So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
Posted Aug 13, 2010 14:52 UTC (Fri) by FlorianMueller (guest, #32048)In reply to: So what do the proponents of Oracle's acquisition of Sun think now? by Karellen
Parent article: Oracle sues Google over use of Java in Android (ars technica)
My reading of that article[0] is that PJ has gone through all the patent pledges, emails, articles and other media, examined them in her professional capability as a paralegal, and come to the conclusion that there is no legal impediment to IBM suing TurboHurcules; i.e. they are free to do so. Note that this is a completely different statement to whether or not she thinks that IBM *should* sue TurboHercules, or whether she thinks it would be good for FLOSS if IBM did sue them.
1. The headline is above the article as a whole and that one not only denies that there's a problem with the pledge (although there is) but also sides with IBM, suggesting that this is just a Microsoft scheme, which is irrelevant anyway for reasons explained here.
2. PJ grossly misrepresented and misinterpreted the pledge. It does have a defense clause, but it was never trigged by anything TurboHercules did, as I explained here. In a nutshell, an antitrust complaint couldn't have triggered the defense clause in IBM's pledge anyway because that one related to the specific scenario of (i) intellectual property (ii) litigation (iii) against open source, and not even one of those three criteria was fulfilled by an (i) antitrust (ii) complaint (iii) against IBM.
IBM have *not* sued or initiated any kind of legal proceedings or threats towards TurboHurcules at this point.
IBM's pledge didn't just say they wouldn't sue. It said they wouldn't assert, which is much broader, as I explained here. Clearly, suing isn't the only way to use patents in a malicious and pernicious manner. The fewest cases ever go to court.
If you are aware of any valid legal reason why IBM could not sue TurboHercules, i.e. why they are not free to do so, could you please post a link to whatever rebuttal you have to PJ's points? (I scanned the comments, but could not see any posts by you there defending your position)
I provided a few links and there's obviously more stuff on the TurboHercules matter on my blog, such as this posting on the entire correspondence between the two parties. There's also the ibmvshercules.com blog that the maintainer of the 11-year-old Hercules open source project set up. He strongly disagrees with PJ and talks about that fact on his blog.
Posted Aug 13, 2010 17:48 UTC (Fri)
by Karellen (subscriber, #67644)
[Link] (22 responses)
Hmmm....but they do own those patents in the area that TurboHercules asked about. How else should they have answered the question?
Should they have denied that they owned any patents in the area, which would be a lie? Don't tell me you wouldn't be ripping even more strips off them than you already are if *that* were the case.
Should they have ignored the question? At which point they'd have had strips ripped off them for failing to communicate in a reasonable fashion with members of the community.
I'll agree that pledging to "not assert" patents is much broader than pledging to "not sue over" patents, but I don't see how pledging to "not assert" must be equivalent to pledging to "deny all knowledge, irrespective of the facts".
Anyway, given that in your opinion IBM answered the question badly, how do you think they should have answered, in a way that is succinct, clear, open, true, and in line with their patent pledge?
Posted Aug 13, 2010 17:59 UTC (Fri)
by FlorianMueller (guest, #32048)
[Link] (21 responses)
@Karellen, as you can see, I'm happy to answer all the questions. One can argue whether this is on-topic or off-topic, but since Oracle vs. Google and IBM vs. TurboHercules are the two most important patents-against-FOSS issues, I can see a connection. In this blog posting I discussed the chronology of events based on the correspondence between the two companies: July 2009: TurboHercules contacts IBM about z/OS, a licensing question that is just a matter of EULA terms, nothing related to patents November 2009: IBM basically says "no your customers won't be allowed to run z/OS in emulation, and it's not a matter of what you or the customers pay, we just don't want it" (by the way, I'm not aware of Microsoft prohibiting the use of Windows in emulation). But additionally, IBM also said: you infringe IP (= patents in this case). So IBM asserted an infringement before TurboHercules asked anything related to patents. You talk about patents IBM has in that area. That's not even what TurboHercules asked in response to IBM's first letter. TurboHercules wanted to know what "IP" it was considered to infringe. Please imagine their situation: the Hercules open source project has been around since 1999, it was mentioned in an IBM RedBook (documentation material) in 2002, and then suddenly in 2009, after so many years without complaining, IBM claimed there was an infringement. As everyone here certainly knows, patents can be infringed unknowingly, while software copyright -- in practical terms -- is only infringed by knowingly copying something. TurboHercules never doubted that IBM had IP in connection with the mainframe. But not having committed any wrongdoing and considering that IBM had not complained in a decade, they certainly were surprised and asked IBM to clarify what IP they might infringe. Then IBM sent the patent list. TH didn't ask for a patent list. The term "IP" could have meant many different things in theory. That's why Richard Stallman rejects it by the way.
Posted Aug 14, 2010 9:42 UTC (Sat)
by Wol (subscriber, #4433)
[Link] (20 responses)
With both patents and copyright, an owner is entitled to ignore (or even encourage?) certain uses. Certainly IBM didn't object to their customers using Hercules (a HOBBYIST, NON-COMMERCIAL TOY).
Then along comes TurboHercules, wanting to make money off of Hercules and IBM. OF COURSE they're going to object.
TH are naive, stupid, or malicious. Take your pick.
Cheers,
Posted Aug 14, 2010 12:34 UTC (Sat)
by FlorianMueller (guest, #32048)
[Link] (19 responses)
TurboHercules's founder, Roger Bowler, founded the Hercules open source project in 1999. Neither he nor anyone he knows received any complaints during that decade. Since TurboHercules's software is 100% the Hercules open source project, we talk about the same patent issues (should there be any infringement, which would have to be proven), and no complaints in about a decade. Please don't forget that you're a hardcore Groklaw user and Groklaw opinion-leader, and here you're on LWN, an open source website. So in a way what you say here reflects on Groklaw, especially because you use Groklaw talking points on the TurboHercules case. And what you say is a total rejection of the concept of software freedom and of most of the basic rights defined by both the Free Software Definition and the Open Source Definition. You argue that patents can be used to restrict use. The same reasoning could be used so that Microsoft, Oracle, Apple etc. should use patents to prohibit the use of Linux on smartphones, doing away with Android at the same time, or that Oracle, IBM and Microsoft should have jointly killed MySQL and PostgreSQL a long time ago. So much for what you said about the right to restrict or encourage uses with patents. The patents IBM listed aren't patents on IBM's proprietary z/OS operating system. They are patents that are either also infringed when z/Linux gets emulated or they're never infringed. Now concerning copyright, as you can see, the European Commission launched a formal investigation because IBM is suspected of illegally tying its mainframe hardware to its z/OS operating system. So copyright holders certainly have to respect antitrust law. When the European Commission required Microsoft to provide a Windows version without the Media Player and to provide, document and license certain interfaces to Samba (and others), that also meant to impose rules on a right holder for the sake of fair and effective competition. That's the rule of law. And since it is the rule of law, right holders aren't just expropriated without compensation the way an authoritarian regime could do it. The law says that in such cases the commercial terms have to be fair, reasonable and non-discriminatory. In other words: they will get paid, but they can't overcharge. I had quoted this as part of the previous quote, but it's so fascinating that I have to quote it again here as a stand-alone quote. Is that what you're going to present as a pro-open-source point of view? Open source is OK as long as it's hobbyist and non-commercial? So the same for Linux, MySQL, Apache, PHP, everything? It comes down to Groklaw's approach: Rule one - patents are bad and open source is good. Rule two - when IBM's patents are involved, patents are good and open source is bad. I think you may have to make more facts-based postings if you want to steer clear of anyone using any of those attributes in connection with the content of your postings.
Posted Aug 19, 2010 0:37 UTC (Thu)
by Wol (subscriber, #4433)
[Link] (2 responses)
Linus studiously avoids having anything to do with anything except Linux. He pays no attention to Windows, AIX or whatever. He just improves his "toy", with the result it is taking on the world on its own merits.
Roger, on the other hand, is paying an unhealthy amount of interest to what someone else (namely IBM) is doing. And (Turbo)Hercules is seriously suffering in consequence.
Cheers,
Posted Aug 19, 2010 3:42 UTC (Thu)
by FlorianMueller (guest, #32048)
[Link] (1 responses)
For someone who touted on this very page his "native" English language skills your orthography leaves a lot to be desired. I'm not talking about a typo but about a mistake you made consistently (twice) in the part quoted above, suggesting that one might add your "native" claim to the long list of untruths you spread here. If you don't know which spelling mistake I mean, look up a dictionary. It doesn't have to be the Merriam-Webster or American Heritage Dictionary since you reject those. While you're at it, you might also want to look up the definition of "to assert" (something you raised elsewhere on this page). If you need help, there's a link in this posting.
Posted Sep 2, 2010 23:18 UTC (Thu)
by nix (subscriber, #2304)
[Link]
(I'm a freak: English and can spell.)
Posted Aug 21, 2010 9:27 UTC (Sat)
by pdickey (guest, #69690)
[Link] (15 responses)
IMHO, there are three things that you're overlooking (from my understanding of this entire case), although if I'm wrong about this, I'd like to see direct clarification (quotes from reputable sources like a licensing agreement).
1. IBM, in their initial response didn't assert patent infringment upon the Hercules project in general. They were suggesting that TurboHercules' implementation of the emulation would be infringing (why? See my #2)
2. In IBM's licensing agreement (from my understanding, and I'm too tired at 0354 to dig the information out for sure), you are allowed to emulate their OS on non-IBM hardware for certain limited uses (disaster recovery, for example). TurboHercules wants to expand that (and change the licensing agreement) to mean that you can run their OS on non-IBM hardware for any reason at all.
3. "Third, we do not see how your product contributes to our business, and IBM offers product and service solutions ourselves for all of the situations you mention in your letter, including training, demonstration, pre- and post processing, data preparation, disaster recovery, archiving, development, and testing. In fact, we would be concerned about loss of functionality and quality that could lead to customer disatisfaction mongst licensees of IBM operating systems."
My amateur interpretation of that quote is this:
"You're suggesting that people may want to use our operating systems in certain fashions. We already provide them a means to do so. So, why should we go into a partnership with someone else to provide something that we are already providing? Plus, if people have problems with YOUR product, they will blame US, because in their eyes, they're not using a TurboHercules product. They're using a licensed copy of our operating system."
In other words, it's like people saying that it's Microsoft's fault that their video card broke. Or more accurately (since this is an Open Source forum), it's like people bitching that it's Ubuntu's fault that their ATI video cards don't work properly (when it's AMD that won't provide the proper drivers for the card).
As for Groklaw, yes I do find some of the points there a little extreme. I don't think that Microsoft is necessarily behind every bad thing in the world (but I do think that as far as tech goes, they aren't upset when the bad things happen to their competitors-- just like any company wouldn't be upset).
That being said.....
She puts up the facts. She links to them, so you can read them for yourself. You are free to form your own conclusions, whether they agree with hers or not. But, she's also free to post her opinions about the facts. After all, you post your opinions about the articles you're writing about (I know I do).
I've learned a few things from Groklaw that aren't just the latest happenings in the legal and tech world... I've learnd that I'm not a lawyer, and I have no desire to become one. I've learned that laws (as they're written) and lawyers make absolutely no sense in the world. And I've learned, especially from reading the ramblings by SCO's lawyers, that Motrin doesn't always cure the headache that I got reading court filings.
And I've learned that you don't have to agree with her. In fact, as long as you do so respectfully, you can tell her you don't agree. You said you've got reams of proof that she censors, and I'll raise you reams of proof that she doesn't (you can find it all on the site). I'm willing to bet if you dig into the site, you'll find comments that disagree with her and other commenters. But, you'll also find that they were respectful of her position on the subject.
Anyhooo, I'm off to bed. (I realize that statement blew what little credibility I have on this site, but oh well. I'm too tired to care).
Have a great weekend:)
Posted Aug 21, 2010 13:06 UTC (Sat)
by FlorianMueller (guest, #32048)
[Link] (13 responses)
You made a long and thoughtful comment and raised many questions. My comments: TurboHercules's implementation is 100% the same software the Hercules open source emulator. We're not talking about a different emulator. In its first reply to TurboHercules, IBM wrote: (quote) First, you state that Turbo "implements the instruction set of IBM mainframes on Intel-based servers". We think that mimicking IBM's proprietary, 64-bit System z architecture requires IBM intellectual property, and you will understand that IBM could not reasonably be asked to consider licensing its operating systems for use on infringing platforms. (end quote) While the word "Turbo" appears there, absolutely everything that is said -- the implementation of the instruction set of the mainframe CPU on Intel-based servers -- relates to the Hercules open source emulator. The maintainer of the Hercules open source project, Jay Maynard, explained this on his blog as well. It's correct that TurboHercules asked for a license agreement in order to run z/OS in emulation for more use cases than disaster recovery. But that's nothing I was overlooking, although you listed it as one of three items I allegedly didn't tell. For an example, my blog posting on the correspondence between the two companies also put this into the context. My very first blog posting on the TurboHercules matter also explained this very clearly: (quote) Hercules just wants to provide customers with an interesting and much-needed choice. In order to do so, Hercules is simply seeking a way that its customers will be allowed to run IBM's z/OS mainframe operating system on Hercules. IBM generally does not allow its customers to run z/OS (hence also the application software those customers internally developed on top of it for trilions of dollars) on non-IBM hardware. When Roger asked IBM for a solution, IBM turned him down and accused Hercules of infringement of intellectual property rights. (end quote)
And here on LWN, I explained on several occasions what the antitrust case is about: it's all about the untying of IBM hardware from the z/OS operating system, in particular in connection with emulation. That's part of one of several reasons IBM gave for turning down the request. It's just a cheap excuse that IBM uses. If people run Windows on virtualization platforms, it may also behave differently than if it resides directly on top of the hardware. Or to address the case that affects almost all of us programmers: if we write something for an operating system, possibly along with other underlying components (database etc.), we have to support our customers (to the extent we're reasonably responsible) although our software isn't "alone" in the system. Note that TurboHercules had not said anything in its letter to IBM about the details of IBM's obligations. TH had only asked for licensing terms in IBM's discretions but fair and reasonable (which is what a monopolist has to be under antitrust rules). In my opinion, "fair and reasonable" obviously means that IBM wouldn't have to end up accepting responsibility for the inner workings of the Hercules emulator. That's where you phrase things in a balanced way but PJ doesn't. The one reason for which I oppose irrational Microsoft paranoia is that if you look the collective power of its key competitors (you could just add up the market capitalizations of those to get an idea that collectively they're several times as powerful as Microsoft), and if you consider that those also commit pretty serious wrongdoings all the time, then it's important to look at all the issues in the industry, emanating from all sources, so as not to be distracted by a bogeyman that actually hasn't become more powerful in about a decade and that has over the last 10+ years had its defining moments in antitrust. Others are yet to experience that, and IBM got off the hook of the 1950's Consent Decree around 2000 and needs to be reminded now. I also link to material I comment on. Nevertheless it's legitimate to criticize someone for out-of-context quoting and gross distortions that are of a nature that an honest activist in her quest for the truth wouldn't make but only someone who's not independent, be it in terms of independent thinking to have reasonable views or otherwise. Linking to the truth isn't enough because most people don't even click through (click-through rates on the Internet are low and everyone knows it). Just an example: in connection with the IBM patent pledge, she started off with something that's utterly disgraceful and no honest FOSS person would do that, only a propagandist of the kind you find (or in the past found) in totalitarian regimes. Frankly, the Iraqi Information Minister was much more honorable with his absurdities than PJ is. So here's just one example of dishonest, undoubtedly intentional propaganda. Her propaganda regarding TurboHercules started with this: So why is this a total Groklie, even though PJ provides links? She quoted from this ZDNet Article of 4 August 2004, but IBM's patent pledge was announced more than five months later on 11 January 2005 (check here or elsewhere on the Web). So she took an IBM statement related to something different -- use of patents against the Linux kernel -- and puts it purposely into a completely different context. If someone doesn't even click through to find out that those are two entirely different contexts, the quote even speaks for itself because it talks about someone's aggression against the Linux kernel and TurboHercules has nothing to do with that. Hercules runs on Linux, peacefully ;-) By contrast, I told people the whole truth: IBM did have a defense clause in its January 2005 open source patent pledge, but there's no way it was triggered by anything TurboHercules did. In this posting I quote from the pledge and explain why. In a nutshell, there's one (and only one) defense clause in IBM's patent pledge. It relates to (i) IP (ii) litigation against (iii) open source. However, all that IBM and PJ could reference in terms of TurboHercules's activities is (i) antitrust [different field of law, entirely] (ii) complaint [just a filing with a regulator who will only pursue if in the public interest; not a court suit] against (iii) IBM [and in particular, its proprietary business]. Any one of the three differences would be enough on its own to make the defense clause in the pledge inapplicable. The combination of three makes it totally absurd to suggest they had a defense scenario under the pledge. Therefore the Groklie that IBM had made a very broadly-phrased exception in terms of defense. That broad "forced to defend" thing was before the pledge, and the language of the open source patent pledge makes it clear that the pledge stands on its own as a commitment. A commitment that IBM betrayed. You can see that PJ was even alerted to that error in the comments. She didn't care. That's another strong indication that she's a propagandist, not an honest open source advocate. An honest person would at least admit the error and correct it, instead of continuing to start off a very long posting with a Groklie. In the discussion on that page she even told the opposite of what the GPL stipulates in terms of selling GPL'd software. Since she talks about the GPL all the time, chances are she knows what the GPL says (and what gnu.org says about selling software). But if it's about defending IBM, then she doesn't care about the truth, really. Finally, our moderator here on LWN asked to stop discussing Groklaw's editorial policy here. So please appreciate I won't comment on your last item here and now (but probably somewhere else, some other time). But regardless, I don't think you've lost your credibility here by simply triyng to find out about many aspects of a big issue.
Posted Aug 21, 2010 13:13 UTC (Sat)
by FlorianMueller (guest, #32048)
[Link]
Posted Aug 22, 2010 0:00 UTC (Sun)
by pdickey (guest, #69690)
[Link] (11 responses)
I took the whole issue (reading your comments, reading different news articles on the issue, and reading about the license agreement) as this:
IBM didn't have an issue with the Hercules Project, because it was mainly (or completely at the time) being used in accordance with the Licensing Agreement (or if it wasn't being used in accordance with the agreement, no one was saying that). Plus the fact that it was free (just my opinion).
Their issue with TurboHercules is different. Not because they have a problem with the Hercules Project, but because TurboHercules wants to use it in a way that is not in accordance with their licensing agreement, and make people pay for that right.
IMHO, isn't that against the code of Open Source software? I read this as:
You can have the software for free, and even see the source code for it (and modify it). However, you have to pay those guys to be able to use the software for anything other than the terms of your License Agreement.
To me, that would be the same as Canonical saying "Here's Ubuntu. You are only allowed to use the non-restricted sources. However, if you want access to the restricted sources, you have to pay LinuxCo2, because they are providing a commercial version of Ubuntu that is allowed to use restricted sources." It wouldn't seem fair, that I'm not legally allowed to put a check in the box next to "Restricted Sources" even though it's available to me--all because I didn't pay someone else for that access.
I would like to point out a bit of irony though. In looking at the letters between TurboHercules and IBM, I noticed that some of the patents were covering Virtual Machines. So, I wonder if those would be capable of being used as Prior Art in the Google v Oracle case. (Which by the way, brings us full circle to the subject of the article we're commenting on). If so, then maybe some good can come of the mess (because IBM essentially laid them out for everyone to see).
Have a great weekend:)
Posted Aug 22, 2010 0:52 UTC (Sun)
by FlorianMueller (guest, #32048)
[Link]
IBM mentioned Hercules in a RedBook (documentation piece) in 2002. It then removed that reference in the next edition. So even many years before the TurboHercules company was founded, IBM changed its stance. In terms of the license agreement at issue, that's the z/OS agreement and chances are that out of many thousands of Hercules users, some perhaps didn't adhere religiously to the z/OS terms. Not that I know of any -- but statistically it's hard to imagine that so many people would all have complied with those terms 100%, all of the time. The "pay" part relates only to z/OS, the proprietary operating system. For proprietary software one has to pay. Of course, if IBM decided to give it away for free, that would be appreciated, but no reasonable person would demand that. Absolutely not. The key requirements for open source are defined by the Open Source Definition (and for free software by the Free Software Definition). This means among other things that you must have the right to modify and to redistribute. You must have access for free -- to the open source program -- if you wish. TurboHercules doesn't violate any of those principles. If you get an Oracle 11g database server running on Linux, the Linux part is open source and Oracle 11g is closed source, running on top of Linux. No. TurboHercules doesn't force you to pay for the Hercules open source program. You can get that free of charge if you wish, and of course you get the source code. The "pay" part relates only to z/OS, which is proprietary. Think of my example with Oracle 11g running on Linux. Oracle 11g is closed source, Linux is open source. To make it even clearer: if you emulate z/Linux with Hercules, then you don't have to pay anyone and there's no restriction preventing you from doing it. Similarly, you can also run MySQL or PostgreSQL on top of Linux, and then it's all free of charge. Those closed-source/open-source combinations are pretty ubiquitous. You only have a problem with a license if it's "copyleft" like the GPL and you don't comply with the requirement to make the source code of a derived work available. In the Linux/Oracle example, note that running applications on top of Linux isn't a "copyleft" case because of an exception. The key respect in which this doesn't relate to the TurboHercules situation is that they don't restrict any Hercules sources or the use of Hercules. If anyone restricts, it's IBM. But again, don't forget that Hercules emulates the CPU instruction set, not the operating system. The operating system runs in emulation. In that emulation you can also run z/Linux. If, however, you want to run z/OS, then that's an issue between you as a customer and IBM. If TurboHercules helps you by organizing a license for all TurboHercules customers, that's also fine -- as long as the open source program continues to be available on open source terms. By running something in emulation you don't create a new combined program. If you run proprietary software on an open source virtualizer like Xen, then the software running in virtualization is also a different thing from the virtualizer. One can be closed source and the other open source. There are probably many many different patents out there on virtual machines. A patent can serve as prior art in principle, but those who now probably look for prior art against those Java patents will look everywhere I'm sure.
Posted Aug 22, 2010 8:09 UTC (Sun)
by anselm (subscriber, #2796)
[Link] (9 responses)
If you want to run Windows (a proprietary OS) in an open-source virtual machine like KVM or VirtualBox, you need to procure a licensed version of Windows. This generally means buying one.
If you want to run z/OS (another proprietary OS) in an open-source virtual machine like Hercules, you need to procure a licensed version of z/OS. This means talking to IBM to see whether they'll let you buy one without an IBM mainframe attached.
There's no difference in my book.
Posted Aug 22, 2010 11:06 UTC (Sun)
by FlorianMueller (guest, #32048)
[Link] (4 responses)
That's exactly right, and the key difference is that Microsoft doesn't leverage its market positions in PC operating systems to tie particular hardware products (or virtualizers) to Windows. But IBM abuses its monopoly that way, and that's why the European Commission's investigation was and is needed. My position is, simply put, that neither of those companies should be allowed to abuse a monopoly. They should have to give customers choice, and offer their operating systems separately from the hardware on fair, reasonable and non-discriminatory terms. I'm very optimistic we'll get there.
Posted Aug 22, 2010 11:20 UTC (Sun)
by dlang (guest, #313)
[Link] (3 responses)
Posted Aug 22, 2010 11:30 UTC (Sun)
by FlorianMueller (guest, #32048)
[Link] (2 responses)
That's new to me. I understand that Windows can be virtualized with Xen (Windows 7 is listed on this page), VMware etc. Wikipedia says "VMware Inc makes VMware Tools available for Microsoft Windows, Linux, Sun Solaris, FreeBSD, and Novell NetWare guest systems."
Posted Aug 22, 2010 12:53 UTC (Sun)
by foom (subscriber, #14868)
[Link] (1 responses)
Posted Aug 22, 2010 12:57 UTC (Sun)
by FlorianMueller (guest, #32048)
[Link]
Thanks. There are always new things one can learn here on LWN :-) From a competition point of view, I wouldn't be concerned about that. It seems reasonable to me based on how you describe it. Since z/OS is by definition the opposite of a "Home" operating system, this kind of distinction is a non-issue for the mainframe matter then. But it's good to know.
Posted Aug 22, 2010 11:09 UTC (Sun)
by dlang (guest, #313)
[Link] (3 responses)
the case of OS/X is even worse in my opinion because it is sold in a retail box set.
In the IBM case, the software isn't sold separately from the hardware (possibly it should be, but in this case, forcing IBM to sell the software to run on turbohurcules is similar to forcing them to support another hardware platform, it's one that attempts emulate the IBM hardware/hypervisor, but since it's an emulation it's not always going to work the same way)
trying to sell the software to run on turbohercules with a caviot that 'it may not always work and IBM won't try to troubleshoot or fix it' would not be soemthing that would go over well.
I could see IBM signing an agreement with turbohercules for them to sub-license the software and provide all the support for it, at that point it would be very clear that if anything didn't work the customer should not contact IBM.
By the way, there's nothing wrong with being a monopoly in a field (like IBM and mainframes), what would be wrong is if IBM abused it's monopoly position. I don't see that happening yet in this case.
Posted Aug 22, 2010 11:18 UTC (Sun)
by FlorianMueller (guest, #32048)
[Link] (2 responses)
This is a matter of market share, and that also marks a fundamental difference between the antitrust aspect of Psystar (it also has a copyright aspect, but that's irrelevant here because TurboHercules doesn't argue on the basis of "fair use") and TurboHercules. I wouldn't necessarily call it "support". I'd say "tolerate". I can't speak for TurboHercules and whether this is a proposal they should accept, but I would like to mention that TH had proposed to IBM to sit down and discuss, and IBM said the door is closed. That's how it led (eight months after TH's original question) to an antitrust complaint. Let's hope that there's going to be an agreement soon in that context. Right on. I usually always attach the word "abuse" to it to make it clear. I actually do. I look at the European Commission's two Microsoft antitrust cases. The first one, besides the Samba/network protocl aspect, was about the tying of the Media Player to Windows, and the second one about browser choice. So there are precedents in the EU. Look at this way: a decision by Microsoft to stop selling Windows separately and instead requiring anyone who wants to purchase a new version of Windows to buy hardware exclusively from Microsoft would easily result in widespread consensus that it can't be accepted. It's just that more people are familiar with Windows than with the mainframe.
Posted Aug 22, 2010 11:30 UTC (Sun)
by dlang (guest, #313)
[Link] (1 responses)
I think that if turbohercules had proposed something like that before going public, they may have had a chance, but after going public the way they have, for IBM to suggest this would smack of backing down to extortion. IBM tends to be very agressive to responding to attacks.
selling the OS to run on turbohercules without supporting it would not work well for anyone (and would generate bad PR when things didn't work). it's not just selling a license and 'tolorating' turbohercules.
no, this isn't anything like if microsoft decided to stop selling windows separatly and only sell it with microsoft hardware, not only because microsoft doesn't sell hardware (and never has), but also because IBM has never sold the software separately, so it's not that IBM is restricting something that was previously open, it's that they are doing the same thing that everyone in the mainframe market has always done, bundle the OS with the hardware.
Posted Aug 22, 2010 11:52 UTC (Sun)
by FlorianMueller (guest, #32048)
[Link]
The whole detailed discussion on this page here has already addressed all of this before in one way or another, and with your comment you just showed that you know hardly anything about the mainframe situation and about the basics of competition rules, but let me explain anyway: 1. TurboHercules sent a polite letter to IBM in July 2009, asking for it. Four months later, IBM turned down. TurboHercules sent another polite letter to IBM then (in November 2009). The public was only informed in late March of an antitrust complaint and in early April of IBM's patent threats: more than eight months after the original, polite contact in which a face-to-face meeting was suggested. 2. Apart from the above, a monopolist is required to treat potential customers in a fair, reasonable and non-discriminatory way. That means if you don't like the color of someone's skin or the shape of someone's nose, as a monopolist you still have obligations to meet unless you want to commit an abuse of your monopoly. See my explanations above. Nothing was discussed in the press until eight months after the original contact was made. No one forced IBM to make two versions of a patent threat, which then of course concerned people in the community, including (among many others) Eric Raymond and me. They did propose negotiations eight months before going public. All four of the letters exchanged between the two companies from July 2009 to March 2010 are discussed in this blog posting, which includes links to the actual letters for everyone to see. The first step is to allow it. The second is how to address the support question in a way that would be fair, reasonable and non-discriminatory under the law. I didn't say there shouldn't be any support. To me, the support question is a logical consequence of the "tolerating" question, but I admit I didn't make it perfectly clear in my previous comment. The question I raised is what would happen, under competition law, if they decided to leverage their market position in software in a way that distorts competition in hardware. If tying is illegal under competition law, then it's illegal regardless of whether previously practiced or newly started. No. Under the US DoJ's Consent Decree of 1956, IBM did provide all components including its operating systems separately for several decades. There were various "plug compatible mainframe" (PCM) manufacturers in the market such as Amdahl and Hitachi. But again, if tying is abuse of a monopoly, it's illegal regardless of history. If you park on the wrong side of your street for 10 years and then in the 11th year the police tells you that it's illegal, you can't argue that you've been doing it for 10 years so it must be legal... See above: the plug compatible mainframe story. The PCMs ran IBM operating systems. The fact that IBM was required to make its operating system available separately is, by the way, considered to have played a key role in IBM's original arrangement with Microsoft that allowed the sale of MS-DOS to other hardware companies. They might have assumed at IBM that sooner or later they'd be required to do so anyway. Now that I've patiently explained everything, it would be very kind of you if you could reconsider your position based on what you know now.
Posted Nov 19, 2010 16:51 UTC (Fri)
by FlorianMueller (guest, #32048)
[Link]
You said you've got reams of proof that she censors, and I'll raise you reams of proof that she doesn't (you can find it all on the site). I didn't just say so -- you can see it yourself now: http://www.scribd.com/doc/43344245/10-11-19-Groklaw-Censorship-Evidence
So what do the proponents of Oracle's acquisition of Sun think now?
IBM's pledge didn't just say they wouldn't sue. It said they wouldn't assert, which is much broader, as I explained here.
and from the post:
I know that some people say that IBM actually just answered a question, and that the infringement they asserted was only hypothetical.
So what do the proponents of Oracle's acquisition of Sun think now?
Anyway, given that in your opinion IBM answered the question badly, how do you think they should have answered, in a way that is succinct, clear, open, true, and in line with their patent pledge?
So what do the proponents of Oracle's acquisition of Sun think now?
Wol
So what do the proponents of Oracle's acquisition of Sun think now?
TurboHercules haven't been around for a decade to receive complaints!
With both patents and copyright, an owner is entitled to ignore (or even encourage?) certain uses. Certainly IBM didn't object to their customers using Hercules (a HOBBYIST, NON-COMMERCIAL TOY).
Certainly IBM didn't object to their customers using Hercules (a HOBBYIST, NON-COMMERCIAL TOY).
TH are naive, stupid, or malicious. Take your pick.
So what do the proponents of Oracle's acquisition of Sun think now?
Wol
So what do the proponents of Oracle's acquisition of Sun think now?
Rather than compare the "hobbiest toys" linux and Hercules, I'll compare the hobbiests Linus Torvalds and Roger Bowler.
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
Patrick.
So what do the proponents of Oracle's acquisition of Sun think now?
1. IBM, in their initial response didn't assert patent infringment upon the Hercules project in general. They were suggesting that TurboHercules' implementation of the emulation would be infringing
In IBM's licensing agreement (from my understanding [...]), you are allowed to emulate their OS on non-IBM hardware for certain limited uses (disaster recovery, for example). TurboHercules wants to expand that (and change the licensing agreement) to mean that you can run their OS on non-IBM hardware for any reason at all.
it's like people saying that it's Microsoft's fault that their video card broke. Or more accurately (since this is an Open Source forum), it's like people bitching that it's Ubuntu's fault that their ATI video cards don't work properly (when it's AMD that won't provide the proper drivers for the card).
I don't think that Microsoft is necessarily behind every bad thing in the world (but I do think that as far as tech goes, they aren't upset when the bad things happen to their competitors-- just like any company wouldn't be upset).
She puts up the facts. She links to them, so you can read them for yourself. You are free to form your own conclusions, whether they agree with hers or not. But, she's also free to post her opinions about the facts.
(Groklie quote) IBM, when it announced the patent pledge, specifically reserved the right to defend itself from attack:
"IBM has no intention of asserting its patent portfolio against the Linux kernel, unless of course we are forced to defend ourselves," said Nick Donofrio, senior vice president for technology and manufacturing, drawing applause in a speech at the LinuxWorld Conference and Expo.
And in the TurboHercules story, who is suing whom? It's not IBM, folks. The complaint against IBM was filed with the EU Commission by TurboHercules. At that exact moment, did they not take themselves out from under the patent pledge's safety umbrella? So be it. (end Groklie quote)
I had explained the whole problem, but if you look again at what PJ wrote, she said: "IBM, when it announced the patent pledge,[...]" and then presents a quote from August 2004 while the pledge was made in January 2005. So just the little part I quoted, and clicking through to the ZDNet link, shows she lied. But with little click-through rates and a number of people who trust the wrong person (a "person" whose identity isn't disclosed to even a remotely appropriate extent, as I explained in this comment here).
Just to add more clarity about the disgracefully untruthful quote
So what do the proponents of Oracle's acquisition of Sun think now?
Patrick.
So what do the proponents of Oracle's acquisition of Sun think now?
IBM didn't have an issue with the Hercules Project, because it was mainly (or completely at the time) being used in accordance with the Licensing Agreement (or if it wasn't being used in accordance with the agreement, no one was saying that). Plus the fact that it was free (just my opinion).
TurboHercules wants to use it in a way that is not in accordance with their licensing agreement, and make people pay for that right.
IMHO, isn't that against the code of Open Source software?
However, you have to pay those guys to be able to use the software for anything other than the terms of your License Agreement.
"Here's Ubuntu. You are only allowed to use the non-restricted sources. However, if you want access to the restricted sources, you have to pay LinuxCo2, because they are providing a commercial version of Ubuntu that is allowed to use restricted sources."
In looking at the letters between TurboHercules and IBM, I noticed that some of the patents were covering Virtual Machines. So, I wonder if those would be capable of being used as Prior Art in the Google v Oracle case
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
I believe the "Home" (cheapest) editions of windows have a license which prohibits the use of them in any virtualized environment. They want you to buy the next tier up if you're doing something complex like running in a VM instead of bare hardware.
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
the case of OS/X is even worse in my opinion because it is sold in a retail box set.
forcing IBM to sell the software to run on turbohurcules is similar to forcing them to support another hardware platform
I could see IBM signing an agreement with turbohercules for them to sub-license the software and provide all the support for it, at that point it would be very clear that if anything didn't work the customer should not contact IBM.
By the way, there's nothing wrong with being a monopoly in a field (like IBM and mainframes), what would be wrong is if IBM abused it's monopoly position.
I don't see that happening yet in this case.
So what do the proponents of Oracle's acquisition of Sun think now?
So what do the proponents of Oracle's acquisition of Sun think now?
why in the world would IBM be willing to enter discussions with a company behaving the way turbohercules is behaving?
Among other things, they could find any part of their negotiations on the newspaper front page taken out of context.
I think that if turbohercules had proposed something like that before going public, they may have had a chance
selling the OS to run on turbohercules without supporting it would not work well for anyone (and would generate bad PR when things didn't work). it's not just selling a license and 'tolorating' turbohercules.
his isn't anything like if microsoft decided to stop selling windows separatly and only sell it with microsoft hardware, not only because microsoft doesn't sell hardware (and never has)
but also because IBM has never sold the software separately,
that they are doing the same thing that everyone in the mainframe market has always done, bundle the OS with the hardware.
So what do the proponents of Oracle's acquisition of Sun think now?