This contraption doesn't look terribly useful to me, because it doesn't take into account all of the other paraphenalia you need when you work. You're going to have a telephone, probably with a headset attachment, a pile of books, papers, meeting notes, etc., a file drawer, and so on. I find that most people working in IT, at least, have a second computer as well, and perhaps even two monitors on their primary computer.
create a version of MS-Office that will only run when a specific library is installed, and make that library available only with MS-Linux. Since it is not part of the kernel, they wouldn't have to release it as free software.
Well, that wouldn't make you use MS-Linux. You could just drop the library into/usr/lib in your own distribution and run it then. You'd have to buy MS-Linux, but not run it.
I do this sort of thing, for example, when I run Solaris programs under NetBSD. I just grabbed the libs and stuff I needed from the free Solaris 7 CD. (Of course, if I were using any of this for commerical purposes, I'd have to go out and buy a Solaris license.)
cjs
Re:SGI could use dual licensing, still get fixes
on
SGI open-sourcing XFS
·
· Score: 1
You'll never beat Microsoft if you write code for them (yes, Microsoft networking has tons of Berkeley code in there, you can tell from the bug-compatibility).
I don't understand this. If Microsoft hadn't been able to get a good IP stack so cheaply, they might have written a very bad one, or not bothered at all. Now where would SAMBA be if Windows didn't have good TCP/IP support? Or do you not consider SAMBA to be excellent competition?
The way you beat someone is to to make sure you've got a playing field on which you can beat them: you convince them to use open protocols. And the best way to do that is to give them the code for those protocols.
Code under the BSD licence has done far more than GPL'd code has (or ever will) to make sure that everyone uses the open protocols that are so absolutely necessary for free software to have a chance.
What you're talking about is a _log structured_ file system.
Right. A few more notes on this: a log-structured filesystem is a great thing if you do a lot of small writes, especially on a lot of different files. (Such as, say, on a news server.) You end up doing far fewer seeks than you would on a more traditional filesystem. On the other hand, you want a lot of buffer cache if you doing a fair number of reads.
NetBSD 1.4 has a log-structured filesystem called LFS, though it may still have a bug here or there. And it really wants a cleaner that also consolidates (defragments) files as it cleans.
But I highly doubt they'll go with anything BSD-like. That would be like telling Sun 'here's our XFS file system, please adopt and expand it for your own proprietary use'.
Would you mind explaining to me why Sun would put development effort into a very stripped-down XFS rather than further increasing their investment in UFS, which is currently far more capable than what SGI is releasing? (They already have large file and filesystem support, journalling, ACLs, and so on.)
What is it with Linux people that they think everybody is going to go around making massive proprietary enhancements to publically available code? To do that is very expensive; there's certainly no point if you already have better technology of your own, and even if you don't, it's often considerably cheaper to give your changes back to the maintainers than to keep them private.
What happens if your raid card goes? Use software raid to mirror accross two controllers and you've eliminated a single point of failure.
Buying a controller card with RAID built in is not typically how you'd set up a high-end server.
Instead, you buy a box from someone like Baydel or Sun Storage Solutions. This box has a bunch of disk drives in it and a couple of controller boards that have SCSI controllers to talk to the disks, RAID hardware, and a SCSI controller to talk to the host. This disk box deals with all of the RAID stuff (striping, parity, rebuilding failed disks on hot spares, etc.). The host sees the entire array as one or more disks.
For the host, you just buy a couple of regular SCSI controllers and plug each one of these into one of the controllers on the disk array box. Then you just have to set it up to fail over to the second controller if the first one fails in some way (be it the host controller, the cable, or the controller in the disk array box).
It seems to me that as long as I am not making money by distributing the code I should be able to do this?
It's got nothing to do with whether you're making money from the code or not. The restriction the GPL places on you is that if you give someone--anyone--a binary of a program, you must give him the source as well. That's all.
If Linux had been under a BSD style license, MS would have an easy and clear way of destroying it as a competitive threat. They would create a prorietary fork, flood the market with it, and we would all live in an MS world for the indefinate future.
I find it rather ironic that Linux folks regularly scream about FUD from MS and others, and then go and spread it themselves whenever they see the letters BSD.
The free versions of BSD haven't suffered at all because of the proprietary, commercial version out there. In fact, there's been a moderate amount of benefit from it in terms of broadening the user base and code and knowledge sharing. Nor has XFree86 (which is a notable part of Linux!) been destroyed by MS or anyone else, despite not being under the GPL.
If Linux weren't under the GPL, and a proprietary version were released by MS, would you use it? Would anyone you know use it? It's only if you would that Linux could be killed. Free software projects can't be killed from the outside, no matter what license they're on; as long as there are coders who are willing to work on the free versions, they will survive. But a loss of interest from those coders will always kill a project, even if it's under the GPL.
Now are you Linux folks finally going to stop spreading FUD about your fellow free software projects?
And don't give me that, "you can add it to xfree86" crap:) PC's running Linux or *BSD* aren't the only machines on the planet running X.
No, they're not: you're quite right about that. There would also be sparc workstations running BSD, Alphas running BSD, Sun 3s, Amigas, Ataris, and various other m68k boxes running BSD, Pmaxen and other MIPS boxes running BSD, ARM32 machines running BSD....and all running XFree86.
NetBSD runs a ton of platfroms under XFree86.
cjs
Re:Why not contact the Author?!
on
BSD vs GPL
·
· Score: 1
If the person refuses, then I suspect that either you approached the person or was a bully to him/her in another lifetime.=)
Ha ha! Not quite. NetBSD has a completely re-written version of the gnu readline library because Richard Stallman won't license it under the LGPL, just the GPL. He explicitly does this to try and force people to bring their entire program under the GPL if they want to use readline. (And this has worked; at least one program has converted from a non-GPL licence with BSD-like freedoms to the GPL just to be able to use readline.)
It has, look at BSDi and all the chunks of BSD code tucked away in all sorts of operating systems like the various win flavours.
And you are aware that this is a key thing responsible for the growth and popularity of Linux, right?
If you want widely to promulgate a particular piece of software, the GPL is a reasonable means of doing that. On the other hand, if you're trying to promulgate an open standard, the BSD licence is a much better strategy. When a company is making a network device, they need a network stack. The most expensive option is usually writing it yourself. A less expensive option is to go to a company that already has one (say, Microsoft or Novell) and license that. However, if there is an even less expensive option yet, that being taking a free implementation of TCP/IP and putting it in your proprietary product, it makes sense to go for that instead.
It's these commercial users `stealing' (in fact, taking what's freely given) the BSD networking code that widely deployed TCP/IP, and thus created the conditions necessary for a non-proprietary Internet. The Internet has (IMHO) done more to increase free software development than anything else out there. (I personally can't imagine how more than a hundred geographically separated developers could work on a system like NetBSD without the easy CVS access we all have via the Internet.)
cjs
Re:For another comparison...
on
BSD vs GPL
·
· Score: 1
Well, that was an interesting article, but its conclusion that the Gs will win makes a few assumptions that don't appear to be valid.
Ts never contribute back to the Os. In fact, they often do contribute back.
The primary source of contributions is through taking them from other programs. This doesn't appear to be the case, perhaps because it's not always easy to take good features out of one program and add them to another, due to design changes and the like.
My experience is that ext2 is way faster then solaris' ufs. Especially in high-performance queuing situations.
I would be very surprised the Linux FS was not faster than FFS or UFS--the Linux FS writes metadata asynchronously! This gives you a speed boost, though you stand a much greater chance of uncorrectable filesystem corruption if you have a crash. Try mounting the Solaris or FreeBSD or whatever filesystem async next time you do the comparison, and you should see similar speeds.
In my testing with bonnie on NetBSD (where there are no metadata writes), I find I get filesystem throughput very, very close (a few percent slower) than what I get doing raw I/O to the disk.
No, you don't likely have to statically link the open source libraries. Under NetBSD and FreeBSD, for example, you just download and pkg_add the linux compatability package, which installs all the Linux stuff you need in/emul/linux, and you're set.
Well, someone appears to be pissed off that his favourite Linux distro didn't get the offer from Usenix!
NetBSD will be pressing its own CD-ROMs, actually, though it will likely wait for the 1.4.1 release, since it will be out in the fairly near future (a couple of months) and it seems a bit of waste to press a huge pile of 1.4 disks only to have them go obsolete a month later. And I doubt that Walnut Creek is lacking in the resources to press FreeBSD CDs for the new release, either.
cjs
Re:I want to try BSD .. but which one?
on
*BSD News
·
· Score: 2
OpenBSD - The FreeBSD ports with NetBSD's selection of different hardware. Has auited the code for security. FreeBSD - Optimized for x86 opcodes. Easy to use ports/packages.
Actually, the NetBSD ports tree is right up there now, too. In the eighteen months since it was created it's come along fantastically.
Note that OpenBSD doesn't have as broad multi-platform support as NetBSD any more; OpenBSD started with the NetBSD source code a couple of years ago, but they've not imported most of the major NetBSD changes in the last year and a half (and there have been some big ones in the kernel, such as a complete new VM system).
Another problem that OpenBSD has had in the past (and may still have) is getting code compiled regularly on all their platforms; sometimes they appear to go a couple of months without a compile of some ports. So even if you're not a programmer, you can be a big help to them just by doing regular builds of OpenBSD-current on non-i386 hardware. (This help would also be welcome in the NetBSD camp, of course, though having more developers they aren't as needy in this regard.)
It should work under NetBSD/i386 as well as it works under Linux, FreeBSD, BSDI, SCO or Solaris/x86. And under NetBSD/sparc as well as it works under SunOS or Solaris. I'm not entirely convinced it `works' on any of these platforms, though.:-)
I do run Solaris and SunOS versions of Netscape Navigator quite successfully on my NetBSD/sparc system, and tons of Linux applications (including Communicator and RealAudio player) under NetBSD/i386.
It seems to me, in fact, that one of the best contributions Linux may be making is to bring forth a broadly available ABI--essentially an informal standard ABI for the Unix community. The lack of this has always been a complaint in the past.
understand that one of the problems with the BSD license is that a company such as AOL can take the existing code base, modify it to their needs, and copyright the resulting fork.
No, that's not an accurate statement; you are misunderstanding the issue.
In terms of copyright, AOL owns the copyright on any changes it makes, but not on the original code. This is true under both the BSD licence and the GPL. In both cases they are also free not to distribute their changes, if they want to keep them for in-house use.
The difference between the Berkeley licence and the GPL is that under the former they may distribute binary versions of their software without distributing the source, under the latter they must distribute the source as well.
In the end, this isn't as big a deal as one would think. There are very strong financial incentives to contributing changes back to the open source maintainers because it's actually quite expensive to maintain your own separate tree, especially in terms of debugging. Take a look at how far behind NetBSD OpenBSD now is, despite a couple of dozen developers working on it--they simply don't have the manpower to keep merging in the NetBSD changes, especially the larger ones.
Regardless, a project cannot be `taken over' by someone who's keeping his changes private; the original source code doesn't disappear. A project can only be `taken over' if people stop working on the free version. And the mere presence or absence of someone else using that code base for their ends is not going to do that.
I know this is an emotional issue for some people; that's why it's important to keep in mind that, while you may have problems with other people making money from you code, many BSD contributors don't have a problem with this. I'm perfectly happy to see someone take free code I've developed and make money from it; I put it out there for others to use, after all. That, to my mind, is the whole point of code being free: people can use it without the sort of restrictions that the GPL puts on you.
Actually that would be using the same principal of the PZM (Presure Zone Mic).
No, it wouldn't. A PZM doesn't rely on the resonance or movement of the surface it's on; in fact, it's best if it doesn't resonate at all. PZM microphones just get better base response on large surfaces because the surface will intercept the longer waves (rather than them going right around) and thus create a pressure differential in the gap between the microphone and the surface.
At least, this is as best I recall. I've not been an audio engineer for almost a decade now, and I can feel my knowledge deteriorating....
Equally, one might ask why HTTP was created at all... Since FTP does nearly everything that HTTP does.
There are a few things it doesn't do; there's no specification within FTP for the header information transferred with HTTP requests, for example. You could get this by layering HTTP on top of FTP, but then you've got the same problem you've always had with FTP: a lot of complexity that you don't need for this particular application.
I also find it odd that people find HTTP less reliable than FTP for downloading large files. It really shouldn't make any difference. (Of course, if you've got an unreliable connection, FTP is better assuming both the server and client support resumption of interrupted transfers.)
The main point the editorial was making is that there's a fairly large market segment that Linux doesn't (and can't, at the moment) address: on-line transaction processing where a consumer is waiting for that transaction to complete.
The post office application isn't like that. If the system shuts down for fifteen minutes, you won't notice they difference; they just fix it and start again. On the other hand, if your bank machine shuts down for a half hour when you're trying to get money out for a cab to the aeroport to catch a flight that leaves in forty minutes, you're going to be pretty darn upset (and rightfully so). People don't generally even use Unix in these sorts of applications.
Well, this seems a reasonable enough editorial to me. They're right: Linux isn't ready to do the sort of things that Tandem Himalaya, IBM Sysplex and DEC VMS cluster systems can do. I don't think that this is a big deal; nobody wants to pay for a Sysplex system to be a departmental file server, either. Each type of system has its role, and I don't see Linux as going any further than Sun currently does with their systems.
But too, while the hype is annoying, I don't see it doing any real long term damage to Linux. Linux is just the IT industry buzzword du jour, as `client server' and `data warehousing' once were. Eventually people will figure out that it's not the silver bullet that's going to cause a miracle in the industry, and things will settle down. Linux will have reasonable success and lots of people will be happily using it.
The only reason for moving it in the kernel is performance. But that is an indication to me that the system APIs simply aren't complete or efficient enough to allow implementation of something like NFS.
I disagree. While a better API (such as one that avoids kernel/userland memory copies) can help, part of this is not an API problem at all: on every processor out there (at least every one I've ever seen) the switch between kernel and user mode is very expensive. There is no fix for this, except not to switch.
Sure, there would be screams of agony if the GPL were shot down. But how many of you folks who work on GPL'd software would stop contributing to the free software world? Raise your hands. I thought so.
As someone else here pointed out, the BSD projects are getting along just fine without the GPL, leaving companies free to `steal' code (in fact, we *BSD developers give it to them freely). And we even get changes back; Apple basically dedicated several weeks of developer time to bringing userland changes that Apple had made back into the NetBSD source tree.
I think people here really underestimate the amount of work required to privately maintain your own changes to public source. Most of the time it's just not worth it, and it's much cheaper to contribute your changes back. If you don't contribute them back, you lose in two ways:
You don't get the rest of the community debugging and fixing your code; you have to do it all yourself
You have to dedicate a lot of time and effort to merging the community's changes into your source tree, and testing the changes within your product.
The latter point is really important. For an example of what happens when you try this sort of split, have a look at NetBSD and OpenBSD. When OpenBSD split from NetBSD a couple of years ago, they took a complete copy of the code base as it stood. For a while they attempted to pull in all NetBSD changes, as well as making their own changes, but eventually it overwhelmed them (even with a couple of dozen developers working on this!) and they gave up. Now NetBSD has a new machine-independent device driver and DMA structure, a new VM system, and a lot of other changes, and OpenBSD has very little of that. They've fallen far behind and haven't much hope of ever catching up. The same would have happened to a commercial user without spending a large amount of money--and avoiding spending a large amount of money was probably the point of using free source in the first place.
In Canada the pistol may not have been illegal in and of itself, but carrying it around in public (concealed or not) certainly is! Unlike the US, nobody seems to protest this law much.
Having now spent some time in the US, (I grew up and spent most of my adult life in Canada), the difference in gun laws seems now to me more a symptom than a cause. There appears to be a substantial minority of Americans who have an attitude about weapons that's quite rare in Canada: it's almost like they worship the guns and the right to own and do whatever they want with them, without having to commit to some responsibility for their use. They loudly and vocally protest the placing on guns the sort of `harsh' restrictions placed on things like automobiles. I really don't know why this is.
This sort of thing seems to be almost a general trend in the US: people yell a lot about freedom, but are willing to give up massive amounts of it in most areas. The US is far closer to a national ID system than Canada or England; the number of people who want (and get!) your SSN absolutely apalls me; there are all sorts of people and institutions requiring an SSN down here under circumstances where it would be illegal even to ask for it in Canada. And I don't see many signs of protest of it.
Anyway, a sensible gun control law to my mind might be to allow weapon ownership, concealment, and the like, but to force registeration of all weapons in the same way we force registeration of all cars and drivers. Then put in place some reasonable penalties for the owner if the weapon is ever used in committing a crime (regardless of whether the owner had anything to do with it or not). Add smaller penalties for selling a weapon to someone who then uses it to commit a crime. That should make gun owners and sellers take a bit of responsibility for the care and movement of their weapons, providing society with better protection against misuse.
Slashdot Mirror
This contraption doesn't look terribly useful to me, because it doesn't take into account all of the other paraphenalia you need when you work. You're going to have a telephone, probably with a headset attachment, a pile of books, papers, meeting notes, etc., a file drawer, and so on. I find that most people working in IT, at least, have a second computer as well, and perhaps even two monitors on their primary computer.
cjs
I do this sort of thing, for example, when I run Solaris programs under NetBSD. I just grabbed the libs and stuff I needed from the free Solaris 7 CD. (Of course, if I were using any of this for commerical purposes, I'd have to go out and buy a Solaris license.)
cjs
The way you beat someone is to to make sure you've got a playing field on which you can beat them: you convince them to use open protocols. And the best way to do that is to give them the code for those protocols.
Code under the BSD licence has done far more than GPL'd code has (or ever will) to make sure that everyone uses the open protocols that are so absolutely necessary for free software to have a chance.
cjs
NetBSD 1.4 has a log-structured filesystem called LFS, though it may still have a bug here or there. And it really wants a cleaner that also consolidates (defragments) files as it cleans.
cjs
What is it with Linux people that they think everybody is going to go around making massive proprietary enhancements to publically available code? To do that is very expensive; there's certainly no point if you already have better technology of your own, and even if you don't, it's often considerably cheaper to give your changes back to the maintainers than to keep them private.
cjs
Instead, you buy a box from someone like Baydel or Sun Storage Solutions. This box has a bunch of disk drives in it and a couple of controller boards that have SCSI controllers to talk to the disks, RAID hardware, and a SCSI controller to talk to the host. This disk box deals with all of the RAID stuff (striping, parity, rebuilding failed disks on hot spares, etc.). The host sees the entire array as one or more disks.
For the host, you just buy a couple of regular SCSI controllers and plug each one of these into one of the controllers on the disk array box. Then you just have to set it up to fail over to the second controller if the first one fails in some way (be it the host controller, the cable, or the controller in the disk array box).
cjs
cjs
The free versions of BSD haven't suffered at all because of the proprietary, commercial version out there. In fact, there's been a moderate amount of benefit from it in terms of broadening the user base and code and knowledge sharing. Nor has XFree86 (which is a notable part of Linux!) been destroyed by MS or anyone else, despite not being under the GPL.
If Linux weren't under the GPL, and a proprietary version were released by MS, would you use it? Would anyone you know use it? It's only if you would that Linux could be killed. Free software projects can't be killed from the outside, no matter what license they're on; as long as there are coders who are willing to work on the free versions, they will survive. But a loss of interest from those coders will always kill a project, even if it's under the GPL.
Now are you Linux folks finally going to stop spreading FUD about your fellow free software projects?
cjs
NetBSD runs a ton of platfroms under XFree86.
cjs
cjs
If you want widely to promulgate a particular piece of software, the GPL is a reasonable means of doing that. On the other hand, if you're trying to promulgate an open standard, the BSD licence is a much better strategy. When a company is making a network device, they need a network stack. The most expensive option is usually writing it yourself. A less expensive option is to go to a company that already has one (say, Microsoft or Novell) and license that. However, if there is an even less expensive option yet, that being taking a free implementation of TCP/IP and putting it in your proprietary product, it makes sense to go for that instead.
It's these commercial users `stealing' (in fact, taking what's freely given) the BSD networking code that widely deployed TCP/IP, and thus created the conditions necessary for a non-proprietary Internet. The Internet has (IMHO) done more to increase free software development than anything else out there. (I personally can't imagine how more than a hundred geographically separated developers could work on a system like NetBSD without the easy CVS access we all have via the Internet.)
cjs
Well, that was an interesting article, but its conclusion that the Gs will win makes a few assumptions that don't appear to be valid.
cjs
In my testing with bonnie on NetBSD (where there are no metadata writes), I find I get filesystem throughput very, very close (a few percent slower) than what I get doing raw I/O to the disk.
cjs
No, you don't likely have to statically link the open source libraries. Under NetBSD and FreeBSD, for example, you just download and pkg_add the linux compatability package, which installs all the Linux stuff you need in /emul/linux, and you're set.
cjs
Well, someone appears to be pissed off that his favourite Linux distro didn't get the offer from Usenix!
NetBSD will be pressing its own CD-ROMs, actually, though it will likely wait for the 1.4.1 release, since it will be out in the fairly near future (a couple of months) and it seems a bit of waste to press a huge pile of 1.4 disks only to have them go obsolete a month later. And I doubt that Walnut Creek is lacking in the resources to press FreeBSD CDs for the new release, either.
cjs
Note that OpenBSD doesn't have as broad multi-platform support as NetBSD any more; OpenBSD started with the NetBSD source code a couple of years ago, but they've not imported most of the major NetBSD changes in the last year and a half (and there have been some big ones in the kernel, such as a complete new VM system).
Another problem that OpenBSD has had in the past (and may still have) is getting code compiled regularly on all their platforms; sometimes they appear to go a couple of months without a compile of some ports. So even if you're not a programmer, you can be a big help to them just by doing regular builds of OpenBSD-current on non-i386 hardware. (This help would also be welcome in the NetBSD camp, of course, though having more developers they aren't as needy in this regard.)
cjs
I do run Solaris and SunOS versions of Netscape Navigator quite successfully on my NetBSD/sparc system, and tons of Linux applications (including Communicator and RealAudio player) under NetBSD/i386.
It seems to me, in fact, that one of the best contributions Linux may be making is to bring forth a broadly available ABI--essentially an informal standard ABI for the Unix community. The lack of this has always been a complaint in the past.
cjs
In terms of copyright, AOL owns the copyright on any changes it makes, but not on the original code. This is true under both the BSD licence and the GPL. In both cases they are also free not to distribute their changes, if they want to keep them for in-house use.
The difference between the Berkeley licence and the GPL is that under the former they may distribute binary versions of their software without distributing the source, under the latter they must distribute the source as well.
In the end, this isn't as big a deal as one would think. There are very strong financial incentives to contributing changes back to the open source maintainers because it's actually quite expensive to maintain your own separate tree, especially in terms of debugging. Take a look at how far behind NetBSD OpenBSD now is, despite a couple of dozen developers working on it--they simply don't have the manpower to keep merging in the NetBSD changes, especially the larger ones.
Regardless, a project cannot be `taken over' by someone who's keeping his changes private; the original source code doesn't disappear. A project can only be `taken over' if people stop working on the free version. And the mere presence or absence of someone else using that code base for their ends is not going to do that.
I know this is an emotional issue for some people; that's why it's important to keep in mind that, while you may have problems with other people making money from you code, many BSD contributors don't have a problem with this. I'm perfectly happy to see someone take free code I've developed and make money from it; I put it out there for others to use, after all. That, to my mind, is the whole point of code being free: people can use it without the sort of restrictions that the GPL puts on you.
cjs
At least, this is as best I recall. I've not been an audio engineer for almost a decade now, and I can feel my knowledge deteriorating....
cjs
I also find it odd that people find HTTP less reliable than FTP for downloading large files. It really shouldn't make any difference. (Of course, if you've got an unreliable connection, FTP is better assuming both the server and client support resumption of interrupted transfers.)
cjs
But the post office isn't OLTP.
The main point the editorial was making is that there's a fairly large market segment that Linux doesn't (and can't, at the moment) address: on-line transaction processing where a consumer is waiting for that transaction to complete.
The post office application isn't like that. If the system shuts down for fifteen minutes, you won't notice they difference; they just fix it and start again. On the other hand, if your bank machine shuts down for a half hour when you're trying to get money out for a cab to the aeroport to catch a flight that leaves in forty minutes, you're going to be pretty darn upset (and rightfully so). People don't generally even use Unix in these sorts of applications.
cjs
Well, this seems a reasonable enough editorial to me. They're right: Linux isn't ready to do the sort of things that Tandem Himalaya, IBM Sysplex and DEC VMS cluster systems can do. I don't think that this is a big deal; nobody wants to pay for a Sysplex system to be a departmental file server, either. Each type of system has its role, and I don't see Linux as going any further than Sun currently does with their systems.
But too, while the hype is annoying, I don't see it doing any real long term damage to Linux. Linux is just the IT industry buzzword du jour, as `client server' and `data warehousing' once were. Eventually people will figure out that it's not the silver bullet that's going to cause a miracle in the industry, and things will settle down. Linux will have reasonable success and lots of people will be happily using it.
cjs
cjs
Sure, there would be screams of agony if the GPL were shot down. But how many of you folks who work on GPL'd software would stop contributing to the free software world? Raise your hands. I thought so.
As someone else here pointed out, the BSD projects are getting along just fine without the GPL, leaving companies free to `steal' code (in fact, we *BSD developers give it to them freely). And we even get changes back; Apple basically dedicated several weeks of developer time to bringing userland changes that Apple had made back into the NetBSD source tree.
I think people here really underestimate the amount of work required to privately maintain your own changes to public source. Most of the time it's just not worth it, and it's much cheaper to contribute your changes back. If you don't contribute them back, you lose in two ways:
- You don't get the rest of the community debugging and fixing your code; you have to do it all yourself
- You have to dedicate a lot of time and effort to merging the community's changes into your source tree, and testing the changes within your product.
The latter point is really important. For an example of what happens when you try this sort of split, have a look at NetBSD and OpenBSD. When OpenBSD split from NetBSD a couple of years ago, they took a complete copy of the code base as it stood. For a while they attempted to pull in all NetBSD changes, as well as making their own changes, but eventually it overwhelmed them (even with a couple of dozen developers working on this!) and they gave up. Now NetBSD has a new machine-independent device driver and DMA structure, a new VM system, and a lot of other changes, and OpenBSD has very little of that. They've fallen far behind and haven't much hope of ever catching up. The same would have happened to a commercial user without spending a large amount of money--and avoiding spending a large amount of money was probably the point of using free source in the first place.cjs
In Canada the pistol may not have been illegal in and of itself, but carrying it around in public (concealed or not) certainly is! Unlike the US, nobody seems to protest this law much.
Having now spent some time in the US, (I grew up and spent most of my adult life in Canada), the difference in gun laws seems now to me more a symptom than a cause. There appears to be a substantial minority of Americans who have an attitude about weapons that's quite rare in Canada: it's almost like they worship the guns and the right to own and do whatever they want with them, without having to commit to some responsibility for their use. They loudly and vocally protest the placing on guns the sort of `harsh' restrictions placed on things like automobiles. I really don't know why this is.
This sort of thing seems to be almost a general trend in the US: people yell a lot about freedom, but are willing to give up massive amounts of it in most areas. The US is far closer to a national ID system than Canada or England; the number of people who want (and get!) your SSN absolutely apalls me; there are all sorts of people and institutions requiring an SSN down here under circumstances where it would be illegal even to ask for it in Canada. And I don't see many signs of protest of it.
Anyway, a sensible gun control law to my mind might be to allow weapon ownership, concealment, and the like, but to force registeration of all weapons in the same way we force registeration of all cars and drivers. Then put in place some reasonable penalties for the owner if the weapon is ever used in committing a crime (regardless of whether the owner had anything to do with it or not). Add smaller penalties for selling a weapon to someone who then uses it to commit a crime. That should make gun owners and sellers take a bit of responsibility for the care and movement of their weapons, providing society with better protection against misuse.
cjs