun gets royalties to make a profit, the developers make money on their product, and Sun's hold to the standard of certain things, like Java, means the market Sun deals with will grow, as will its platform. This license is for Sun's market to increase, but to make it easier for the developers.
That, IMO, is the essance. That is not to say Sun has some evils in the license, its grip is to strong. I think for a coprerate license, for someone like Sun, the goals above are very good, and do help its community. That doesn't do any good for the open source / free software community, but Bill Joy said they were different. Its using open source, in terms of the ability to see the code and hardware, and that's quite useful. I do think Sun could improve the license, but Sun has always liked to have control.
So, its not open source in how we view it. Its a better method than closed source, and if Sun had jumped on the open source scene, with the GPL or BSD, I wouldn't have believed it. Its not logical in their revenue for the most part. SGI seemed to do it only for a last shot... though IBM has been the best of the bunch.
SVLUG member.. just now in chicago..
Well, I already replied to one poster on the whole article bit, so I'm going to shoot a bit of a tangent off now. For a while now I've been trying to get support for a site that tries to educate people about licenses - no bias. When I say no bias, I mean that anything contriversal would have to be footnoted with research. Unfortunately, not to many people have been to eager to help, other than say.. sounds useful.
Obviously people disagree on things, from ignorance or real problems. Would it not be better to give more of a resource? Even here people are mixing up what FSF's and BSDL's license goals our and the SCSL's our.. which is why the emails went back and forth with RMS, ESR, and Bill Joy.
If anyone wants to help, or other info, email me.
First off, when Bill Joy talks about "improving" Linux, it's not clear whether he is talking about a sold distribution or an internal one. The distinction DOES matter.
I believe it doesn't matter. The GPL does not allow modifactions to be made that would co-exist under a seperate license. A module is not part of the kernel, it is a peice of code that is called by the kernel, loaded into memory, etc. If you changed code in the kernel, that must be GPL'ed. The same is with other programs under the GPL. The GPL does not make a distinction between used externally and internally.
Secondly, he does not make it clear what kind of "improvements" he is referring to. Again, this matters.
Why should he limit it to a list of types? He might not make a complete list, etc. He is talking in a general sense, for a profit-driven change, that the SCSL is better for th companies because other than bug fixes, they can make capital off their new code. The SCSL requires compatability so that things are not forked. The Linux community fixes this by "ignoring" these changes, as ESR says. Sun has to deal with HP, MS, and others who would be quite happy to "break" Java or other Sun innovations.
Anything that can be compiled and inserted as modules (INCLUDING code which modifies the kernel, as it's running) is NOT covered by the Linux GPL. Thus, such improvements CAN be shipped as closed, proprietary binaries. There's NOTHING to stop you from doing that.
I'd be quite interested in your explanation of how this is true.
here's nothing in the GPL that is inherently "anti-sell".
Its only anti-sell in that if you GPL your product, why would some one pay thousands for it other than support, documentation, etc? They are not paying for R&D, or the code. That makes it a bit harder to gain capital, especially since others will happily distribute your code, and compete in support, documentation, etc. That is not saying you can't sell GPL'ed programs, but its not as profitable.
This guy says he worked on the BSD licence. So why not use that?
This guy? Where have you been? BSD is one of the founding figures of open source, as is FSF. Bill Joy made numerous improvements to BSD, which played a major role in UNIX in general (as AT&T UNIX later incorperated BSD innovations). BSD has always been an open source license, except in the eye's of FSF, which cannot dictate as OSI has approved the BSDL and the community approves OSI's definition of open source.
Sun doesn't use the BSDL because the SCSL has different goals, just like the GPL. Sun wants, IMO, this:
1. Developers to have access to the code for further understanding how to work ontop of Sun's products/innovations.
2. Developers can improve Sun's code and make a profit. This is only restricted in selling bug fixes.
3. Developers cannot create forks, in the essense of creating incompatable versions of the Java programming language, etc.
4. Sun can make royalties, a profit, while also bringing developers to working with Sun, and at times on Sun's platform.
The GPL does not do this. The BSDL does not do this. Sun's philosophy here is different than the GPL's or the BSDL's. Sun understands the open source movement, and is not jumping on for free development as many companies hope to do (ie, Netscape had hoped to increase market share by free development). Sun rather wants to target for free developement, but just developement based on its technologies. This would increase Sun's market, and thus the company's profits.
We DON'T need Yet Another Licence
We need just the number of usuable licenses that fit the goal. Look at BSD OSes. Free is targeted for Workstation/Servers on the i386 platform (targetted, not limmitted to). Net is for a BSD on all platforms, and Open is for a secure BSD on many platforms. BSDI is targetted for an i386 BSD server with corperate support. You could not have 1 BSD that could do al of these as well. Linux can do all of these adaquately, but that does not mean its the bet at security, as server performance, at desktop performance, etc. It does a good job, ok job, or lousy job in every catagory. So do BSD OSes. Each OS targets a different use, and thus is better than another at that area.
The BSD and GPL and SCSL target different things. Each are useful. SCSL is not open source in the OSI definition, but its goals are different than OSI's definition. Each are needed. Otherwise, they wouldn't be used.
How was the comment not refering to views of any community?
More anti-Linux dribble and crap from the FreeBSD zealots. Why dont you people stop with the jealousy, get off your knees and do something productive instead of the constant jabs at Linux ("bastard OS").
"anti-Linux dribbe," "crap from...," "constant jabs at Linux." Those lead me to believe that that AC things the FreeBSD, or more correctly the BSD community is attacking Linux. The article was refering to the black-and-white structure of the UNIX famility tree.
And of course, your right on the branches. I'm not to keen on the idea of calling Linux a "bastard OS," nor would I call it a new branch. Both titles could be accurate, but I'd refrain from using them. Quick question, was it System V or System VII that incorperated BSD features? I remember something about AT&T UNIX incorperating a good deal from BSD Unix...
hmm.. the BSD community wasn't quoted by saying that. Its the UNIX community, the article was saying consider Linux a "bastard OS." That's pretty true. Sun hasn't 'seen the light' and dumped its decrepit (hah!) OS in terms to Linux, neither has IBM. Those UNIXes are qutie good, and its a bit sad many people view UNIX dead, and only Linux (and BSD) as server OSs of choice. Sun, SCO, etc are only now starting to have some Linux support. How old and well creditted is FreeBSD Linux support? So, umm.. your wrong.
Its always a bit disturbing when almost every article on BSD gets Linux thrown in, whether just to say "its like linux" (which gets extremely annoying..), or comparisions. The reason for Linux's mentioning was alright, and that was to say Jordon and others want enough word out so BSD can follow, safely, behind the Linux trend. They don't want BSD to be overshadowed, or to become extreme.
I agree, though, its pretty bad few people recognize Linux and BSD as all open source. That's not a BSD side problem, its both. With the larger userbase for Linux, there's an idea that seems to float around that open source = GPL = 'free' software.. and the rest are just annoyances that 'ride' along with Linux. It should be a joint thrust, Linux people should advocate Linux, but not condem but even advocate BSD. Same with BSD people. All it takes is education, and the will to admit we're friends with simiar ideology. Friends can always disagree on a few points.. but still friends.
You have to be kidding. Why would this be a violation if MS is secretly, publicly, or not at all porting GPL code from a UNIX platform to the Windows CE platform? That doesn't mean they take the code and incorperate it in their work, it means they get more, and supposedly better development tools on one of their platforms. MS would do this secretly because they've accepted the role as anti-open source (which I think was forced on them, btw). If they 'embraced' open source in publicly by doing the porting publicly, wouldn't every slashdot zealot automaticly claim this proves open source (and likely I'd guess they'd mean GPL open source) is superior?
They're likely not doing the porting just for internal use, because they want others to develope for CE so its superior on embedded applications. MS already has development tools, but 3rd party groups may not want to pay for them, etc. Microsoft is merely using the tools at hand, and one of them is open source. The more choice for developers, and the more freedom, the faster and more successful a platform evolves. Thus, automaticly creating FUD (the propiganda that MS might be intentially breaking the GPL, though without any hint of proof, or an argument for it) is just useless.
Anyways, I assume MS will silently get the open source development tools ported and public, and do their best to make it look like it comes from a 3rd party. That, or they neglect the idea entirely.
With Ati-Vec.. ehh, "velocity engine" Dragon should port to the Mac. It would follow Apple's goals, and the CPU is perfect for it. Dragon now supports KMI SIMDs, which improve performance by a great deal (supposedly). The Alti-Vec instructions are far superior, and are not just marketting ploys. With them, and the far better quality of the PowerPC design, I'm sure a speech recognition program would be superb.
... course, a lot of us don't have Macs and want to use another OS anyways.
I found this old tidbit again, out of sheer luck looking for something else.. (today).. and forgot about it. Its a bit interesting, pretty old. Not sure what else to say.. just that anyone bothering to read the forum....
Well, I agree with the MS bashing. I view it as hypocritical. When I got a bit out of my mind, and asked why on SVLUG, along the attacks and such, I did get one (well, also Rick's first was good.. 2nd a bit brutal :^). I can't find the message, of course... :-) To sum it up in a far worse way than it was said, MS's web server never stops cranking out the FUD. The web page doesn't get tired, it just keeps going on and on. Because Microsoft is such a big entity, the Linux community (as of yet) can't just have a page do the same fashoin to counter MS's. So.. screaming and shouting and the rest of it are needed to counter. Until Microsoft stops, the Linux community can't... (of course, one can always say the Linux community started it.. but in return, MS started it by making poor software and doing various unethical business practices)
Now it's just a bunch of zealots trying to dominate the world and following some stupid dictator. I'd rather work with FreeBSD.
heh. well, I started thinking the same thing a while back. Which is basicly why I lost it a bit... BSD in general, seems calmer and more orionted towards coding and progress, while Linux/GPL seems bent on good code, but more importantly to get a real fat ego boost. The latter can over shadow the former.. I emphasized the seem because it may be judging the Linux community harshly. The BSD community doesn't generally scream and shout, while at least some of the Linux community does. That may be the wrong impression... Rick's good responce on that part...
But just remember... Rick's reply was not meant as an article, or some Slashdot post, or anything else. I was quite surprised ot see it on Slashdot... It was just a reply off the thread in SVLUG... nothing more. It wasn't an article, it wasn't meant to be anything I think that Rick expected so much responce from.
Learning by understanding their code (which is freely viewable), and then re-engineering it.. but.. that likely wouldn't work, because the developer is agreeing to certain conditions, and thus may make his code truly Sun's. That's why companies must inform developers that they cannot agree to such licenses or else the company may have to give Sun whatever code they create...
Also, Sun does better SMP both because Solaris is superior in that area, and Sparc is far better than x86 in MP configurations. It would surely help, in general. But, licenses are always a problem.
I would think so.. but Bill Joy seems to ignore open source, and has adopted the idea that you use open source when playing with the little guys, and when you need the big ones, you go to Sun. That's why they support Linux, Sun believes that if your done playing, you get Solaris. Same with PCs. From the few times I've read Bill Joy make remarks on open source in recent press, he seems to have forgotten that he pioneered it, along with FSF and others. He did much of the work on BSD.. and its sad Sun went System V, because it seemed entirely for advertising rather then users (as many clung to SunOS and disliked Solaris..)
I know. However, human nature makes that impossible. People will do something because they want to, or are forced to, or see something making it worthwhile. That's only logical, though not the best. Sure, I wouldn't want companies to use child labor (ie, Industrial Rev), but they saw money as the factor. Over time, enough workers protested to change that.. because it was in their interest not to have 16 or so hours a day, etc. A things can't come by saying.. I would hope man would be better.. because man only can when things lead up to it. The open source revolution is making open source a real option. In the 80s Sun never would have even used the SCSL, yet now they do. You keep on inching, and eventually you meet your goal.
:-) Better shutup now.
In that way, I think RMS was before his time, but in his time he was to extreme.. yet he did change the way people viewed software. I'm almost afraid to compare RMS to Marx (and if I get flamed, you better have actually read Marx).. but both changed people's views, seemed to understand in better detail then fellow revolutionaries (I guess you might say BSD) around him, and outlined away towards an eventual utopia, if you will. Sheesh, I'm fearing for my life because I said that.
I got the IBM UltraStar 9zx back when it was a pre-release.. The one problem was that it got way to hot, but it was IBM's first 10k rpm drive, and Seagate was definately in the lead in solving minor problems like that (they had worked the kinks out then, as they had unleashed their 2nd generation). The drive did die, sadly, but not horribly. It died because I was to slow finding a solution for heat, and even though I had, months later it gave up. It tried, failing to spin up. It was still in good enough shape that I saved all the data I needed... which is the good thing about networks/2nd hard drives.
:) Course, now with the 72/34 gbs.... (BTW.. people never noticed its Ultra3 / Fibre only. Hard to find a cheap (~$200) scsi card for that...)
IBM's tech support didn't even care about hearing my problem.. they just said, sure, well send you a new drive, it will be there in 3-7 days, and you have 2-3 weeks to get the defective drive to us. That was it... an having two 9gb UltraWide SCSI 10k rpm drives.. when both working.. you feel a bit bad about sending one back.
Ok, when anyone does anything, its generally because they see an advantage in it. RMS created FSF due to his morals, Linus created Linux because he was interested in it, etc. Sun would embrace the Open Source movement if it saw an incentive for it. The article basicly says.. Sun is inept by openning its code, but doesn't give it out freely for the developers to tweak. I don't see the incentive, I generally see the same people saying "Linux rules.. we'll kill all other UNIXes (and OSes)." That's one way to solve the UNIX frgmented market, I guess.
Lets examine how SUN would make money (more money) off putting everything under the GPL or the BSDL. The Solaris/Sparc combination is highly used at ISPs because of stability and performance. Many of the same ISPs that use Sun boxes use FreeBSD on lower end PCs, because FreeBSD has great performance/stability, and can handle what a Sparc isn't needed for.
Sun made Solaris's code open, which IMO was to make it easier for developers to write on the Solaris/Sparc platform, rather than getting them to go find bugs and enhance the OS. I doubt that if Sun made the OS open source, that anything more than leeching of code would take place, just like the Slashdot forums were full of comments about trying to GPL open source BSDs (for good and down right sickening reasons) when the BSDL changed. This merely improves Linux (& BSD), but on a range of platforms. Linux gets stabler, and now Sun has to deal with competing with PCs running an OS closing in on Solaris's quality, rather then NT which many have gripes towards. That, IMO, would increase compitition, and as PCs are dominent, it would not create a bigger market for Sun hardware.
Lastly, if Sun did do that, its competetors (IBM, Compaq, Microsoft) could take the code too. That is, unless they use the GPL due to its restrictions, and thus taunt its competitors with access.. yet not without them opening up too. Instead, Sun is using the SCSL to help developers use its technologies, but Sun gets some revenue and doesn't see splintering. Sun never gave up its power over Java, and has bitterly fought Microsoft for trying to make it platform dependant. By opening its technologies under the SCSL, Sun also increases its market by not having increasesd compitition, but rather having more out there for customers.
It could easily be argued that SUN would help the community by opening up, but when you look at the bottom line, I have not yet seen a good argument on why Sun's revenue would sky rocket. If someone has an answer, I'd love to see it.
Seagate is part of Microsoft? I've never heard that one. lol.. unix friendly hard drives. Any IDE/SCSI drive that complies to the standard, with the controller supported, will work... And before I ask relavent questions, if you have problems theres always email support (ie freebsd-questions@freebsd.org) or newgroups.. even irc and perhaps a BSD UG in your area.
:^)
Since the drives new, my first responce (1024 cyulinder limit) is useless. I'm not quite sure what to ask.. just a dumb one, you did set an active partition, right?
Why didn't you (or did you?) try the tricks on the Linux Laptop pages? There's a solution for the Inspiron 7000 on the LT chip. Of course.. I've got the new media-P (I think they ran out of stock, so mine was a roll over). I still need to play more with xfree to get the tricks listed to work... its not fun getting Caldera at Linux World and than seeing the nice, pretty KDE desktop mangled and unusable.
I've been tempted a few times to go download a demo of AccelX just because I followed the thread on the Inspiron / media-P off their site...
Now in defense of NT you say that NT is the old code base. Can't you MS Defenders at least make up your mind.
Well, I never would have called Linux 30 year old code, because that's not true.. Linux is UNIX-like, it doesn't contain any 30 year code in the kernel to my knowledge. Due to the model of Linux, which was my point, it changes a monthly, and thus is not old code, but a mixture of 90s code. (You should also note that BSD programmers like to say how stable and efficent their base code is, some which is quite old.. and Linus in Open Sources counters by saying old code isn't good.. code must always be dynamic. In many ways, both views are correct)
On the "you MS defenders" crack, I merely don't bias my OS because some guy on tv or where ever else said 'I hear its kinda cool.' I've used NT, Windows, DOS, Linux, FreeBSD, and Solaris (and others.. but some, like Domain/OS, were to painful to go into). I still like DOS's simplicity with adaquate power.. but that doesn't mean I dislike Solaris, BSD, etc.
Seems to me this is a perfect argument FOR Linux/Open Source, not against it.
When was I against? I never, ever, said that. If you looked in my postings (what.. 40 are on the current list), you'd see I'm not anti-Open Source. Rather, I merely thought, and still do, that the different models do give reason why you can't compare NT to Linux off the bat, and make obscure statements about how one is worse than the other. The both models have their benefits, and as everything is being networked, a large number of common people use computers, etc. the Open Source model will prevail. I do believe that the closed source model was useful before, simply because it was hard enough getting the system to work.. updating its base files monthly or such would be more painful.
I saw this too.. but I've heard of NetMax. They did FreeBSD way before linux, and only started linux fairly (meaning I believe near the beginning of this year) recently. Their setup is supposed to be quite nice. That's really all I know, but they've been in the game for a long time, and seem well respected.
I know this article is on Daily Daemon News, which many BSD readers go to. I wouldn't which board had the article first, but that doesn't mean OpenBSD people hunted it down for Slashdot.. they may have hunted it down for BSD people.
Also, I thought NT people where well-aware of how many friends they had. I think on slashdot, both NT and BSD have been given bad tones, with BSD getting some lenancy as its open source and one can't play the FUD games as easily. The post of the top 10 list showed a good number of NT users, and less bashing than one would think possible on slashdot. Maybe readers are growing up, or more likely, there's a whole mess of people linux posters (the FUDders & zealots) don't realise exist.
ahh, true. I forgot all about Fibre Channel. It would be nice if FW could take the middle ground between IDE and SCSI. SCSI is expensive and few make adaquate drives at a resonable price, and IDE is still to cpu intesive...
I also haven't seen any good IDE w/ scsi chip drives for a long time. Those were great for home systems, not much more and all/most of the benefits of scsi. Nowadays its to split.. its either IDE for cheap storage, or scsi for fast/reliable storage...
That's because everyone knows its just so damn cool we can just sit back and share a smile. :-)
NT came before FAT32 and that is the reason it cant read it.
Exactly, I figured that was an obvious point I didn't have to make. However, the developement models are entirely different, which is why NT didn't adapt. SP are Service Packs.. they are bug fixes, not special enhancements or anything else. When Linux updates, they're patches, enhancements, etc. The two are different. For the model NT uses (in both design and development), updating the kernel every 6 months could cause conflicts, making the system weaker and unstable.
Both Linux and NT evolve at a given rate, the difference is that NT (like other products) releases its next version at one point, Linux is always released. Windows 2000, or NT5, has been in the works for years, back in '97 is was due out in the beginning of '98.. and now its been pushed back.
In your reference of comparing NT SP 4 to 2.0.36 Linux, you pretend that they are on the same cycle. If you compared NT5/w2k at that time period to linux, you would be showing two systems (both under development by programmers). NT4 is only bug fixed, and while things do change in SPs other than bugs, they are to minor to count. If I compared W2k to Linux, Linux would win from my various uses of both, and that is a real comparision of up to date code on both ends.
My point was that you cannot compare code bases from different time periods and say.. 'wow.. I'm surprised x is so horrible to y, where y just was released 15 days ago.. and x.. well, its 5 years old. But still.. x sucks because it doesn;t support this new z that came out 4 months ago.' But wait.. x v2 does support it, has been in the works.. your just neglecting a lot. When w2k comes out.. cry your heart out of how horrible it is to Linux, because that's really comparing codes. When people think of Linux today, 5 years ago, 5 years from now.. they think of it as one thing, because every week it improves by a small ammount, not every 2 years by a large one.
NT doesn't support Fat32, yet NT 5 beta1 did. In an extreme example.. your method of thinking is that the 8088 should be like the pentium (or athlon, or alpha, or...), and your bitching because Intel was so ignorant and lazy to implement all these things into the 8088, even though they either didn't exist, or weren't feasabe. But the 286, the 386, and so on came out and did make improvements. So the 8088 didn't have an MPU... later chips did. When Intel's working on a new chip, why should it go back to the old one and re-invent it to have all the aspects of the new one.. and then release both?
Ahh, good comment (well, cept for 2nd to last line.. *shrug*). Here's where I differ, which of course is my opinion.
I tend to include BSD work before the free Net/1 release because institutes were freely able to obtain the code, and work on it, others easily (though illegally) obtained BSD from UGs (as BSD was *very* popular), and companies were able to get the BSD work, while paying AT&T the license fees. Thus, I see it as inclusive, not exclusive.
As the average user could not obtain the code, the average user was not going to be writing the code. The people writing the code were from universities, as RMS left the MIT AI lab just so he didn't have any legal problems. These were the the programmers, and they could see the code and contribute code. Looking at the copyright, it begins 1979 (though I thought BSD began earlier), which predates FSF's founding.
Of course, your more in the right than I am. In the 80s, it was far different than it is today, so I tend to view BSD as open from the beginning, even though it was not.
Why don't you think about this a bit before you say it? Otherwise, it ignorant FUD.
First, NT can read Fat32 - a simply program for NT has existed for years. You do have to pay for write, but that's fine. Now, why does NT not support Fat32 natively? Well, lets see. NT came out in 1996, so it is 3 year old technology, plus bug fixes. Back in 1996, Linux didn't support NTFS, it didn't support Fat32, it didn't do a lot of things. I can't understand why people love to compare NT to Linux, while NT is an old code base (most of W2k, remember, is new code).
Yes, comparing the two is handy, because NT4 is the current NT - but NT's code evolves only in MS's labs for the next version, and new releases don't come out every three months, or the like. Compare the two where it counts, but don't compare Linux's new code to NT's old code. When w2k comes out, then compare on merits of the developers. If you want to compare the code and functionality of NT to Linux in such an obscure place, go load up a copy for Linux from 1996, and tell me which is superior, and who supports MS better.
Oh, and for NTFS. Aspects of NTFS are good for NT, and not for consumer windows. The reverse is true for Fat32. Supporting reading of NTFS on windows (the same guys with the fat32 reader make an NTFS dos reader) is only a security problem. If your at the machine locally.. you could grab a simple dos disket and read/write (with that could can read, slowly, and have to *pay* for write ability). Also, NT is still a networkable fs (it has permissions, shares, etc) and so reading it natively from windows 9x generally is worthless to users...
really, I can't go on and convince you because your giving me a headache. Basicly, if you say something, think about it. Don't make FUD and degrade something you have no clue about. You may sound smart to those who know nothing.. but you overly foolish to those with any sense at all.