Just to be clear, the BSD license is also a Free Software license as defined by the FSF, even though they would prefer software be released under copyleft licenses like the GPL. In general, the differences between the OSI definition of Open Source licenses and the FSF definition of Free Software licenses are very small and tend to turn on relatively minor details.
Correct, although it's user-mode only and so cannot run e.g. an x86 OS. This simplification goes a long way to allowing good performance even without the costly hardware-assist mechanisms Transmeta provided for CMS.
One of the advantages of being Intel is that people will port their OS to your new instruction set.
Around 2000, Linux was already reported at over 30% and to rise even more.
By whom? Do you have a link?
How is it possible that it is expected to account for only 26% of shipments in 2008?
Easy: IDC changed their counting methods in the meantime, while the earlier numbers are about shipments, the current numbers are about revenue and only for server-hardware that actually ships with Linux.
Sorry, no. The 25.7% figure is by unit shipments, not revenue. If you work out the numbers, it looks like they're predicting pre-installed Linux to have ~17% revenue share of the server market in 2008.
Not an emulator; they are porting OpenVMS to Itanium. (They are not, as far as I know, emulating the Alpha instruction set; apps will need to be recompiled.)
The CDDL holds that anyone redistributing CDDL-licensed software cannot assert a patent claim against any other contributor to that software without breaching the license and forfeiting his ability to redistribute the software.
This constitues a "further restriction" on the recipient of the software beyond the restrictions set forth by the GPL, and is thus incompatible with section 6 of the GPL.
These sorts of patent amnesty clauses are generally considered a Good Thing, and are common in many newer open source licenses (the MPL of course, and the CPL, etc.); something like this is pretty likely to be part of v3 of the GPL. On the other hand, very clearly Sun deliberately chose to use a GPL-incompatible license because they don't want Solaris's unique features ending up in Linux. (Otherwise they could have dual-licensed under CDDL and GPL, like Mozilla does with the MPL.)
Bug #253831 is the tracking bug for sites that evade the pop-up blocker. You should add any new sites in a comment there. (Incidentally, you should try to make sure your PC isn't infected with spyware first, as none of the examples there seem to correspond to what you're reporting.)
Unless they've changed it since the last time it was exposed in the preferences (0.9.3 I think), auto-update checks for new browser releases once a day and new revisions of extensions and themes once a week.
How can you proclaim a browser's strength as its extensibility, and then have this 'don't use extensions' nonsense every time you upgrade version?
Becasue this is a testing release only. Users are encouraged to download RC1 only if they are willing to file bugs; normal end users are supposed to stay with 1.0 PR until 1.0 final is available.
That's why this story needed to provide an FTP link to RC1--it's not available via the mozilla.org web page or the Firefox autoupdate. The next couple weeks between RC1 and final are intended to allow extension authors to certify compatability and/or make needed changes. Then when 1.0 final hits, autoupdate should upgrade 1.0 PR users and their extensions automagically.
It's fixed in the trunk, but will not be fixed in the FF 1.0 branch because the current patch caused too many regressions and would require a serious QA effort to track them all down.
Feel free to vote for it, but it's way too late to make it into FF 1.0. Besides, it's already recognized as a high-priority bug. If you really want to help move this into Firefox as quickly as possible, download a trunk build and isolate test cases for regressions.
It would be terrible from MS's standpoint if some GPL'ed code accidentally made its way into Windows, which results in the open-sourcing of some component they don't want open.
Putting GPL'd code in non-GPL'd software does not result in that software being automagically GPL'd. It is just a simple case of copyright infringement.
While GPLing the software in question would bring the infringer into compliance, the more likely outcome is that they pay a settlement to the copyright holder and agree remove the infringing code in a reasonably quick fashion. Even if the case goes to court, there is no way to compel the infringer to GPL their code, only to remove the infringing code.
This is a common misconception, and it may be one that certain people at Microsoft are laboring under, but it is simply untrue.
I was under the impression that the entire notice is presented under penalty of perjury which was the only reason the whole "guilty until proven innocent" was allowed.
There is no "guilty until proven innocent" aspect to the law, at least not in any criminal sense. Perhaps you mean that, in order to retain their "safe harbor" status, an ISP is obligated to take down the alleged infringing material as long as the person distributing it does not protest the takedown notice. If so, realize that that involves a civil contract between the user and the ISP, not a criminal judgement of infringement.
If, on the other hand, a DMCA notice is contested and criminal charges result, then of course the standard is "innocent until proven guilty."
Under the DMCA, the notices are sent under the penalty of perjury.
The only part of the letter that has to be declared under penalty of perjury is that the letter writer is the authorized representative of the copyright holder--which is true in this case. The assertion that the files in question violate that copyright does not have to be made under penalty of perjury.
I don't know about how it works where he is from, but in my town, there's a good chance that you pay the cabbie with a credit card.
It was a New York City cab. They don't take credit cards (yet).
Aren't cabbies legally required to record pickup and destination sites?
Yes. That information would be of essentially no use in tracking down someone in New York.
Incidentally, I have two "stuff left in an NYC yellow taxi" stories, and they both ended well. One night I left my cell phone in a cab. 45 minutes later I was picking it up from the apartment of the guy who got in the cab after me, but of course that was easy for obvious reasons.
The second story is more impressive: a good friend in from Europe on business left her day's purchases in the back of a cab. Three or four shopping bags worth, some relatively expensive clothes, credit card receipts (several of which had full CC information), the whole works. Not good.
Luckily she was expensing the cab ride, so she had taken what passes for a receipt from the cab meter; apparently that includes the cab number, so a few days later (she'd gone back to Italy in the meantime) she was able to call the Taxi and Limousine Commission and get the cabbie's contact information, which she gave to me, which eventually ended up in the cab driver dropping her stuff off at my apartment about a week later.
The amazing thing is, as near as I could understand his overexcited broken English, the cab driver had actually been trying to get ahold of her, via the name on her credit card. (VISA security was--as they ought to have been--entirely unhelpful; but the attempt was at least impressive.) And he wouldn't even accept a fare or a tip for driving her stuff to my apartment! (He was in the neighborhood anyway, he said.)
What conclusions can we draw from all this?
Cab drivers like to see left items returned to their rightful owners; but there is no systematic way to do this (the LTC doesn't collect lost and found)
New Yorkers--or at least the ones taking cabs late at night--tend to be very honest people
Barring the extremely unlikely chance that Jordan took his receipt with him from his cab ride (unless he was expensing it, which seems doubtful, almost no one does), there is no concievable way he could have recovered his memory card
If they're going to produce a 'component-ized' Mozilla, why don't they also put out a stand-alone version of Composer??
They do. Sort of. Nvu is a standalone web authoring system based on a fork of Mozilla Composer. The project head is Daniel Glazman, who was lead developer for Composer. I have not used Nvu, but it seems to have added a number of significant (and IMO much needed) features to Composer's base (e.g. CSS editing and site managment).
The development is sponsored by Linspire, not the Mozilla Foundation, so the project arguably loses out in terms of branding, marketing, integration with Mozilla's bugzilla, etc. to an official Foundation project like Sunbird (the standalone calendar component). On the other hand, planet.mozilla.org (and thus the mozilla.org frontpage) syndicates Glazman's blog posts.
Just so it's clear, the $31M cap does not include payments that have already been made; the "total" in "total legal costs" refers to the fact that the cap would apply to all the firms representing SCO, not just Boies, Schiller and Flexner.
This is all made a bit more complicated by the facts that SCO currently owes something like $8M in unpaid bills for legal services already rendered--which apparently is covered by the cap; that the detailed terms of the deal haven't been released yet; and that in fact the details haven't been agreed upon yet (so far there's only a signed letter of intent). But the bottom line, confirmed at yesterday's conference call, is that SCO currently has ~$43M of cash on its balance sheet, and that assuming their future legal expenses hit the cap, they have ~$12M left to run the rest of their business--roughly 4 months of operating expenses at last quarter's burn rate.
In other words, barring some last minute capital infusion, SCO will run out of cash well before they hit the spending cap--unless, of course, they plan on dropping the pretense of running a business outside of their lawsuits.
If the author provides a "time limited" or "evaluation" version, but only provides source code upon purchase, then he's in a gray area.
No he's not. If he's distributing a binary containing GPL'd code he needs to make the source available to anyone to whom he distributes that binary. Period, end of story.
(Yes, the GPL does allow one to charge for the service of distributing source, but "no more than your cost of physically performing source distribution", which in this day and age is $0.)
And they hold no debt, making it appear that the market is giving them a negative enterprise value.
But not quite: that's in terms of their balance sheet as of the end of the last quarter. They don't have $61M any more--beside the legal expenses of ~$4M/quarter, they've spent $13M buying out part of Baystar's investment.
So sadly it appears the market still thinks SCO is worth something other than the cash they have on hand. After all, there's also office furniture, real estate, etc.
As for liquidating their "business" and distributing it all as a one-time dividend, I wonder if an admission at this point that they have absolutely no case might further expose them to shareholder lawsuits and so forth
BTW, besides acting as a licensing go-between, SCO also bought the exclusive right to develop and sell licenses to any new versions of Unix System V.
Correct, this is what SCO sells under the name UnixWare. (Indeed, Novell had already started marketing under the UnixWare name before they sold it to SCO; version wise, it was System V version 4.2 at the time of the sale.)
So SCO owns Unix System V Release 4. Although I don't believe they would own the copyright on any code in Unix SVR4 that was previously in Unix SVR3.2(or anything before SVR4 for that matter). So, to the extent they added new code SCO owns that but nothing previous to it.
Well not exactly. The original Novell-SCO Asset Purchasing Agreement specifically stated that Novell maintained all copyright in System V software. A year later, the infamous Amendment 2 to the APA changed that to state that Novell would retain all UnixWare copyrights except those "required for SCO to exercise its rights with respect to the acquisition of UNIX and UnixWare technologies." Those particular copyrights were never identified, nor were they explicitly transferred. Of course, this dispute is at the center of the SCO-Novell Slander of Title case, although it probably won't be adjudicated there because all Novell needs to get the Slander of Title claim thrown out is for their interpretation to be reasonable, not that it necessarily be correct.
Perhaps that's what you're saying, with the caveat that what SCO bought from Novell was System V 4.2/UnixWare and not System V 3.2.
they did pay $100 million for something.
Right. They got the UnixWare source code and all in-progress development, the UnixWare trademark, the full right to any new licenses of UnixWare or any UnixWare derivative they created, the full right to Novell's exclusive service and maintenance contracts with all its licensees, copyright to all the various UnixWare manuals, the 5% fee for administrating pre-existing licenses on Novell's behalf, and other assorted odds and ends.
Whether all that (without the copyright to the pre-existing System V/UnixWare codebase) was worth $100 million in 1995 is an interesting question. Hindsight would seem to indicate not. On the other hand, hindsight has vindicated the size and importance of the Unix-on-x86 market, which at the time, between Xenix/SCO Unix/OpenServer (which SCO already owned) and System V/UnixWare, SCO seemed to have all to itself. (haha)
Plus they didn't need the System V 4.2 and earlier copyrights to execute the UnixWare business, because they had a perpetual right to use, modify, and license the System V/UnixWare code. Really the only reason they would need the copyrights is if they wanted to sue people.
"Second, even if the AT&T agreements could be read to preclude the disclosure of homegrown code - and they cannot be - any breach based upon such a reading has been waived by Novell on behalf of SCO, and by SCO itself," IBM says in the filing, referring to the time in which Novell owned the rights to Unix before SCO.
Totally wrong. Instead, IBM is referring to the fact that Novell only sold some Unix rights to Old SCO, and that among the rights they retained was the right to direct SCO to "amend, supplement, modify or waive any rights under, or assign any rights to, any System V License". And, in the event that SCO refused, to do so themselves.
More broadly, SCO did not buy the existing System V licensing business from Novell. They only bought the right to sell new Unix licenses, and also to handle administrative duties on existing licenses in exchange for a 5% commission. IBM's Unix license is still with Novell, not SCO. SCO acts as a go-between on Novell's behalf, but that's the extent of their role.
The amount of abject ignorance on display in this thread is staggering, even for Slashdot. Just to hit a quick list of misconceptions and misanalysis:
a) A unified Xeon/Itanium socket for Tukwila has been on Intel's official roadmap for at least a year now, and obviously has been in the works for much longer. This is not news and indeed Intel has been hyping the hell out of it for some time now.
b) The point of this is not to somehow "retire" Itanium (what, by giving Itanium customers the option to "upgrade" to Xeon??) but to drive it into the mainstream by dramatically lowering platform costs. Intel claims Itanium will get 2x the performance on the same socket as Xeon in 2007. Obviously this will depend on the workload, but with Tukwila going against an dual-core MPU based on the ancient NetBurst core, it's not unlikely either. Of course the reason to stay x86 will be for binary compatability, but in many if not most server situations IA-32 EL on Tukwila should provide better x86 performance than the top end Xeon. The point of a unified socket is to phase out Xeon, not Itanium.
Meanwhile, Slashdot has managed to miss today's announcement that the likely future fastest supercomputer in the world will be running Linux. Seems like a slam dunk, right? Linux running the fastest single-image computer in the world? What's the catch? It's running on Itanium, of course.
Must be tough trying to come up with a negative spin on that one. From the but-if-it-was-opteron-it-would-cost-$100,000-less dept., perhaps??
All the same, it took a volunteer paralegal 24 hours to come up with overwhelming evidence against their latest claims. Certainly they could have at least come up with something more substantial, or in lieu of that, more vague.. to base their claims on.
So what are they trying to pull? There's more involved here, and I think that it might be really important to understand what it is before they show their hand.
There's a hearing scheduled for Aug. 4 on IBM's request for summary judgement on its 10th counterclaim, which states that Linux does not violate any of SCO's copyrights. The summary judgement phase comes at the end of the discovery phase, and before the trial phase of a court case. Now the trial phase is when controversial issues of fact are decided. So a claim can only be ruled on in summary judgement if, based only on issues of law (interpreted neutrally), with the evidence presented in discovery interpreted in as favorable a light as possible towards the losing side, the judge still concludes that the losing side has no chance of succeeding at trial on the merits.
In other words, the way you avoid losing on summary judgement is to make it seem like there is a controversy of facts that it will take a trial to resolve. And that's what SCO's trying to do by throwing in as many crap theories of why Linux might be infringing their copyrights as possible. So long as the judge thinks there's the tiniest chance they could prove any of those claims, she has to deny IBM's summary judgement motion.
Now, SCO knows full well that, even if they make it to trial, they're not going to win any copyright claims against IBM regarding Linux. Indeed, one of their tactics in arguing against IBM's counterclaim is to say that they haven't based their case against IBM on a copyright argument but rather on an argument that IBM somehow violated their UNIX contract with AT&T, which SCO is now successor in interest to.
The problem for SCO is that their SCOSource "business" (such as it is) relies entirely on the fear that Linux does infringe on SCO's copyrights. Even if SCO somehow did win against IBM on the contract violation issue, that wouldn't make anyone other than IBM liable for using or distributing Linux. (Nor would a copyright-based win make them liable for merely using Linux, but it sounds plausible.)
If IBM were to win summary judgement on Aug. 4, it would be very widely publicized that SCO's copyright claims on Linux had been completely thrown out, and the entire premise for SCOSource would be wiped out. So, even though they don't intend to argue any copyright claims against Linux in this or any other case--for the simple fact that they don't have any and they know it--SCO really doesn't want to lose this counterclaim, and moreover even if they have to eventually lose it after the trial's over a couple years from now, they really really don't want to lose it on summary judgement in a few weeks.
Hence the scramble to muddy the waters and create whatever potentially controversial facts they can. Of course the sort of analysis on Groklaw should--assuming it holds up and IBM presents it well--carry the day and allow for summary judgement. But the main tactic of civil law is to throw around as much shit as possible and hope that some of it sticks; worst case scenario you waste your opponents' time having to defend against it.
Weren't the Autozone and DaimerChrysler suits very similar in nature?
SCO represented them that way in the media, but in fact the only similar thing about them was that they were baseless fishing expeditions that involved former customers of SCO/SCO's predecessors in Unix interest.
The AutoZone case is about SCO's charge--based on absolutely no evidence--that AutoZone simply could not have migrated their custom POS system from SCO OpenServer to Linux without also copying the OpenServer static libraries it was previously compiled against. Of course AutoZone insists the ported application is compiled against Linux libraries. SCO has also argued to some degree that AutoZone's very use of Linux violates their copyrights, which is why the case is stayed pending resolution of IBM's 10th counterclaim, but they have been backing away from this argument lately to try to avoid the stay and because the IBM case has shown they have no evidence that Linux infringes their copyrights.
The Daimler case was brought because as successor in interest to Chrysler, who at one point a very long time ago had had a Unix System V license from AT&T, Daimler had failed to respond to a letter from SCO's legal department demanding they tell SCO the serial numbers of all the computers they had running Linux, among other things. In court today the judge ruled that the only thing Daimler could possibly have had to tell SCO was what the AT&T-Chrysler contract required, namely what computers they had running Unix System V, which at this point is, obviously, none. She has not ruled on the question of whether Daimler was at fault in not responding until SCO filed a suit against them, although given that SCO did not bother following up with a second letter or phone call, and that the AT&T contract did not specify a time period in which to respond, it doesn't seem like there's much to that side of the case either.
When you say RedHat and Autozone are stayed until the "copyright issues" are resolved, does that mean that they could potentially start moving again once the August 4 judgement comes in the IBM case?
Yes, particularly if IBM wins. It's worth noting that if IBM does not get summary judgement on their 10th counterclaim (that Linux does not infringe any SCO copyrights) that doesn't mean they've lost the counterclaim, only that there are facts in dispute and thus evidence must be presented for a jury to decide.
When you say it's "scheduled", do you mean the judge will release a decision or do you mean that there will be a hearing and the decision will come about a month later?
A hearing is scheduled for the 4th. The judge can issue a decision then, as the Daimler judge did today, or think about it and decide later.
Slashdot Mirror
Just to be clear, the BSD license is also a Free Software license as defined by the FSF, even though they would prefer software be released under copyleft licenses like the GPL. In general, the differences between the OSI definition of Open Source licenses and the FSF definition of Free Software licenses are very small and tend to turn on relatively minor details.
Correct, although it's user-mode only and so cannot run e.g. an x86 OS. This simplification goes a long way to allowing good performance even without the costly hardware-assist mechanisms Transmeta provided for CMS.
One of the advantages of being Intel is that people will port their OS to your new instruction set.
Not an emulator; they are porting OpenVMS to Itanium. (They are not, as far as I know, emulating the Alpha instruction set; apps will need to be recompiled.)
The CDDL holds that anyone redistributing CDDL-licensed software cannot assert a patent claim against any other contributor to that software without breaching the license and forfeiting his ability to redistribute the software.
This constitues a "further restriction" on the recipient of the software beyond the restrictions set forth by the GPL, and is thus incompatible with section 6 of the GPL.
These sorts of patent amnesty clauses are generally considered a Good Thing, and are common in many newer open source licenses (the MPL of course, and the CPL, etc.); something like this is pretty likely to be part of v3 of the GPL. On the other hand, very clearly Sun deliberately chose to use a GPL-incompatible license because they don't want Solaris's unique features ending up in Linux. (Otherwise they could have dual-licensed under CDDL and GPL, like Mozilla does with the MPL.)
Bug #253831 is the tracking bug for sites that evade the pop-up blocker. You should add any new sites in a comment there. (Incidentally, you should try to make sure your PC isn't infected with spyware first, as none of the examples there seem to correspond to what you're reporting.)
3 83 1
https://bugzilla.mozilla.org/show_bug.cgi?id=25
Yes, thanks for the tip. And the default value for extensions.update.interval is 604800000 milliseconds, which is one week.
Unless they've changed it since the last time it was exposed in the preferences (0.9.3 I think), auto-update checks for new browser releases once a day and new revisions of extensions and themes once a week.
Becasue this is a testing release only. Users are encouraged to download RC1 only if they are willing to file bugs; normal end users are supposed to stay with 1.0 PR until 1.0 final is available.
That's why this story needed to provide an FTP link to RC1--it's not available via the mozilla.org web page or the Firefox autoupdate. The next couple weeks between RC1 and final are intended to allow extension authors to certify compatability and/or make needed changes. Then when 1.0 final hits, autoupdate should upgrade 1.0 PR users and their extensions automagically.
217527.
It's fixed in the trunk, but will not be fixed in the FF 1.0 branch because the current patch caused too many regressions and would require a serious QA effort to track them all down.
Feel free to vote for it, but it's way too late to make it into FF 1.0. Besides, it's already recognized as a high-priority bug. If you really want to help move this into Firefox as quickly as possible, download a trunk build and isolate test cases for regressions.
While GPLing the software in question would bring the infringer into compliance, the more likely outcome is that they pay a settlement to the copyright holder and agree remove the infringing code in a reasonably quick fashion. Even if the case goes to court, there is no way to compel the infringer to GPL their code, only to remove the infringing code.
This is a common misconception, and it may be one that certain people at Microsoft are laboring under, but it is simply untrue.
If, on the other hand, a DMCA notice is contested and criminal charges result, then of course the standard is "innocent until proven guilty."
Yes. That information would be of essentially no use in tracking down someone in New York.
Incidentally, I have two "stuff left in an NYC yellow taxi" stories, and they both ended well. One night I left my cell phone in a cab. 45 minutes later I was picking it up from the apartment of the guy who got in the cab after me, but of course that was easy for obvious reasons.
The second story is more impressive: a good friend in from Europe on business left her day's purchases in the back of a cab. Three or four shopping bags worth, some relatively expensive clothes, credit card receipts (several of which had full CC information), the whole works. Not good.
Luckily she was expensing the cab ride, so she had taken what passes for a receipt from the cab meter; apparently that includes the cab number, so a few days later (she'd gone back to Italy in the meantime) she was able to call the Taxi and Limousine Commission and get the cabbie's contact information, which she gave to me, which eventually ended up in the cab driver dropping her stuff off at my apartment about a week later.
The amazing thing is, as near as I could understand his overexcited broken English, the cab driver had actually been trying to get ahold of her, via the name on her credit card. (VISA security was--as they ought to have been--entirely unhelpful; but the attempt was at least impressive.) And he wouldn't even accept a fare or a tip for driving her stuff to my apartment! (He was in the neighborhood anyway, he said.)
What conclusions can we draw from all this?
They do. Sort of. Nvu is a standalone web authoring system based on a fork of Mozilla Composer. The project head is Daniel Glazman, who was lead developer for Composer. I have not used Nvu, but it seems to have added a number of significant (and IMO much needed) features to Composer's base (e.g. CSS editing and site managment).
The development is sponsored by Linspire, not the Mozilla Foundation, so the project arguably loses out in terms of branding, marketing, integration with Mozilla's bugzilla, etc. to an official Foundation project like Sunbird (the standalone calendar component). On the other hand, planet.mozilla.org (and thus the mozilla.org frontpage) syndicates Glazman's blog posts.
As of version 0.10 (a.k.a. 1.0PR) Firefox does this as well.
Just so it's clear, the $31M cap does not include payments that have already been made; the "total" in "total legal costs" refers to the fact that the cap would apply to all the firms representing SCO, not just Boies, Schiller and Flexner.
This is all made a bit more complicated by the facts that SCO currently owes something like $8M in unpaid bills for legal services already rendered--which apparently is covered by the cap; that the detailed terms of the deal haven't been released yet; and that in fact the details haven't been agreed upon yet (so far there's only a signed letter of intent). But the bottom line, confirmed at yesterday's conference call, is that SCO currently has ~$43M of cash on its balance sheet, and that assuming their future legal expenses hit the cap, they have ~$12M left to run the rest of their business--roughly 4 months of operating expenses at last quarter's burn rate.
In other words, barring some last minute capital infusion, SCO will run out of cash well before they hit the spending cap--unless, of course, they plan on dropping the pretense of running a business outside of their lawsuits.
No he's not. If he's distributing a binary containing GPL'd code he needs to make the source available to anyone to whom he distributes that binary. Period, end of story.
(Yes, the GPL does allow one to charge for the service of distributing source, but "no more than your cost of physically performing source distribution", which in this day and age is $0.)
And they hold no debt, making it appear that the market is giving them a negative enterprise value.
But not quite: that's in terms of their balance sheet as of the end of the last quarter. They don't have $61M any more--beside the legal expenses of ~$4M/quarter, they've spent $13M buying out part of Baystar's investment.
So sadly it appears the market still thinks SCO is worth something other than the cash they have on hand. After all, there's also office furniture, real estate, etc.
As for liquidating their "business" and distributing it all as a one-time dividend, I wonder if an admission at this point that they have absolutely no case might further expose them to shareholder lawsuits and so forth
Correct, this is what SCO sells under the name UnixWare. (Indeed, Novell had already started marketing under the UnixWare name before they sold it to SCO; version wise, it was System V version 4.2 at the time of the sale.)
Well not exactly. The original Novell-SCO Asset Purchasing Agreement specifically stated that Novell maintained all copyright in System V software. A year later, the infamous Amendment 2 to the APA changed that to state that Novell would retain all UnixWare copyrights except those "required for SCO to exercise its rights with respect to the acquisition of UNIX and UnixWare technologies." Those particular copyrights were never identified, nor were they explicitly transferred. Of course, this dispute is at the center of the SCO-Novell Slander of Title case, although it probably won't be adjudicated there because all Novell needs to get the Slander of Title claim thrown out is for their interpretation to be reasonable, not that it necessarily be correct.
Perhaps that's what you're saying, with the caveat that what SCO bought from Novell was System V 4.2/UnixWare and not System V 3.2.
Right. They got the UnixWare source code and all in-progress development, the UnixWare trademark, the full right to any new licenses of UnixWare or any UnixWare derivative they created, the full right to Novell's exclusive service and maintenance contracts with all its licensees, copyright to all the various UnixWare manuals, the 5% fee for administrating pre-existing licenses on Novell's behalf, and other assorted odds and ends.
Whether all that (without the copyright to the pre-existing System V/UnixWare codebase) was worth $100 million in 1995 is an interesting question. Hindsight would seem to indicate not. On the other hand, hindsight has vindicated the size and importance of the Unix-on-x86 market, which at the time, between Xenix/SCO Unix/OpenServer (which SCO already owned) and System V/UnixWare, SCO seemed to have all to itself. (haha)
Plus they didn't need the System V 4.2 and earlier copyrights to execute the UnixWare business, because they had a perpetual right to use, modify, and license the System V/UnixWare code. Really the only reason they would need the copyrights is if they wanted to sue people.
Totally wrong. Instead, IBM is referring to the fact that Novell only sold some Unix rights to Old SCO, and that among the rights they retained was the right to direct SCO to "amend, supplement, modify or waive any rights under, or assign any rights to, any System V License". And, in the event that SCO refused, to do so themselves.
More broadly, SCO did not buy the existing System V licensing business from Novell. They only bought the right to sell new Unix licenses, and also to handle administrative duties on existing licenses in exchange for a 5% commission. IBM's Unix license is still with Novell, not SCO. SCO acts as a go-between on Novell's behalf, but that's the extent of their role.
The amount of abject ignorance on display in this thread is staggering, even for Slashdot. Just to hit a quick list of misconceptions and misanalysis:
a) A unified Xeon/Itanium socket for Tukwila has been on Intel's official roadmap for at least a year now, and obviously has been in the works for much longer. This is not news and indeed Intel has been hyping the hell out of it for some time now.
b) The point of this is not to somehow "retire" Itanium (what, by giving Itanium customers the option to "upgrade" to Xeon??) but to drive it into the mainstream by dramatically lowering platform costs. Intel claims Itanium will get 2x the performance on the same socket as Xeon in 2007. Obviously this will depend on the workload, but with Tukwila going against an dual-core MPU based on the ancient NetBurst core, it's not unlikely either. Of course the reason to stay x86 will be for binary compatability, but in many if not most server situations IA-32 EL on Tukwila should provide better x86 performance than the top end Xeon. The point of a unified socket is to phase out Xeon, not Itanium.
Meanwhile, Slashdot has managed to miss today's announcement that the likely future fastest supercomputer in the world will be running Linux. Seems like a slam dunk, right? Linux running the fastest single-image computer in the world? What's the catch? It's running on Itanium, of course.
Must be tough trying to come up with a negative spin on that one. From the but-if-it-was-opteron-it-would-cost-$100,000-less dept., perhaps??
Atleast AMD has a good idea of what they are going to be doing a few years from now.
Really? Can you point me towards those official public AMD roadmaps detailing their platform infrastructure in 2007?
Give me a break.
All the same, it took a volunteer paralegal 24 hours to come up with overwhelming evidence against their latest claims. Certainly they could have at least come up with something more substantial, or in lieu of that, more vague.. to base their claims on.
So what are they trying to pull? There's more involved here, and I think that it might be really important to understand what it is before they show their hand.
There's a hearing scheduled for Aug. 4 on IBM's request for summary judgement on its 10th counterclaim, which states that Linux does not violate any of SCO's copyrights. The summary judgement phase comes at the end of the discovery phase, and before the trial phase of a court case. Now the trial phase is when controversial issues of fact are decided. So a claim can only be ruled on in summary judgement if, based only on issues of law (interpreted neutrally), with the evidence presented in discovery interpreted in as favorable a light as possible towards the losing side, the judge still concludes that the losing side has no chance of succeeding at trial on the merits.
In other words, the way you avoid losing on summary judgement is to make it seem like there is a controversy of facts that it will take a trial to resolve. And that's what SCO's trying to do by throwing in as many crap theories of why Linux might be infringing their copyrights as possible. So long as the judge thinks there's the tiniest chance they could prove any of those claims, she has to deny IBM's summary judgement motion.
Now, SCO knows full well that, even if they make it to trial, they're not going to win any copyright claims against IBM regarding Linux. Indeed, one of their tactics in arguing against IBM's counterclaim is to say that they haven't based their case against IBM on a copyright argument but rather on an argument that IBM somehow violated their UNIX contract with AT&T, which SCO is now successor in interest to.
The problem for SCO is that their SCOSource "business" (such as it is) relies entirely on the fear that Linux does infringe on SCO's copyrights. Even if SCO somehow did win against IBM on the contract violation issue, that wouldn't make anyone other than IBM liable for using or distributing Linux. (Nor would a copyright-based win make them liable for merely using Linux, but it sounds plausible.)
If IBM were to win summary judgement on Aug. 4, it would be very widely publicized that SCO's copyright claims on Linux had been completely thrown out, and the entire premise for SCOSource would be wiped out. So, even though they don't intend to argue any copyright claims against Linux in this or any other case--for the simple fact that they don't have any and they know it--SCO really doesn't want to lose this counterclaim, and moreover even if they have to eventually lose it after the trial's over a couple years from now, they really really don't want to lose it on summary judgement in a few weeks.
Hence the scramble to muddy the waters and create whatever potentially controversial facts they can. Of course the sort of analysis on Groklaw should--assuming it holds up and IBM presents it well--carry the day and allow for summary judgement. But the main tactic of civil law is to throw around as much shit as possible and hope that some of it sticks; worst case scenario you waste your opponents' time having to defend against it.
Weren't the Autozone and DaimerChrysler suits very similar in nature?
SCO represented them that way in the media, but in fact the only similar thing about them was that they were baseless fishing expeditions that involved former customers of SCO/SCO's predecessors in Unix interest.
The AutoZone case is about SCO's charge--based on absolutely no evidence--that AutoZone simply could not have migrated their custom POS system from SCO OpenServer to Linux without also copying the OpenServer static libraries it was previously compiled against. Of course AutoZone insists the ported application is compiled against Linux libraries. SCO has also argued to some degree that AutoZone's very use of Linux violates their copyrights, which is why the case is stayed pending resolution of IBM's 10th counterclaim, but they have been backing away from this argument lately to try to avoid the stay and because the IBM case has shown they have no evidence that Linux infringes their copyrights.
The Daimler case was brought because as successor in interest to Chrysler, who at one point a very long time ago had had a Unix System V license from AT&T, Daimler had failed to respond to a letter from SCO's legal department demanding they tell SCO the serial numbers of all the computers they had running Linux, among other things. In court today the judge ruled that the only thing Daimler could possibly have had to tell SCO was what the AT&T-Chrysler contract required, namely what computers they had running Unix System V, which at this point is, obviously, none. She has not ruled on the question of whether Daimler was at fault in not responding until SCO filed a suit against them, although given that SCO did not bother following up with a second letter or phone call, and that the AT&T contract did not specify a time period in which to respond, it doesn't seem like there's much to that side of the case either.
When you say RedHat and Autozone are stayed until the "copyright issues" are resolved, does that mean that they could potentially start moving again once the August 4 judgement comes in the IBM case?
Yes, particularly if IBM wins. It's worth noting that if IBM does not get summary judgement on their 10th counterclaim (that Linux does not infringe any SCO copyrights) that doesn't mean they've lost the counterclaim, only that there are facts in dispute and thus evidence must be presented for a jury to decide.
When you say it's "scheduled", do you mean the judge will release a decision or do you mean that there will be a hearing and the decision will come about a month later?
A hearing is scheduled for the 4th. The judge can issue a decision then, as the Daimler judge did today, or think about it and decide later.