Given that media PCs have been poor sellers for a decade, now, Microsoft may be trying to _force_ media PC market penetration via the XBox. They would have to do this to compete against PVRs, for example.
So...what would the kids say if daddy is recording American Idol when they want to play? Or what about when mommy wants to get some crystal off of ebay?
Putting licensing peeing contests aside, Solaris is UNIX complaint, implements POSIX, and lots of other standards (listed in the standards(5) manual page). Even better, OpenSolaris will give people access to the implementations of those standards.
Both hardware _and_ software, there really is no lock-in, especially in the Microsoft sense.
You keep mulling about that using GPL code in a larger product means the entire product has to be released under the GPL,...
I think you are missing that _linking_ (using ld and ld.so in UNIX and Linux) is sufficient for something to be declared a larger work regarding the GPL. If I have a 50,000,000 line program that compiles into one binary, and I put 5 lines of GPL code into it (modified or not), I have to then put 50,000,005 LOC under the GPL in order to distribute it legally. The FSF FAQ is quite clear on this--no legal counsel needed.
They will tell you directly, that each and every GPL violation that is investigated, usually ends up in a compromise between the author of the GPL code, and the offending party...
The compromise is a mutual agreement. If one party persists on their claim (e.g., SCO), the case has to continue until the courts decide the result.
So please, get off the "viral" rant, its getting old, and its just propagating more misinformation that simply isn't true.
Sorry, but it seems you are muddling the issue. The GPL itself is quite clear on the matter.
It seems the whole point to your posts above is that the GPL isn't viral because you don't have to GPL the code--you just have to hire a lawyer and go to court or just opt out entirely. I guess that's a compromise.
You are also arguing based on a slightly different definition of 'viral'. In most discussions, it is used in the context of making larger works based on source code, not simply installing binaries that link against LGPL shared libraries.
...the code you incorporated into your proprietary product.
If that company ever distributed their proprietary code, it would have to be GPL.
(without looking at the GPL code as a baseline, because that becomes a copyright infringement, as it isn't a cleanroom implementation)
How is this possible after determining the GPL code isn't workable?
If you borrow a socket interface from a GPL product and put it into your application, does your entire application become a derivitive work of that socket interface? Of course not!
If you re-implemented the interface, no. If you borrowed the GPL code that implemented the interface, the resulting program has to be GPL.
Also, GPL != LGPL. GNU specifically uses the GPL on shared libraries where they want to enforce linked programs be GPL.
It grants rights, it does not take them away.
Pedant: It takes away rights that would otherwise belong to the public domain.
Your link to the GPL FAQ does not support your arguments:
There's more but seven is enough. In short, any time they hint at not requiring re-licensing under the GPL, there is a catch. Either the author has to dual-license (one license being the GPL) or amend the GPL (no guaranteed propogation). Always, the GPL is in the equation.
Look, I am not anti-GPL (it _is_ a useful license), but I do understand the rationale behind the per-file licenses like the MPL (and its derivatives). If I write code today based on an existing code base, I really can't predict my future needs whether to redistribute it or not. If I go the GPL route, that changes the game plan. With the CDDL, the game plan is more controllable, if I organize my files properly and make a modular program.
"Many licenses are compatible with this, and the "viral" clause does not stop you from making the code available under any other license you like as well."
Non-GPL code becomes GPL. It's like an irreversible chemical reaction, unless you happen to own the copyright to all the code to begin with.
That's probably because they can't. Sun started working on OpenSolaris five years ago, building upon the big UNIX license they got back in the early 90s.
Sure they ride the linux wave for corporate profit, but they put alot of cash into building that wave too.
Sun is doing the exact same thing with many (even GPL) projects. Sun pays people to work on OpenOffice.org. They do the same for GNOME. Sun even supports their middleware stack on Linux. Rumor has it they'll open source their middleware, too.
Their motives are different.
Red Hat and IBM are publically trade for-profit companies. Their motives are all the same. Why else would Red Hat go the RHEL/Fedora route? Also, Solaris has always been there, and Sun is only improving it (they certainly aren't taking anything away).
I don't see a double standard at all.
Really? You don't see all the different definitions of 'freedom' flying around? The slanted importance placed on 'GPL compatibility'? Those are some pretty impressive blinders.
What about OpenSolaris implies lock-in, and who modded this insightful? In fact, there was a mention at Blastwave that some people are interested in a PowerPC port of OpenSolaris. That's on top of all the x86 hardware it runs on, along with all the SPARC hardware.
The CDDL would be, because it is a file-based license, but the GPL does not allow it.
The only thing you lose is the patent litigation protection.
You can always purchase an expensive insurance policy from OSRM...or get it for free in OpenSolaris.
That's not really big on freedom.
OSI fully recognizes that Sun is acting quite freely. Your sentiment is extremely GPL-centered.
GPL-compatible
You've used forms of this word at least five times in your post. You do understand that GPL-compatibility is a one-way street and that combining code with GPL code requires that the larger work be released under the GPL, right? This carries implications that some people are uncomfortable with, and these people are accomodated under the CDDL.
One worry for MS though, if this catches on, might the ease of administration, standardised licensing etc, start to hurt full Windows sales ?
What about the permanent full-time employment enjoyed by many thousands of Windows admins, whose job it is to troubleshoot every little problem on all the Windows desktops?
Nor do past successes guarantee future successes. For example, just how much longer can Microsoft continue charging money for their software? Eventually people will look at the Microsoft tax versus IBM+Linux or Sun+Linux+OpenSolaris and do a double-take at Microsoft's price tag. It's inevitable. It doesn't hurt that Linux and OpenSolaris not only are cheaper but are years ahead of Windows NT/2000/XP technology-wise, too.
Microsoft's been doing this for twenty years. As the "oops, sorry" goes, please read the end user license agreement (EULA) that comes with any Microsoft product.
Also, no company can reliably distribute patches daily (the QA effort is non-trivial) and no customer wants to patch daily (that's just a waste of time).
This is why no one worth anything uses beta software in production. They want the majority of issues resolved first!
This name is long dead. People using it just look silly.
Also, the cost of a super-computer would accomodate a new Opteron server plus attached storage for NFS. Solaris 10's NFS implementation is the best in the industry, so why not use it? Various places, I've seen that Solaris 10 on a good 2-way server can easily saturate a gigabit network (meaning we need even faster networking to accomodate it).
A deterrant. Retailers don't care if they alarms are reliable; they use them to deter many of the casual wannabe shoplifters out there. And apparently they work, because the big retailers with analysts out the wazoo use them.
SGI could build a 50,000 node single-image system with 15,000TB of RAM, but it doesn't matter if no one buys it. Sun has already shipped tens of thousands of Opteron servers just in the last six months.
So, you're saying Windows XP doesn't append the disk storage to it's virtual memory space? That's a fairly inefficient way to use storage, if true. Solaris (Linux?) appends swap space to the physical RAM, so I can literally have a program use nearly RAM+swap amount of space without a hitch. Solaris'/tmp directory is also mounted as a RAM disk into virtual memory, which makes for an awesome place to put intermediate files from scripts. Solaris also uses all unused physical RAM as a disk cache.
I know this sounds like a solaris-solaris-solaris fanboy post...but it just seems that Windows XP is still seven years behind UNIX in kernel-land.
If you bothered to read my post, you would see that I was differentiating between the language (C#) and the platform (.Net).
For all practical purposes, they are one and the same. It's just like trying to get a lot of milage out of Lisp without POSIX or GUI layers. It just doesn't happen often. Microsoft understands this 100%.
If C# standardisation is purely a marketing ploy, how come so many 'big-guns' of the Open Source movement have adopted it and have been involved in porting to unix/linux?
If by 'big guns' you mean Mono, that's a pretty small faction of FOSS. Sure, there's some seriously influential people in there, but the project remains controversial in the FOSS world at large. Anyone who implicitly trusts Microsoft these days is a huge risk taker.
Anyone can produce a C# compiler and call it C#. That cannot be said of Java.
What does the 'j' in GCJ stand for? Sun just doesn't want people touting the Java Platform brand without conformance testing. The name of the language is not an issue.
Having a "standard" platform is what has driven the computer business for the last 15 years and has made the low cost PC possible.
No, IBM clones made the low-cost PC possible. They just threw Microsoft on there, because not many options existed and UNIX was still expensive. Microsoft rode the wave, finding ways to lock-in their customers, and, now, there is a genuine backlash against Microsoft.
The developing generation of GNOME/KDE-based desktops are going to drive the sub-$200 PC. Not Microsoft, not by a long shot.
Slashdot Mirror
Watch out, though, the license agreement is a vial you have to drink!
I just solved the problem: PIPIP (picture in picture in picture) with three keyboards, mice, and game controllers. All on the family 25" TV.
I propose: XBox360 and XBox 360 MEE (More Expensive Edition).
Given that media PCs have been poor sellers for a decade, now, Microsoft may be trying to _force_ media PC market penetration via the XBox. They would have to do this to compete against PVRs, for example.
So...what would the kids say if daddy is recording American Idol when they want to play? Or what about when mommy wants to get some crystal off of ebay?
Putting licensing peeing contests aside, Solaris is UNIX complaint, implements POSIX, and lots of other standards (listed in the standards(5) manual page). Even better, OpenSolaris will give people access to the implementations of those standards.
Both hardware _and_ software, there really is no lock-in, especially in the Microsoft sense.
You keep mulling about that using GPL code in a larger product means the entire product has to be released under the GPL, ...
I think you are missing that _linking_ (using ld and ld.so in UNIX and Linux) is sufficient for something to be declared a larger work regarding the GPL. If I have a 50,000,000 line program that compiles into one binary, and I put 5 lines of GPL code into it (modified or not), I have to then put 50,000,005 LOC under the GPL in order to distribute it legally. The FSF FAQ is quite clear on this--no legal counsel needed.
They will tell you directly, that each and every GPL violation that is investigated, usually ends up in a compromise between the author of the GPL code, and the offending party...
The compromise is a mutual agreement. If one party persists on their claim (e.g., SCO), the case has to continue until the courts decide the result.
So please, get off the "viral" rant, its getting old, and its just propagating more misinformation that simply isn't true.
Sorry, but it seems you are muddling the issue. The GPL itself is quite clear on the matter.
It seems the whole point to your posts above is that the GPL isn't viral because you don't have to GPL the code--you just have to hire a lawyer and go to court or just opt out entirely. I guess that's a compromise.
...the code you incorporated into your proprietary product.
You are also arguing based on a slightly different definition of 'viral'. In most discussions, it is used in the context of making larger works based on source code, not simply installing binaries that link against LGPL shared libraries.
If that company ever distributed their proprietary code, it would have to be GPL.
(without looking at the GPL code as a baseline, because that becomes a copyright infringement, as it isn't a cleanroom implementation)
How is this possible after determining the GPL code isn't workable?
If you borrow a socket interface from a GPL product and put it into your application, does your entire application become a derivitive work of that socket interface? Of course not!
If you re-implemented the interface, no. If you borrowed the GPL code that implemented the interface, the resulting program has to be GPL.
Also, GPL != LGPL. GNU specifically uses the GPL on shared libraries where they want to enforce linked programs be GPL.
It grants rights, it does not take them away.
Pedant: It takes away rights that would otherwise belong to the public domain.
Your link to the GPL FAQ does not support your arguments:
1
2
3
4
5
6
7
There's more but seven is enough. In short, any time they hint at not requiring re-licensing under the GPL, there is a catch. Either the author has to dual-license (one license being the GPL) or amend the GPL (no guaranteed propogation). Always, the GPL is in the equation.
Look, I am not anti-GPL (it _is_ a useful license), but I do understand the rationale behind the per-file licenses like the MPL (and its derivatives). If I write code today based on an existing code base, I really can't predict my future needs whether to redistribute it or not. If I go the GPL route, that changes the game plan. With the CDDL, the game plan is more controllable, if I organize my files properly and make a modular program.
"Many licenses are compatible with this, and the "viral" clause does not stop you from making the code available under any other license you like as well."
Non-GPL code becomes GPL. It's like an irreversible chemical reaction, unless you happen to own the copyright to all the code to begin with.
IBM isn't open sourcing AIX, or trying to.
That's probably because they can't. Sun started working on OpenSolaris five years ago, building upon the big UNIX license they got back in the early 90s.
Sure they ride the linux wave for corporate profit, but they put alot of cash into building that wave too.
Sun is doing the exact same thing with many (even GPL) projects. Sun pays people to work on OpenOffice.org. They do the same for GNOME. Sun even supports their middleware stack on Linux. Rumor has it they'll open source their middleware, too.
Their motives are different.
Red Hat and IBM are publically trade for-profit companies. Their motives are all the same. Why else would Red Hat go the RHEL/Fedora route? Also, Solaris has always been there, and Sun is only improving it (they certainly aren't taking anything away).
I don't see a double standard at all.
Really? You don't see all the different definitions of 'freedom' flying around? The slanted importance placed on 'GPL compatibility'? Those are some pretty impressive blinders.
What about OpenSolaris implies lock-in, and who modded this insightful? In fact, there was a mention at Blastwave that some people are interested in a PowerPC port of OpenSolaris. That's on top of all the x86 hardware it runs on, along with all the SPARC hardware.
It's not GPL-compatible.
The CDDL would be, because it is a file-based license, but the GPL does not allow it.
The only thing you lose is the patent litigation protection.
You can always purchase an expensive insurance policy from OSRM...or get it for free in OpenSolaris.
That's not really big on freedom.
OSI fully recognizes that Sun is acting quite freely. Your sentiment is extremely GPL-centered.
GPL-compatible
You've used forms of this word at least five times in your post. You do understand that GPL-compatibility is a one-way street and that combining code with GPL code requires that the larger work be released under the GPL, right?
This carries implications that some people are uncomfortable with, and these people are accomodated under the CDDL.
One worry for MS though, if this catches on, might the ease of administration, standardised licensing etc, start to hurt full Windows sales ?
What about the permanent full-time employment enjoyed by many thousands of Windows admins, whose job it is to troubleshoot every little problem on all the Windows desktops?
Past failures don't guarantee future failure.
Nor do past successes guarantee future successes. For example, just how much longer can Microsoft continue charging money for their software? Eventually people will look at the Microsoft tax versus IBM+Linux or Sun+Linux+OpenSolaris and do a double-take at Microsoft's price tag. It's inevitable. It doesn't hurt that Linux and OpenSolaris not only are cheaper but are years ahead of Windows NT/2000/XP technology-wise, too.
Didn't Google start all this?
Microsoft's been doing this for twenty years. As the "oops, sorry" goes, please read the end user license agreement (EULA) that comes with any Microsoft product.
Also, no company can reliably distribute patches daily (the QA effort is non-trivial) and no customer wants to patch daily (that's just a waste of time).
This is why no one worth anything uses beta software in production. They want the majority of issues resolved first!
Spoke a bit too soon. It looks like they're already pushing 10Gbit for regular traffic and getting >200MB/sec on NFS.
Slowaris
This name is long dead. People using it just look silly.
Also, the cost of a super-computer would accomodate a new Opteron server plus attached storage for NFS. Solaris 10's NFS implementation is the best in the industry, so why not use it? Various places, I've seen that Solaris 10 on a good 2-way server can easily saturate a gigabit network (meaning we need even faster networking to accomodate it).
A deterrant. Retailers don't care if they alarms are reliable; they use them to deter many of the casual wannabe shoplifters out there. And apparently they work, because the big retailers with analysts out the wazoo use them.
However, linux will still operate the NFS server because Apple's implementation of NFS server only allows fifty clients.
This is interesting, because Solaris 10 would be the best choice for an NFS server (especially for a super-computer, where these things matter).
Apparently, Sun's next-gen Galaxy servers were designed in-house. It'll be interesting to see how they spec out whenever they start shipping.
SGI could build a 50,000 node single-image system with 15,000TB of RAM, but it doesn't matter if no one buys it. Sun has already shipped tens of thousands of Opteron servers just in the last six months.
Windows XP doesn't swap, it uses a page file...
/tmp directory is also mounted as a RAM disk into virtual memory, which makes for an awesome place to put intermediate files from scripts. Solaris also uses all unused physical RAM as a disk cache.
So, you're saying Windows XP doesn't append the disk storage to it's virtual memory space? That's a fairly inefficient way to use storage, if true. Solaris (Linux?) appends swap space to the physical RAM, so I can literally have a program use nearly RAM+swap amount of space without a hitch. Solaris'
I know this sounds like a solaris-solaris-solaris fanboy post...but it just seems that Windows XP is still seven years behind UNIX in kernel-land.
No, the GP poster is now correct--the moderators have voted!
If you bothered to read my post, you would see that I was differentiating between the language (C#) and the platform (.Net).
For all practical purposes, they are one and the same. It's just like trying to get a lot of milage out of Lisp without POSIX or GUI layers. It just doesn't happen often. Microsoft understands this 100%.
If C# standardisation is purely a marketing ploy, how come so many 'big-guns' of the Open Source movement have adopted it and have been involved in porting to unix/linux?
If by 'big guns' you mean Mono, that's a pretty small faction of FOSS. Sure, there's some seriously influential people in there, but the project remains controversial in the FOSS world at large. Anyone who implicitly trusts Microsoft these days is a huge risk taker.
Anyone can produce a C# compiler and call it C#. That cannot be said of Java.
What does the 'j' in GCJ stand for? Sun just doesn't want people touting the Java Platform brand without conformance testing. The name of the language is not an issue.
Having a "standard" platform is what has driven the computer business for the last 15 years and has made the low cost PC possible.
No, IBM clones made the low-cost PC possible. They just threw Microsoft on there, because not many options existed and UNIX was still expensive. Microsoft rode the wave, finding ways to lock-in their customers, and, now, there is a genuine backlash against Microsoft.
The developing generation of GNOME/KDE-based desktops are going to drive the sub-$200 PC. Not Microsoft, not by a long shot.
In the end, some school official found out what was happening and put a stop to it.
Welcome to the wide wonderful world of NCLB, where individualism and excellence are illegal.