You know, it's not just a LISP or Python operator... some of us use it to write in our languages. Tres frequemment, sometimes. (I'm not French, but, similarly to French, my native language uses the grave accent - just not as often.)
But more importantly, it is used in a PERL script to run another application, ie, `wget slashdot.org`; or whatever you want to run. How its used in lisp or python I don't know. Anyone?
Those of us that take pride in our total lack of programming skills find it very handy. It again confirms PERLs place as the "Swiss Army Chainsaw" of programming languages. I have several PERL scripts that use this throughout, in a very ugly but effective way.
Except where I live... Freaking bastards. there is *NO* guinness in Santa Maria, California, except one place-- Who sells so little of it, it tastes stale and awful...
I find it tastes that way everywhere. For a real beer, I prefer Red Oak which is brewed locally at their brewery/pub. Getting a fresh one isn't an issue. In kegs only, since it isn't pasturized. I would imagine most decent sized cities have their own breweries as well.
I said the concept of 'public domain' applies purely to copyright law, not patent law.
You are correct that the traditional PD applies to Copyright, and I was about to agree with you that it doesn't apply to patents (to be honest, I had not thought about the distiction until this conversation) but then I started Googling.
Now, usually when googling, you search and find the answer, but what I found was a muddy mess. It appears some people have applied for patents in the name of PD, but the links are broken. This seems to imply you can put a patent into PD, sort of.
While I agree technically with what you are saying, there has to be come mechanism to donate a patent to the Public Domain, albeit differently than Copyrighted works. If nothing else, donating it to the FSF. I had hoped that 15 minutes of Googling would provide a definitive answer, but no luck.
My guestimation is that we are both a little right and wrong on this one.
It's either a trick or the guy doesn't know anything about patents. Someone needs to email him and figure it out.
We have a few.
You can attempt to patent something that is in the Public Domain, and actually get the patent through. There are tons of examples of this on Slashdot. Thats the whole idea about "prior art". So obviously, there ARE patents out there with prior art. So yes, it can, and does, happen. When it does, it get very expensive, especially if it is something you are giving away and not making any profit on.
The defensive patent is less expensive than the potential legal wrangling that could easily cost up to $100,000. Ten times the amount of a patent.
IBM does this, but doesn't PD them. This is exaclty what they just countersued SCO for. Patent infringement. The only difference is if you PD it after you patent it, you guarantee that no one can ever make a claim on it, unless they demonstate prior art. The burden is on them.
This is a very common, common concept. They are usually just unenforced, rather than PDed, but the concept is very simple and cost effective.
And then when those ISP's are blacklisted, and no new ISP's who refuse to adopt the new protocol are allowed in, all is well.
You guys gotta get over the concept that using any service from another company on the internet is a RIGHT. AOL doesn't have to accept mail from anyone. Or allow sending to anyone. They may lose customers, but they are not obligated to do so. Nor is any other ISP. I use AOL as the largest example only.
You don't have a Right for AOL to accept mail from your server. My server refuses mail from hundreds of domains. So can their servers. They accept my mail because we both use the same protocol and standards of conduct, as to not get on anyone's blacklist. If you don't, you don't get access.
This is not a forced list, its a way for ISP's to stop spam with fraudulent addresses that point to them. You can still send all the mail you want, as long as you don't use aol.com as a return address (assuming you are not on aol's backbone). It is a voluntary system that is being developed by many different people in the communtity. Its not Microsoft forcing something on everyone. It is totally voluntary and developmental.
Put it in perspective. If it sucks, it won't get used very long by anyone.
you can have port 25 open, but only accept mail from certain ip ranges or from certain domains (forcing a dns lookup by sendmail). Linuxconf can even set this up. You can't firewall it off or YOU can't send mail.
Once you cut the amount of spam, by simply making it harder to spam, it becomes easier to filter out the open relays. The problem now is there is so much noise.
The bottom line is this. SMTP servers should ONLY send messages that originate from TRACEABLE sources (that can be held accountable). It makes NO difference whether the source of this accountability is a RESPONSIBLE (e.g. trusted or owned by the sysadmin of the SMTP) IP address, or a RESPONSIBLE username/password.
Yes. This system would help by adding another layer to demonstrate that they source is accountable.
Hate to see what Hotmails DNS will look like with a few million: 1.7.168.192.in-addr._smtp_client.hotmail.com. TXT "spf=allow" entries in it . ..
I have no idea on this, as I understand dns at a minor level, but wouldn't it be entirely possible to patch NAMED to send ranges and not just individual addys? IE: 24/1.7.168.192.in-addr._smtp_client.hotmail.com. TXT "spf=allow" -and patch sendmail (which will already have to be patched to use this system) to understand the network bit?
Again, I have no idea how trivial this would be, but it seems that it would not be the most difficult change. It just requires SENDMAIL to do a bit of math once it gets the answer.
That doesn't sound right. If "public domain" meant "no patent", then they'd save $10,000 by making it public domain without patenting it.
Its called a defensive patent. You can release a concept into the public domain, but if I patent it one month later, I can tie your ass up in court for years, get injunctions against anyone using it, SCO companies to get some money, and make general havoc. If I was a spammer, and you didn't patent it, I would just because I could do this.
Yes, in the end, I would lose the case after you prove prior art (years later), but I would make a fortune in the interum.
So ISPs will no longer have to deliver spam? Please. I doubt it will even make a dent in the number of unsolicited bulk messages.
What is your basis for this?
This is just ridiculous. When I send mail to abuse@ they have logs and know who to go after too. How would they suddenly be able to identify spam themselves if it were authenticated? They would have nothing more than they do now. Tracing an IP/time to a username is a very minimal amount of time.
Tracing often leads to open relays in China, which those servers would now be blocked with this list.
SMTP servers now only allow IPs from within their controlled subnet to send messages.
Sorry, not true. That is the problem. There are open relays everywhere, and SMTP with Auth allow you to send from anywhere. This is just an oversimplification.
This isn't a cure all, but if you can reduce 50% to 75% of the spam, it becomes much easier to track down the other 25% to 50%. Anything that increases accountability is a good thing.
Nothing will cure the spam problem by itself, but according to AOL, they get 1 billion emails a day, and over 2/3rds are spam. Just cutting it in half saves them handling 330 millions pieces of junk, which IS significant. Think about it: 2/3rds of their computer bank that is used for mail is to purely handle spam, and cutting it in half would reduce their mail handling needs in a way that saves them millions a year. Like I said, its not a cure, but its significant.
You raise a good point. I found my sdsl ip range was getting blocked because it was "possible" that spam could come from it. We had maybe 20 emails a day going through that particular network is all, so I got to go rent a rack instead, since I could not send mail. Moving it all over to a pair of T1s now, so it shouldn't be a problem. But many larger ISP's are now blocking all sdsl IP ranges. bummer.
Actually, you just have to setup your home computer as a mail relay, which is not that big of a deal.
As to setting up your own server, if you can setup the box on the network and get through the firewall, then your college doesn't have to authorize anything per se. You just need a rent DNS service or use your own, and have it point to the IP bank from the college as a legitimate IP address range for that domain, and use your new domain name in your mail. You can get DNS service pretty cheap. Its not that hard, just takes some creativity.
Actually, that brings something important to mind: Here in Australia a very large proportion of mail servers are Debian boxes. If that patent idea gets taken up, I can't see Debian including SPF; it'll be poison.
He refered to patenting it, and immediately releasing it as Public Domain. This means it can be used by anyone, in any license, even Microsoft. Actually, you NEED Microsoft to use it if you want it to work anyway. But there is already lots of PD in Linux, including Debian, so no worries.
From the original article: The rebate program by the Lexington, Ky.-based company offers an upfront discount to consumers who agree to return used cartridges only to Lexmark for refilling or recycling.
This doesn't change anyone's ability to refill their own cartridges, and ties the rebate to the promise to use only Lexmark authorized supplies. HP has sensors in its newer printers that can tell if you use their supplies or others, which affects your warranty, similar. This is not as big of a deal as it is played out to be here, since you can always choose to not buy Lexmark. They don't have an monopoly, after all.
This said, its still a crappy ruling, and it IS being appealed. Unfortunately, here in the States, most cases like this are finally settled on appeal, seldom at the first trial. Case in point is the recent attempt to put off the California recall. Our legal system may have plenty of flaws, but eventually it works (most of the time).
Errr, since the code was released under the GPL without SCO's permission, yes - either pony up or stop using XFS.
No, technically, if SGI released it and it was SCO's code, then its not GPL because they didn't have the legal authority (copyright) to release it under any license. It has to be "recalled" so to speak, meaning removed and SGI would be liable for any real damages.
Where's the guarantee that Linux is not tainted with some other obscure company's (SOOC for short) code? Where's the guarantee that programmers working on the kernel didn't use good old copy & paste in "tough" places? And most importantly, when Linux is mainstream enough, where's the guarantee that SOOC will not sue for frikkin billions for code theft / patent infingement / licensing agreement violations?
Read the GNU GPL itself. Their is no guarantee, although a company can legally offer a guarantee if they choose. The guarantee that SCO won't sue frikkin billions is the fact that you can't hold an end user liable, even though SCO talk like they are. This is demonstrated by the fact that they have not sent out invoices. They can hold RedHat responsible for the copy RedHat sold me, but I didn't not profit from the code financially (unless I resell it). Unless someone can point you to a legal precedent, don't believe it.
If you buy a Sony TV, and its found that Sony stole some resistors and used them to built your TV, they don't come in and make you pay for the resistors or take them out of your TV. They bust Sony's ass. SCO wants you to *THINK* they can sue you, but they can't. This is part of their financial strategy, but it has no legal precedence.
Actually, it was released, but if code inside cannot be GPL'd, then the release could not be GPL'd, therefore the non-GPL-able code has not been released under the GPL. Quite simple, really.
Congrats, you appear to be the only one that got it right. If it legally could not have been release under the GPL, then it is not, even if it was.
At that point it is discovered and disclosed that the code could not legally be GPL'ed, it is not GPL and never has been. Simple, yes. Wierdly recursive, but simple.
Part of what you would expect in pay would depend on the license they want. If they want you to dual license the software, (GPL and BSD for example) so they can avoid releasing changes, then I would expect more simply because you are offering them more value, with no requirement that they put back into the community.
I personally don't see a problem dual licensing, but you have to be very careful to not put stuff you added in THEIR licensed version, in your GPL version, or they could get pissed and take action. You would have to decide if its worth the potential risk.
If SCO goes under SysV doesn't just disappear, somebody will own it, and as long as that somebody isn't releasing it under an open source license the possibility remains for this same situation to come up yet again.
Actually, SCO couldn't release SysV into the Public Domain if they wanted to. SCO doesn't exactly own every right to SysV, they do own the right to license it. If you PD code, that means you can strip the copyright notices off, since PDing code means waiving the copyright. The problem is, they don't own the copyright to every piece, just the legal and exclusive right to license it. You would have to get the permission of the original copyright holders, who still legally own the original copyright, even if they licensed the right to use it to someone else.
Again, IANAL, but I have seen several references to this, and it kind of makes sense since they don't *really* own all the code, but rather just own the right to make money off of it. Some of the code is BSD, for example, and they can not PD code that is BSD without Berkeley's permission.
This means that it is likely that SysV will never be open sourced in any way. If my understanding is correct, it would be virtually impossible to do so. The best you could hope for is whoever BUYS the rights that SCO has now, will publicly declare that they will not enforce the copyrights or assign ownership to FSF, but this would not make it GPL or PD, just unenforced.
Still, I don't see any prohibition against including a few segments of PD code into a GPL'd package.
First, IANAL. Second, maybe I didn't make myself clear. It is absolutely legal and justifiable to use PD code in GPL programs. PD code is the ultimate expression of freedom, more free than GPL, since you don't have ANY restrictions. There is a significant amount of PD code in Linux. It is also perfectly legal to have PD code in SysV, or Windows, or in your Tivo without any disclosure.
But SCO has no cause of action that Linux has the same PD code, since PD by its nature HAS NO RESTRICTIONS. That was the big deal with the Greeked code SCO used as a demonstration recently, it was PD code. This is part of the reason SCO doesn't want to show code, because if you can show it was released under PD, then there is no actionable cause.
If you PD some of your code, you can NEVER enforce any copyright over it, since by the nature of PD, you waive your right to copyright. As a matter of fact, you can write code and release it under both the GPL and BSD license (for example) but you can't release it into the Public Domain and any other license, technically. This doesn't stop someone from using that PD code in a proprietary software package and copyrighting that package, but technically the PD portion is not copyrighted, but you just don't know it because you don't have the source. Weird, but that is the way I understand it.
But if they have already released it into the public domain...
Putting something in the Public Domain specifically means you are giving up any copyright to it. NO ONE owns the Copyright for Public Domain works, thus, anyone can use it for any reason whatsoever. Microsoft can use it, you can use it, Saddam Hussan can use, all legally, and no one has any authority to limit how you use it.
The GPL relies on Copyright, and is very unrelated to PD. If you never have read it, go to www.gnu.org and read the GPL. It is very much like any other software license in a legal sense, only the terms are different. Its an interesting read. GPL does NOT mean you can do what you want with it. If you change GPL software, sell or give away a modified binary but refuse to show me the source, you are infringing the copyright, for instance, and subject to getting your butt sued by the owner of the copyright.
Slashdot Mirror
You still haven't shown me anything that looks like a failure of the open source development model.
:)
Caldera Linux
You know, it's not just a LISP or Python operator... some of us use it to write in our languages. Tres frequemment, sometimes. (I'm not French, but, similarly to French, my native language uses the grave accent - just not as often.)
But more importantly, it is used in a PERL script to run another application, ie, `wget slashdot.org`; or whatever you want to run. How its used in lisp or python I don't know. Anyone?
Those of us that take pride in our total lack of programming skills find it very handy. It again confirms PERLs place as the "Swiss Army Chainsaw" of programming languages. I have several PERL scripts that use this throughout, in a very ugly but effective way.
Except where I live... Freaking bastards. there is *NO* guinness in Santa Maria, California, except one place-- Who sells so little of it, it tastes stale and awful...
I find it tastes that way everywhere. For a real beer, I prefer Red Oak which is brewed locally at their brewery/pub. Getting a fresh one isn't an issue. In kegs only, since it isn't pasturized. I would imagine most decent sized cities have their own breweries as well.
I said the concept of 'public domain' applies purely to copyright law, not patent law.
You are correct that the traditional PD applies to Copyright, and I was about to agree with you that it doesn't apply to patents (to be honest, I had not thought about the distiction until this conversation) but then I started Googling.
Now, usually when googling, you search and find the answer, but what I found was a muddy mess. It appears some people have applied for patents in the name of PD, but the links are broken. This seems to imply you can put a patent into PD, sort of.
While I agree technically with what you are saying, there has to be come mechanism to donate a patent to the Public Domain, albeit differently than Copyrighted works. If nothing else, donating it to the FSF. I had hoped that 15 minutes of Googling would provide a definitive answer, but no luck.
My guestimation is that we are both a little right and wrong on this one.
It's either a trick or the guy doesn't know anything about patents. Someone needs to email him and figure it out.
We have a few.
You can attempt to patent something that is in the Public Domain, and actually get the patent through. There are tons of examples of this on Slashdot. Thats the whole idea about "prior art". So obviously, there ARE patents out there with prior art. So yes, it can, and does, happen. When it does, it get very expensive, especially if it is something you are giving away and not making any profit on.
The defensive patent is less expensive than the potential legal wrangling that could easily cost up to $100,000. Ten times the amount of a patent.
IBM does this, but doesn't PD them. This is exaclty what they just countersued SCO for. Patent infringement. The only difference is if you PD it after you patent it, you guarantee that no one can ever make a claim on it, unless they demonstate prior art. The burden is on them.
This is a very common, common concept. They are usually just unenforced, rather than PDed, but the concept is very simple and cost effective.
And then when those ISP's are blacklisted, and no new ISP's who refuse to adopt the new protocol are allowed in, all is well.
You guys gotta get over the concept that using any service from another company on the internet is a RIGHT. AOL doesn't have to accept mail from anyone. Or allow sending to anyone. They may lose customers, but they are not obligated to do so. Nor is any other ISP. I use AOL as the largest example only.
You don't have a Right for AOL to accept mail from your server. My server refuses mail from hundreds of domains. So can their servers. They accept my mail because we both use the same protocol and standards of conduct, as to not get on anyone's blacklist. If you don't, you don't get access.
This is not a forced list, its a way for ISP's to stop spam with fraudulent addresses that point to them. You can still send all the mail you want, as long as you don't use aol.com as a return address (assuming you are not on aol's backbone). It is a voluntary system that is being developed by many different people in the communtity. Its not Microsoft forcing something on everyone. It is totally voluntary and developmental.
Put it in perspective. If it sucks, it won't get used very long by anyone.
you can have port 25 open, but only accept mail from certain ip ranges or from certain domains (forcing a dns lookup by sendmail). Linuxconf can even set this up. You can't firewall it off or YOU can't send mail.
Once you cut the amount of spam, by simply making it harder to spam, it becomes easier to filter out the open relays. The problem now is there is so much noise.
The bottom line is this. SMTP servers should ONLY send messages that originate from TRACEABLE sources (that can be held accountable). It makes NO difference whether the source of this accountability is a RESPONSIBLE (e.g. trusted or owned by the sysadmin of the SMTP) IP address, or a RESPONSIBLE username/password.
Yes. This system would help by adding another layer to demonstrate that they source is accountable.
Hate to see what Hotmails DNS will look like with a few million: 1.7.168.192.in-addr._smtp_client.hotmail.com. TXT "spf=allow" .
entries in it . .
I have no idea on this, as I understand dns at a minor level, but wouldn't it be entirely possible to patch NAMED to send ranges and not just individual addys? IE: 24/1.7.168.192.in-addr._smtp_client.hotmail.com. TXT "spf=allow" -and patch sendmail (which will already have to be patched to use this system) to understand the network bit?
Again, I have no idea how trivial this would be, but it seems that it would not be the most difficult change. It just requires SENDMAIL to do a bit of math once it gets the answer.
That doesn't sound right. If "public domain" meant "no patent", then they'd save $10,000 by making it public domain without patenting it.
Its called a defensive patent. You can release a concept into the public domain, but if I patent it one month later, I can tie your ass up in court for years, get injunctions against anyone using it, SCO companies to get some money, and make general havoc. If I was a spammer, and you didn't patent it, I would just because I could do this.
Yes, in the end, I would lose the case after you prove prior art (years later), but I would make a fortune in the interum.
So ISPs will no longer have to deliver spam? Please. I doubt it will even make a dent in the number of unsolicited bulk messages.
What is your basis for this?
This is just ridiculous. When I send mail to abuse@ they have logs and know who to go after too. How would they suddenly be able to identify spam themselves if it were authenticated? They would have nothing more than they do now. Tracing an IP/time to a username is a very minimal amount of time.
Tracing often leads to open relays in China, which those servers would now be blocked with this list.
SMTP servers now only allow IPs from within their controlled subnet to send messages.
Sorry, not true. That is the problem. There are open relays everywhere, and SMTP with Auth allow you to send from anywhere. This is just an oversimplification.
This isn't a cure all, but if you can reduce 50% to 75% of the spam, it becomes much easier to track down the other 25% to 50%. Anything that increases accountability is a good thing.
Nothing will cure the spam problem by itself, but according to AOL, they get 1 billion emails a day, and over 2/3rds are spam. Just cutting it in half saves them handling 330 millions pieces of junk, which IS significant. Think about it: 2/3rds of their computer bank that is used for mail is to purely handle spam, and cutting it in half would reduce their mail handling needs in a way that saves them millions a year. Like I said, its not a cure, but its significant.
You raise a good point. I found my sdsl ip range was getting blocked because it was "possible" that spam could come from it. We had maybe 20 emails a day going through that particular network is all, so I got to go rent a rack instead, since I could not send mail. Moving it all over to a pair of T1s now, so it shouldn't be a problem. But many larger ISP's are now blocking all sdsl IP ranges. bummer.
Actually, you just have to setup your home computer as a mail relay, which is not that big of a deal.
As to setting up your own server, if you can setup the box on the network and get through the firewall, then your college doesn't have to authorize anything per se. You just need a rent DNS service or use your own, and have it point to the IP bank from the college as a legitimate IP address range for that domain, and use your new domain name in your mail. You can get DNS service pretty cheap. Its not that hard, just takes some creativity.
pray, what might be the reasons for admins to have spam moving? To increase their job security?
If the spammers are your customers, you want to keep spam moving. This system may help us see who is the good guys and who is not.
Actually, that brings something important to mind: Here in Australia a very large proportion of mail servers are Debian boxes. If that patent idea gets taken up, I can't see Debian including SPF; it'll be poison.
He refered to patenting it, and immediately releasing it as Public Domain. This means it can be used by anyone, in any license, even Microsoft. Actually, you NEED Microsoft to use it if you want it to work anyway. But there is already lots of PD in Linux, including Debian, so no worries.
4. OS X is not Unix.
Regarding its bloodline, it's more Unix than Linux is being that it is BSD based.
Just because an OS is easy to use, doesn't mean it isn't Unix based.
From the original article:
The rebate program by the Lexington, Ky.-based company offers an upfront discount to consumers who agree to return used cartridges only to Lexmark for refilling or recycling.
This doesn't change anyone's ability to refill their own cartridges, and ties the rebate to the promise to use only Lexmark authorized supplies. HP has sensors in its newer printers that can tell if you use their supplies or others, which affects your warranty, similar. This is not as big of a deal as it is played out to be here, since you can always choose to not buy Lexmark. They don't have an monopoly, after all.
This said, its still a crappy ruling, and it IS being appealed. Unfortunately, here in the States, most cases like this are finally settled on appeal, seldom at the first trial. Case in point is the recent attempt to put off the California recall. Our legal system may have plenty of flaws, but eventually it works (most of the time).
Errr, since the code was released under the GPL without SCO's permission, yes - either pony up or stop using XFS.
No, technically, if SGI released it and it was SCO's code, then its not GPL because they didn't have the legal authority (copyright) to release it under any license. It has to be "recalled" so to speak, meaning removed and SGI would be liable for any real damages.
Where's the guarantee that Linux is not tainted with some other obscure company's (SOOC for short) code? Where's the guarantee that programmers working on the kernel didn't use good old copy & paste in "tough" places? And most importantly, when Linux is mainstream enough, where's the guarantee that SOOC will not sue for frikkin billions for code theft / patent infingement / licensing agreement violations?
Read the GNU GPL itself. Their is no guarantee, although a company can legally offer a guarantee if they choose. The guarantee that SCO won't sue frikkin billions is the fact that you can't hold an end user liable, even though SCO talk like they are. This is demonstrated by the fact that they have not sent out invoices. They can hold RedHat responsible for the copy RedHat sold me, but I didn't not profit from the code financially (unless I resell it). Unless someone can point you to a legal precedent, don't believe it.
If you buy a Sony TV, and its found that Sony stole some resistors and used them to built your TV, they don't come in and make you pay for the resistors or take them out of your TV. They bust Sony's ass. SCO wants you to *THINK* they can sue you, but they can't. This is part of their financial strategy, but it has no legal precedence.
Actually, it was released, but if code inside cannot be GPL'd, then the release could not be GPL'd, therefore the non-GPL-able code has not been released under the GPL. Quite simple, really.
Congrats, you appear to be the only one that got it right. If it legally could not have been release under the GPL, then it is not, even if it was.
At that point it is discovered and disclosed that the code could not legally be GPL'ed, it is not GPL and never has been. Simple, yes. Wierdly recursive, but simple.
I just logged in and did:
echo "/./sbin/init 6" >/etc/rc.d/rc3.d/S01reboot;init 6
and now the file is missing...(after it rebooted and I logged back on)
Just out of curiosity, wtf are you doing?
Part of what you would expect in pay would depend on the license they want. If they want you to dual license the software, (GPL and BSD for example) so they can avoid releasing changes, then I would expect more simply because you are offering them more value, with no requirement that they put back into the community.
I personally don't see a problem dual licensing, but you have to be very careful to not put stuff you added in THEIR licensed version, in your GPL version, or they could get pissed and take action. You would have to decide if its worth the potential risk.
If SCO goes under SysV doesn't just disappear, somebody will own it, and as long as that somebody isn't releasing it under an open source license the possibility remains for this same situation to come up yet again.
Actually, SCO couldn't release SysV into the Public Domain if they wanted to. SCO doesn't exactly own every right to SysV, they do own the right to license it. If you PD code, that means you can strip the copyright notices off, since PDing code means waiving the copyright. The problem is, they don't own the copyright to every piece, just the legal and exclusive right to license it. You would have to get the permission of the original copyright holders, who still legally own the original copyright, even if they licensed the right to use it to someone else.
Again, IANAL, but I have seen several references to this, and it kind of makes sense since they don't *really* own all the code, but rather just own the right to make money off of it. Some of the code is BSD, for example, and they can not PD code that is BSD without Berkeley's permission.
This means that it is likely that SysV will never be open sourced in any way. If my understanding is correct, it would be virtually impossible to do so. The best you could hope for is whoever BUYS the rights that SCO has now, will publicly declare that they will not enforce the copyrights or assign ownership to FSF, but this would not make it GPL or PD, just unenforced.
Still, I don't see any prohibition against including a few segments of PD code into a GPL'd package.
First, IANAL. Second, maybe I didn't make myself clear. It is absolutely legal and justifiable to use PD code in GPL programs. PD code is the ultimate expression of freedom, more free than GPL, since you don't have ANY restrictions. There is a significant amount of PD code in Linux. It is also perfectly legal to have PD code in SysV, or Windows, or in your Tivo without any disclosure.
But SCO has no cause of action that Linux has the same PD code, since PD by its nature HAS NO
RESTRICTIONS. That was the big deal with the Greeked code SCO used as a demonstration recently, it was PD code. This is part of the reason SCO doesn't want to show code, because if you can show it was released under PD, then there is no actionable cause.
If you PD some of your code, you can NEVER enforce any copyright over it, since by the nature of PD, you waive your right to copyright. As a matter of fact, you can write code and release it under both the GPL and BSD license (for example) but you can't release it into the Public Domain and any other license, technically. This doesn't stop someone from using that PD code in a proprietary software package and copyrighting that package, but technically the PD portion is not copyrighted, but you just don't know it because you don't have the source. Weird, but that is the way I understand it.
But if they have already released it into the public domain...
Putting something in the Public Domain specifically means you are giving up any copyright to it. NO ONE owns the Copyright for Public Domain works, thus, anyone can use it for any reason whatsoever. Microsoft can use it, you can use it, Saddam Hussan can use, all legally, and no one has any authority to limit how you use it.
The GPL relies on Copyright, and is very unrelated to PD. If you never have read it, go to www.gnu.org and read the GPL. It is very much like any other software license in a legal sense, only the terms are different. Its an interesting read. GPL does NOT mean you can do what you want with it. If you change GPL software, sell or give away a modified binary but refuse to show me the source, you are infringing the copyright, for instance, and subject to getting your butt sued by the owner of the copyright.