Discussion on:

The Novell deal is done: Let the patent scandals begin

Microsoft will only care if chrome becomes popular. Otherwise they will do nothing. It has been 15 years since Linux and Microsoft's NT platform have jointly ruled the X86 server business. Also Linux has failed to be a threat to Windows desktop OS. Businesses probably won't switch to Linux for the desktop in mass with only options like Ubuntu availble. Also Linux and Windows server admins will not want to voluntarily switch in mass either. I believe Microsoft would know that long term they would lose out if they made people abandon Linux by gun point. Linux is working well in the server room is a fact I think they would privately admit. So the real concern is google if google continues their plans for a complete monopoly of user IT experience (which includes the desktop via Chrome OS) then Microsoft would be forced to try to play its hand of patent infringement. I think those who go the open source route for the right reasons would be against google's chrome anyway. Well now you have another reason. -

They may also have interest in Android which is already presented a popularity threat. It's much closer to a Linux distribution than ChromeOS apears to be.

Maybe Microsoft purchased the patents to Novell technology that will improve its products. Windows is still inferior to NetWare as a file and print server even though the former has had no new development for seven years.

I mean, really! This has been debunked repeatedly, most notably by Andy Tannenbaum, who wrote Minix. See the wikipedia article. While there's a lot of mutual (if grudging) respect between Torvalds and Tannenbaum, they take very different approaches to OS design.
This is an indication of the difference between journalism and blogging.

The paragraph was constructed like a stream of consciousness monologue, nevertheless I did not see Jack suggest "derived," a code evolution, rather that a few folks had some similar missions and ideas. One of them wasn't handed top shelf equipment on a silver platter and paid for his "time." I think that guy is within right to have his criticism of 'establishment' be heard.

As noted, Torvalds wrote it from scratch to run on x86, a UNIX/MINIX - like OS.

Yes, but Linux is designed from the ground up as a monolithic kernel OS, whereas Minix is designed around a microkernel -- two completely different architectures, and the source of the great Torvalds/Tenenbaum debate.

The whole point is that they are different, and none is derived from another.

Hence, I don't understand the "but".

Please moderate the anti-MS basement conspiracy theorist in yourself before writing.

As much as I enjoy your articles, sometimes you seem to barely contain the anti-MS Mr. Hyde within you. Try letting your Dr Jekyll side proofread and edit any articles before Mr Hyde can post them next time.

Should be ignored should it?

Microsoft are hardly likely to change their ways just because you may like them to. They have a long history in starting Anticompetitive Practices that started with the company who really was responsible for getting them going IBM, and I very much fail to see where in the above Blog anything different is going to happen nore should it be expected.

Col

But MS has already been burned more than once but to accuse them of waiting for a chance to swoop in and dissect the Linux kernel after the death of Open/SUSE is kinda paranoid when the kernel and its source code is out there for anyone to analyze and dissect to their hearts content. And I doubt even MS could do anything with the patents they acquire other than use them to improve their own products.

Edit: Justin James stated the content of my thoughts below far better than I could.

They made Kea!. Our terminal program back in 2000 for Win 95, 98, ME and 2000. Still using it too. Didn't know they still existed.

We were using their terminal emulation software back with Windows 3.1 and WFW 3.11, but I was unaware that they were even still in business. Their purchase of Novell seems real strange to me since it seems way out of their product line. Makes me think they are a front from some other investment group or such.

Take another look at their product line. It is a GREAT fit with Novell, when you consider that Novell employs Miguel de Icaza, who runs GNOME and Mono. Attachment has stuff that lines up perfectly with that kind of tech portfolio.

J.Ja

although I agree with you about Mono.

It's a disease, Often caught by teenagers after playing spin-the-bottle. Here's the cure:

>rpm -e mono

But he has/had a LOT of involvement in it. His experience is invaluable to any company doing X11 work like Attachmate does.

J.Ja

I forgot bout them, and didn't know the company was still in business. That's good news. A company does not have to be in the news to survive. Buying out the competition has been going on since business began. It's had to keep up with acquisitions...for some reason I thought Sun bought Novell. Do we need to care as technicians?

EXTRA! X-treme. Yes, that is what it is called, caps and all. Can emulate terminals for VAX VMS, UNIX, AS/400, and IBM mainframes.

"What is Attachmate? I had never heard of them. What do they do? According to their ?product listings? on their web site they do Terminal Emulation and Enterprise Fraud Management."

I've heard of them. Not much, though. So I took a quick trip to their site, and I can see why the Novell property might be attractive to them. I found that out in under 1 minute of looking at their site.

Regarding "fraud management", here's what their site (again, discoverable in ONE CLICK from the front page) says:

"Luminet enterprise fraud management software sees, records, and analyzes user activity across all applications. In this way, it helps you monitor, detect, and prevent a range of abuses, including insider fraud and regulatory violations. By removing the guesswork from application monitoring, it gives you the intelligence you need to take informed action."

Not exactly a Microsoft conspiracy theory here. The Novell property makes sense. Oh, and they also do stuff with terminal emulation (as you pointed out) which is more than just telnet, and X 11... Novell... Miguel de Icaza... GNOME... ring any bells? See the connection yet?

Also from their products page:

"With the Verastream integration suite, you can expose business processes as reusable web services, XML, JavaBeans, or .NET components. Verastream-generated services can be mixed, matched, and selectively combined to extend legacy functionality to new users or composite applications. No code changes to legacy applications are required."

Hmmm... Novell... Miguel de Icaza... Mono project... get the picture? Again, this is a good fit, not a conspiracy.

"Oh sure, Novell has already published a statement saying that they will retain all of their UNIX patents after the purchase (Read the full statement here). But that leaves behind some 822 other patents owned by Novell that will fall into the hands of Microsoft. I would venture a guess that many of these patents are Linux-based patents that Microsoft (or rather Steve Ballmer) hopes will arm them mightily in their never-ending battle to take down Linux as a whole."

If Microsoft wanted the patents, they could buy the patents. They don't need to help broker a deal with Attachmate for that. Furthermore, how does Attachemate's purchase put these patents in the hands of Microsoft? Finally, Microsoft gets well over 3,000 patents a YEAR. 822 isn't a big deal to them.

"What strikes me as rather odd about this deal is that, in the past, Microsoft has made the claims that it owned the rights to some of the intellectual property in the Linux kernel. If that?s the case, then why are they allowing Novell to retain their UNIX patents? Is this what will become an inside job where MS is attempting to slowly put a strangle-hold on Novell so they can wrestle those UNIX patents out of Novell?s cold, dying hands?"

Learn the difference between "intellectual property" and "patents". There is a VERY important difference between the two! Also, learn the difference between "UNIX" and "Linux", and the difference between "UNIX" the operating system and "Linux" the kernel.

To put it another way...

Microsoft may have intellectual property claims (perhaps patents, perhaps copyrights) on some things within the *Linux* kernel. That does not mean that the *UNIX* patents would have anything to do with those claims on the Linux kernel. The UNIX claims could be, say, things related to the network stack, or perhaps the disk partitioning system...

"Here?s where I get confused: Linux was based on Minix. Minix was an operating system (Mini UNIX) created by Andrew Tanenbaum designed for education and science. Minix is a UNIX-like operating system, but was built from scratch. There was never any IP infringement on the part of Tanenbaum. When Torvalds decided to create his own operating system, he was just trying to create a ?Minix? that would run on x86 hardware (He couldn?t afford the far more expensive hardware required to run Minix). And so Linus Torvalds created Linux. Eventually the GNU applications were added to replace the Minix application. The GNU applications were created by Richard Stallman with the goal of creating free, UNIX-like software from the ground up."

This is the difference between a "patent" and a "copyright". A "patent" doesn't care about the content, it cares about the methodology. It is wholly possible that despite being a ground-up, from-scratch codebase, and despite cloning the Minix kernel's external interfaces (the only part that matters), the Linux kernel could indeed use techniques patented by Microsoft or UNIX, even though Minix does not.

As an analogy:

I'm writing a media handling library. I like one of the GPL'ed media libraries a lot, and I want mine to be a drop in replacement. So I look at the library and find all of the externally callable code interfaces. I give my library the exact same interfaces, so someone can refer to my library instead of the other one, recompile their app, and it works 100% out of the box. So far so good, nothing illegal there. Now, let's say that on the inside of my library, I write the code from the ground up. Again, no problems! But let's say I do some algorithmic tricks that happen to be patented by the MPEG group. Oops. Even though I have a ground up rewrite, I am in the crosshairs of a patent lawsuit, despite the fact that the system I cloned was 100% in the clear.

"Here?s what I think. (I know you?ve been holding your breath, waiting for my take on it?) I believe that Microsoft is, in fact, going to make a grab for those UNIX patents. I believe those, and only those, patents are their one true target in this acquisition. They are going to let SUSE and openSUSE continue on and, when those two products fail to deliver (they assume and/or hope), they will then dive in and claim their prize. Once MS has their hands on those UNIX patents they will comb through the Linux kernel and finally, and absolutely, be able to either lay claim to patent infringement or give up this ghost hunt all together."

Again, how does the Attachemate purchase give these patents to Microsoft?

When Microsoft wants to do something... they do it. They have the money and the manpower. The fact that *not a single thing they produce* (other than .NET) actually originated in-house and was originally an acquisition is proof positive that they are well versed in buying what they need. They don't need to play games to get a hold of patents. They have the money to buy them and the relationship with Novell to do it if they want. Novell is dying, quickly, it's not like they need to retain their rep and hide a patent sale to Microsoft, and it certainly would incur less wrath from DOJ to buy them outright than to use Attachemate as a cutout for patent trolling.

J.Ja

And as you have taken the effort to point out, there does seem to be a good fit in several respects.

As to the MS & patents bit, that was a part of the deal. MS, through a consortium (CPTN Holdings), acquired 882 "other" patents - probably a bunch of Novell stuff, Word Perfect stuff, etc. What some people seem to be worried about is if there is some Suse stuff involved. (Since Suse is supposed to continue normal operations, I don't know why the patents would shift ownership, but there is all that cross-licensing that was up for renewal soon.)

Your illustration of software patents issues is excellent - I don't think enough people understand that about software patents. Almost all it takes is a very generic statement for "A method of...", and you have a patent.

I personally find this to be a problem with patents. They are frequently overly broad, and if two or three entities come up with a similar "method" for doing something, I think this rather fails the "non-obvious" test. And there is quite probably prior art for a lot of software patents. But this a whole other ball of wax.

Yeah, I knew about the patents and the holding group. But here's the questions it raises:

1) Why is Microsoft supporting a consortium for the patents instead of buying them outright? The responder below (Rick S._z) suggests it's to shelter their business from counterclaims. That's an interesting and potentially plausible explanation. But the history is wrong. Microsoft provided SCO with financial backing and encouragement, but SCO had those patents the whole time. It isn't like Microsoft gave SCO money to buy the patents and then told them, "go get 'em tiger!"

2) I STILL fail to see the connection between Attachmate and these patents. Jack's never explained this, and neither have any of the commenters here. What's the connection? It would seem to me that Jack's proper complaint should be able the sale of the patents to the consortium.

3) What do we know about this consortium? I don't know much about it, other than the name and that Microsoft is involved. Who are the other players? What percentage of control do they have? A good example of a well-managed patent consortium is the MPEG group. Regardless of how you feel about them having control of that patent pool and enforcing them quite strictly, the fact is, they are *fair*. No one company dominates them. As a result, no one who wants to license their patents is ever locked out, and they don't bring lawsuits for political reasons. There are just too many cooks in the kitchen for that to happen. If this CPTN consortium is like that... well, it's as good as it's going to get. If Microsoft has 75% of the votes on it and the other 25% are spread between a couple of Microsoft partners... well, that's an obvious bad sign.

4) Is 822 patents really enough to sink Linux? Considering all of the things a kernel (let alone an OS) does, it is inconceivable that a modern kernel can ever be 100% unburdened from patents. I can't imagine that these 822 patents would be the critical key to demolishing, or even wounding Linux.

5) Patents have a shelf life. Any patent in that group, if it is really damaging to Linux, is either close to expired, or was done AFTER Linux was made. So at worst, they cover techniques which *may* be used by Linux.

6) If the patents were registered by Novell, and cover things that Novell did in SuSE, guess what? Anything that was released to anyone should have also come with a patent clause to cover its use too. Let's pretend that one of the patents is for thread scheduling. If that thread scheduling technique ended up in the Linux kernel, then at worst, that thread scheduling need to be removed. At best, the code came with an unlimited license for the patented idea too.

7) There is ALWAYS an implied "buyer beware" with code contributions, especially in open source projects. No project the size of Linux should be accepting code without checking out the patent situation. Look at HTML5 and video codecs. The *entire* hold up there is because the HTML5 group does not want to specify a codec that could be covered by patents. Stuff like WPM/V8 doesn't fix the problem, because as I've already shown, ground up rewrites don't clear you from patent claims. There is 100% no way that *anyone* has found to make a video codec not covered by existing patents in one form or another. Same thing for kernels. So if code ended up in the Linux codebase that was under patents, but the project didn't get the license for the patent, who screwed up there? And who's the bad guy? Not Microsoft, it isn't like they slipped this Trojan horse code in, just to pull the rug out later...

8) If these patents were so important? Why didn't IBM or Oracle, both of whom have the money and familiarity with patent law to do a deal, chase after them?

At the end of the day, the conspiracy theory explanation looks weak. Until Jack (or one of the other commenters) can come up with an even semi-plausible explanation for at least some of these in-line with the "conspiracy theory", I'll say that the idea of a conspiracy holds no water. The whole thing makes no sense.

J.Ja

Recall that during the 90's, they held off a Justice Department suit for nearly a decade until a friendly administration came in and killed it.

They may have nearly infinite resources, but I doubt even MS are anxious to be directly and obviously involved in anything that might raise that specter again.

Yeah, like I've said, I can see in the abstract why they would want a "sock puppet" (or "cutout", "cardboard man", or whatever we want to call it) for patent litigation. That, in and of itself, makes sense.

But no one in this discuss or the original post have really shown if Microsoft actually has a sock puppet here (again, who is in the patent consortium, and how much control does Microsoft have over it?), and no one has explained in ANY form how the Attachmate deal relates to this *at all*.

J.Ja

...was it before we knew for sure SCO was little more than a front for MS? Not all inferences are unreasonable, even if they may not pan out at the end of it all.

can the idea not be mentioned without an already overblown conspiracy theory? There isn't even any point in discussing it until someone comes up with some facts.

And no, I am not buying in to conspiracy theories at the moment, either.

I don't know what Attachmate links there are to MS, if any. The only thing I see is that they split the items for sale, the consortium buying stuff that Attachmate likely does not want and does not want to spend money on. A conspiracy, at this point, in my mind, would work the other way around - MS fostered Attachmate's purchase/merger by buying the garage sale items Attachmate wouldn't have bought, but needed to be sold by Novell for there to be a sale at all. Let's go halfsies.

The patents are explicitly not the UNIX patents, I doubt that tey would be Suse-realted patents as Suse will still be a business unit, so that mostly leaves Novell stuff that Attachmate really did not want. (Maybe including the Word Perfect Office patents Novell still owned.)

Still, "big deal", so far. I think Oracle is still a much better punching bag for those with the need. I just can't see any "Linux-sinking" in the works, here. And, frankly, I think MS would have chosen Nathan Myrvhold
to buy the patents if they wanted a patent-trolling dog. (Who knows, maybe Intellectual Ventures is in the consortium.)

Until we know anything more about this, I see no reason to freak out. And I trust MS about as far as I can comfortably spit the Redmond campus. I just don't see teh kinspearisee.

neither I nor, as far as I can see, Jack, are getting hysterical about this. We're simply recognizing and pointing out past behaviors and watching current activities to see where they go.

The silliest thing we can do is pretend there's no possible problem until we're buried under it.

Going into a panic based on present information is only the second-silliest thing.

Go read his post again. It's a chicken little, sky is falling rant, with ZERO backing evidence. And note his utter silence and failure to respond to those who have criticized him on this.

J.Ja

And, what do you think you are going to do about it, anyway, before you are buried under it?

Again, I have a special place in my heart for MS, and it isn't the nice place. I've had problems with MS long before I was aware of any modern alternatives. Before I was aware of an open source ecosystem.

But what is all the theorizing - nay, this cannot even qualify as hypothesizing - supposed to do? Provide a plant for a possible future "I told you so"? You have absolutely nothing to go on in this case.

Go pick on Oracle.

Microsoft *DOES* need a sock-puppet to execute the lawsuits, because they would risk great damage to their current assets (and future profits) if they were attacked with successful counterclaims and compensatory damages for inflicting misery.

Caldera/SCO was a perfect puppet: Now that it's almost completely over, there's no money left for IBM, Red Hat, or anyone else to go after. If the plaintiff had been Microsoft, instead of it's sock-puppet, counterclaims would probably call for damage amounts in the $Billions.

... or what do you think?

"Microsoft will find nothing useful..." is almost certainly wrong, because Microsoft wins by merely "sucking out the air supply" of Linux Developers, Users, and Supporters. Scaldera presented no specific claims, but still managed to create 8 years of FUD, Court Costs, and distraction from productive work. (And they're STILL not dead.)

"Microsoft has everything they need..." is one of the incorrect answers. ("Who cares" is the other.) When they do attempt to claim "ownership" of something which is currently in Linux, they might find their claim demolished - in one of several ways. (The original creator might have licensed it under the GPL, and "exclusive Novell Ownership" is gone- it's already licensed in perpetuity. Or worse, the findings of fact determine that Novell or MS has infringed a GPL-licensing owner. Or the "infringing code" can be removed and replaced.

Now for the REALLY COMPLICATED one (which I'll attempt to describe correctly, eventouh IANAL): The particular Unix "property" being claimed might be found to have been put under license terms which approach the characteristics of "public domain" writings and ideas- with only minimal requirements to recite certain Copyright notices and Acknowledgments. This was a result of the result of the AT&T vs. California Board of Reagents case. (AT&T had abused Copyright requirements so badly, and so much code had been created by authors elsewhere, that large amounts of it (Code, Methods, and Concepts) were hopelessly intermixed, and "handed out" in classrooms, and revised by numerous Developers, and handed back to USL and the University, and to other Universities, and many other entities (some with license agreements, many without.)

Files were divided into 4 groups, with different terms of usage and licensing each. But overall, a lot of the "Eye-Pee" ended up under the very generous licensing terms I described above: Keep these copyright acknowledgement recitals in place, and you can do whatever you like with the code, methods AND concepts inpmenented in the file, and the license terms "pass through" to anyone you give it to. You and your antecedants will never be sued for using this stuff.

And so, they might own it- but the ownership carries no rights to sue anybody for anything, unless they failed to keep the proper recitations attached.
-----

However, actually arguing about that stuff could wrack up $millions in Court costs. To the "Evil Empire" (don't argue with the description, they HAVE been convicted) that's mere pocket change. But to FSF and the Supporters/Developers of Linux, it's a big problem.

"Who Knows" is 100% correct. Nothing is claimed, it's all FUD so far. But it involves a big stack of genuine copyrights, there's a lot more to fight about.

Linux will almost certainly rise from this with clean hands, it's simply NOT Unix. It emulates UNIX behavior in many ways, based on public, open standards of compatibility. But per above, probably not "unscathed". MS has too much money, too much greed, and too much Evil to argue or fight fairly.

It doesn't matter who is behind! As we all know the Linux/Other OS community is somewhere a cash cow. MS is following (in slow motion) the same path. The "R & D? is done by the community and the big players, i.e. MS, Novell, Redhat, ?, just derive something out of it and patent it. So as long something or someone is a cash cow, it?ll never be killed.

Your inmplication that when Steve Ballmer goes, Microsoft will See the Light appear to me optimistic in the extreme. The history of large corporations is that, when the charismatic founders who cared about the actual product go, they are replaced by bean-counters and lawyers who are MORE likely to protect corporate assets, not having the ablity to innovate themselves. (Assuming that you agree that Microsoft have been innovators in the past anyway!)

The idea of "taking down Linux" is preposterous. Taking down Linux implies that there is some financial loss at stake (note I am saying Linux, not RedHat or Suse Enterprise). Linux, as an open source community owned operating system, has NOTHING TO LOSE. Even if by some miracle Microsoft managed to gain market share to the point where they overtake Linux in whatever key area they are trying to overtake them in, they will have deprived the owner (the community) of nothing. The community will still be in the possession of the most stable server operating system in the world.

This article is a perfect example of why TechRepublic HAS to add a "thumb-down" button together with its "Worthwhile" button.

As a Novell admin for longer than I care to admit I found the article interesting. I'm just glad I sold what little Novell stock I had last year. It's a shame the people at Novell always seemed to do such a lousy job of marketing their products. I still prefer my old Netware server over our Windows 2003 server.

Sounds like you don't understand the difference between a patent and a copyright.

Hey, Jack? If at the end of all the legal wrangling it turns out that Microsoft (Ballmer) is correct in their claims, will you be man enough to extend an apology for your paranoic lamentations? Or, will you continue your quest to slay the dragon by choking it with your sour grapes?

sorry to point this out but the discussion about Minix being build from scratch is completely misguiding. If true, it will only prove there is no copyright infringement. Patents are different: if you invent something from scratch but *after* someone else, then you owe royalties if you use it for business; you can do research about it, and Minix was a research and teaching OS.

Long before they got into Fraud Management, Attachmate was a highly respected company that build software terminal emulators to attach (see where the name comes from) to mainframes, IBM in particular, from PCs. They were expert in various communication protocols, of which Novell was one of the popular ones at the time. So this makes a bit of sense without invoking conspiracy. Where does the puppet-master assertion originate?

You talk about Linus wanting to port Minix to the x86. That makes no sense; Minix even worked on an 8088, out of the "box". But it was a toy OS, I think he was after something more solid, perhaps also taking advantage of the 80386 capabilities.

i have to say this topic really brings about some outstanding discussion.

the most challenging issue with this topic is that it is so heavily couched in legalese. we can all make assumptions and claims, but ultimately this whole mess will fall into the hands of a court and the speculation will finally end.

during the process of writing this particular article i read article after article describing what has, should, can, can not, might, might not happen. it seemed that on an hourly basis everything changed...but the one thing that has remained constant is that the issue has had the Linux community holding their breath, panicking, or anxiously awaiting another failed attempt at proving IP infringement.

i, for one, fall in the latter category. what about you?