Slashdot Mirror
Red Hat Files for Software Patents
Marsala writes "Apparently Red Hat has filed two patent applications for stuff related to the TUX webserver. The patents are for Embedded Protocol Objects and Method and apparatus for atomic file look-up. One has to wonder (if their patents are granted) what their licensing terms will be.... free for open source, or a tool to try and screw other Linux distros?" As reported by Linux Weekly News.
You have to think about this...are they filing for the money or will they open it? If they believe in linux as a principle of their company they won't prevent anyone else from using it, but if they simply want to make shareholders happy they might charge for it.
-CPM
---You're all I need, When the water runs deep, You're all I need, Now I cry my soul to sleep -- Collective Soul, Needs
Is it not agreed in OSS community that software patents are bad?? The argument that their protecting the community is garbage. The money effort should be put into fighting software patents.
Is it just me or do both patents end with the patch for the atomic file lookup?
--
Benjamin Coates
They don't seem to define quite what it means, but as near as I can tell, it's pretty much the same thing as the atomup updates that database systems have been doing for a few decades.
What is there in Red Hat's patent application that is actually new?
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
I just had an unecessarily alarmist thought. Could this be used to defeat the GNU license? Sure, you still release your source code under the GNU license, but also charge everyone for using 'your' patents.
The first, "Embedded protocol objects", seems to be saying that if you have a webpage that consists of dynamic and static content then the static content can be cached for faster access. Hardly novel.
The first, "Method and apparatus for atomic file look-up", basically says that it is a good idea if you can see if something is in a cache before requesting the operation that would put it in the cache. Again, not particularly revolutionary.
Red Hat is the Microsoft of Linux... Ironically, this makes it the best weapon Linux has against Microsoft.
Remember "Bring 'em on"? *sigh
...they are looking towards the future.
If the US patent office allows you to patent knives and forks, RH is assuming that they will have to patent the spork to guarantee their future rights to use it without paying royalties.
This just shows how sick the patent process is at the moment. I only wonder why someone hasn't patented callable functions--prior art hasn't seemed to be much of an obstacle in the past.
G
There's something I do not understand about patents, like, how on earth can this be patented? Specifically the first patent:
"Dynamic and static protocol objects are mixed together at a server and included in a dynamic reply to a communication request made by a client application."
Ahhhhm, so a client requests something and a reply is made that is partially dynamic and partially static? So, like php then? Like perl cgi pages? Like any reporting engine ever written?
And, check this out, step by step explainations of how the code works. This shit only exists to keep lawyers happy.
Don't get it.
Dave
I write a blog now, you should be afraid.
But RedHat is a publicly-traded company now, so it has the same "duty to the shareholders" that every other large corporation in the U.S. has. Hence, I have a lot less faith in the company regarding the use of stuff like software patents only for Good. In short, I won't be surprised at all if they use these patents to smack down other commercial Linux distributions, all for Profit at Any Cost.
Only if a Good Guy retains a controlling share of the company would that not apply.
Use 'slashdot stuff' in the subject line in any email you send me if you want to get past the spam filter.
I don't even like redhat linux. At all.
But, as much as I dislike their product, I just get the strange feeling inside, that their company isn't run by the complete and utter assholes we see everywhere else.
To suspect them of pulling any dirty is just damn wrong. Maybe it's just me wanting at least one company out there to be ethical, some really corny wishful thinking on my part, but what have they ever done to you? They deserve an an apology.
Besides, they might be able to stick it to M$ somehow...
I recall a few year back, this was discussed amongst GNU folks, and others as a mechanism to prevent proprietary vendors from locking 'us' (us being Open Source developers, users and businesses) out of important inventions.
I have only seen one 'iffy' thing ever come out of Red Hat (the RHN server, which is an in-house secret...though they appear to be helpful with the Current developers), so I tend to believe they have no intention of using this offensively against Open Source companies or users.
Besides, I seem to recall the GPL protects us against anyone integrating incompatibly licensed code into GPL software. TUX is part of the Linux kernel, thus it cannot be restricted or fees enforced on its use. Red Hat would have a bear of a time rewriting TUX independently of the kernel (Ingo could do it, of course), but the damage is done. Those inventions (and inventions they are...Ingo does really cool stuff, this isn't a one-click patent folks) are already in the development kernels. We (the community) already own these developments.
So, what I'm trying to say is:
Thank you very much, Ingo and Red Hat. I am very appreciative of the interesting inventions you have given to us. I forward to learning about them, using them, and enjoying the benefits they give us. And boy, that sure was clever of you to patent it, so that Microsoft will have a bit of trouble stepping on your toes in the future. Way to play hardball!
The GPL requires anyone holding a patent on the software to allow others to freely use/modify it.
From the GPL license:
Finally, any free program is threatened constantly by software
patents. We wish to avoid the danger that redistributors of a free
program will individually obtain patent licenses, in effect making the
program proprietary. To prevent this, we have made it clear that any
patent must be licensed for everyone's free use or not licensed at all.
The only thing this patent prevents is from others creating proprietary versions of the technology in question; which, IMO, is a Good Thing(tm). In fact, in the thread about this on the LKML someone brought up that the FSF even encourages doing this.
What's the big deal? The GPL is pretty clear on this: So, Redhat can't use these patents in an abusive manner unless they're going to release this stuff in some kind of software that's released under a GPL-incompatible license.
Last i checked, Redhat had a policy of releasing software they write under Free Software licenses. So unless i'm badly mistaken, or Redhat is moving away from Free Software, which i haven't heard they were doing, then this is nothing at all to worry about.
Now, if redhat were going to announce they were going to start writing software and releasing it under non-free licenses, i would be worried indeed.
But for now.. Haven't you people ever heard of a defensive patent? Maybe a bit unecessary, maybe a bit worrying, but this isn't the end of the world.
- super ugly ultraman anonymous cowards filtered. if their comments aren't worth so much as a nom de plume why should i read them?
So Redhat has done some significant work on the TUX webserver and it's patentable. They should DEFINTELY file those patents.
Now don't get me wrong. Software patents are generally bad news - they are too long (even 10 years in software is an eternity) and a lot aren't even worth the paper they are written on. But for the software industry today they are part and parcel of developing software and protecting the work that a software company has put in.
Patents aren't like trademarks - there is no 'patent dilution' to worry about. So RedHat can be extremely selective about who is allowed to benefit from the patent (even free of charge) and who is never going to get a license. This is particularly important - MS has a vested interest in watching over a lot of the developments in the TUX webserver as TUX is the main performance competitor to IIS. If RedHat patents its best ideas, it makes a public record of the 'invention'. If MS patents something that RedHat came up with, yes, prior art may be used to fight a legal battle to have the patent nulled but that is considerably more expensive than simply patenting the good stuff first.
Now I look forward to the day when software patents last 3 years from filing (because I dont see software patents ever going away totally). I look forward to the day when all software patents are published PRIOR to the patent being awarded so that many eyes may suggest prior art challenges. But until that day comes, Free and Open source companies had better use the weapons at hand to protect their assets. And if that means filing patents so be it.
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
Patent it before someone else does, or lose it.
Get a free ipod.
Where have you been, man? This stuff has been opened up. It is in development kernels already. Heck it might even be in the shipping Red Hat kernel (TUX was offered in my most recent Red Hat installation, and these patents regard nifty stuff Ingo wrote for TUX).
These things are a part of the kernel. Red Hat can't help but open it up.
Gotta hand it to RedHat, they may have finally found a way to make money. In a world where most every commercial distro is loosing money (or barely floating along) by playing ball the old way, RedHat has turned to a new game it would seem. Here we basically have far and away the largest commercial linux distro. They have appeared to many to be the hero of the commercial linux world. Now they are showing they're not in this business to be anyone's hero (unlike Loki). They are in it to make money.
It reminds me of how MS came along in the very early days with dirt cheap prices and was basically the savior of the microcomputer world. before then nothing existed beyond Unix. Unix was EXPENSIVE! DOS was comparatively cheap. Microsoft seemed like everyone's buddy back then, before they started to charge exorbitant fees and put an iron-tight contract in their software in the form of a EULA.
Since RedHat is in this world to turn a profit (and make no mistake about it, offering your products for free isn't very profitable), they are looking at an alternative way to produce. Will they be the next restrictive MS, or is this simply a neccessary "evil"?
Slackware forever. Honestly, what else would you trust when it absolutely positively has to be stable, secure, and easy
If this is a counter action to be used against MS anti-GPL license then it really should be held by a well defined pro GPL party, such as the FSF.
Given that they appear to be the way of the future (even in Europe, the organised eurolinux campaign seems to be struggling to hold off the "overly enthusiastic" patent office), the Free Software community should definitely use them when they discover nifty algorithms.
As someone above pointed out, the GPL guarantees that any GPL-dependent organisation (such as RedHat) will have to license software patents on a royalty-free basis for GPL software. It would be better to license them for all DFSG-free code. Note that such a license would not apply if, for example, Microsoft took some *BSD code an incorporated it in Win2k, because it's no longer DFSG-free.
The only question then is whether you charge royalties to proprietary software firms for use of the patented techniques, or whether you exclude them completely...
Fixing copyright
I think Redhat should Gilmore Patent these ideas. The Gilmore Patent was proposed by John Gilmore ( I saw him present this idea at a Foresight Conference ) Basic concept is that a Gilmore Patent is like the GPL of the patent world, once you Gilmore Patent something only companies who have Gilmore Patents (or no patents at all) may use the patent royalty-free.
The only thing that concerns me about this idea is that it seems like it might be easily circumventable, you could do something like set up a subsidiary that doesn't have any patents, and then funnel the money back to the parent company. Any IP lawyers out there have a feasibility assessment of this idea?
Websurfing done right! StumbleUpon
Could somebody please explain to me why software should be excluded from patenting? If it's a legitimate innovation (i.e. not Microsoft "innovation") why shouldn't it be patentable?
It is a separate matter that the vast majority of the software patents are absolute garbage (like the (in)famous one-click shopping), but so are the patents in other areas (like Rambus, and that guy who patented sideways swinging...)
___
If you think big enough, you'll never have to do it.
Or maybe they are just proving that they are a company attempting to make money. Maybe they are not as open about things as Debian (I use that because that was in your post), but Debian is not a company ... you see what I am getting at ...
man
No manual entry for
It's not a matter of just the current program or making linux more or less useful, it's a matter of setting a precident. What people seem to forget about big things like this is that they're all spawning points for new paths and ideas. If you allow a patent for software that at one point was freely distributable and useable for whatever purpose you see fit. Able to be modified and redistributed as long as the source remained intact, you open a whole can of worms.
Can we limit the use of this software to paying customers (despite the inclusion of source), enforcable through such things as the DMCA.
Can we refuse to let people reverse engineer and redistribute the software?
Can we limit the source to only paying customers?
It's not a matter of closing software and making proprietary software to earn a buck (see Apple's Aqua for that). That tactic is fine with me, release something for an Open Source OS that's proprietary, as long as it's released under that licence. But what happens when you start patenting GPL software? That's what makes this interesting.
T Money
World Domination with a plastic spoon since 1984
The patents will never be granted because they don't represent original work. We can prove this by the fact that open source development consists entirely of chasing tail lights, and never involves doing anything original. We know that's true because Microsoft says so. Q.E.D.
``Life results from the non-random survival of randomly varying replicators.'' -- Richard Dawkins
The difference being, Intel, IBM, Sony, Philips, Microsoft, Apple, EA, Blizzard et al don't release their software and hardware as open source / open spec.
T Money
World Domination with a plastic spoon since 1984
No. Trademarks can be abandoned (i.e., revert to the public domain) if they are not defended against dilution. Patents don't work that way. They can lie in wait for the technology to be more widely used, even if by independent invention. The most public instance of such a "submarine patent" is the GIF patent. Slashdot has also reported on BT's hyperlink patent. Ditto Rambus's SDRAM patents.
The only obstacles to late enforcement of a patent are public opinion and laws against unfair business practices. The latter is mentioned in the Rambus link above.
If you're opposed to (software) patents, I hope you won't limit free licensing to GPLed software. While it may be difficult to implement, mutual defense is the appropriate patent analog to the GNU GPL.
The intersection of copyright and patent opponents is smaller than either on its own. If you are in both camps, support them separately with copyleft and mutual defense. The enemy of my enemy is my friend, or something like that.
Maybe someone at Red Hat thought they should get a patent on it before some asshole from another company did. If Red Hat patents the technology they can let anybody or nobody use it - at least they have control. Some outside group might have gotten a patent on the concept and held it hostage, forcing all users to pay.
I'm not a fan of silly IP actions, but Red Hat filing for a patent is NOT the same as Red Hat preventing others from using the technology. There are occasionally good reasons for this stuff and you should wait and see what the company does before you jump on them.
== Paul Rickard, Editor of The Microsoft Boycott Campaign ====
Personaly, I've like SuSE, I love QNX (it's different), OS X is just plain cool and then there's the mini ditro over at toms.net (tomsrtbt)
T Money
World Domination with a plastic spoon since 1984
Working within the IT industry is can be frustrating. If your job is to integrate and maintain technology, you will be dependent on vendors to produce that technology. Due to economy of scale, resources, the demands of R&D, and simply "how business is done", these vendors tend to be corporations. And eventually some of them are going to screw you over.
At this point it might be worth touching on the subject of morality. When something is labeled as "evil" it is a moral judgment. Something is opposed to one's sense of morality. One moral code often expressed within tech circles is functionality. Interoperability and functionality is a goal - anything that intentionally interferes with that goal is "evil". Business morality is often centered around profit. Anything that makes profit can be approved. Tech and business moral codes clash when technology is used to create a (often profitable) dependence on a product by limiting functionality and interoperability through technical or legal means.
Vendors will screw over the average tech worker when the company's moral code moves from a technical one to an aggressive business one. Take a look at Silicon Valley's history. It is chock full of techies starting a business, the business growing, and then the techie is pushed out as the business-types move in. The only question is how aggressive the business types get. It is a part of how corporations work and how "business is done." A company that is "cool" now can change based on who runs it and who decides on company policy and direction.
The double-cross might not even come from the trusted company in question. A company's assets (products, patents, and other intellectual property) often survive the company. The patent that is held "just in case" by the "cool" company of today can be sold off at pennies-on-the-dollar in tomorrow's liquidation and become a tool for its new owner's questionable, aggressive business tactics.
Why the distrust of Redhat? Because we almost expect a double-cross. Its history, nothing personal.
It may not be fair. But then, its not the game that we created. But the rules are well known. And Redhat should be very familiar with them by now. Knee-jerk journalism aside, Redhat should be prepared to explain their situation, their tactics, and provide a system of assurance to their customers and community they work with.
Unless, of course, their moral code has shifted.
Ok, Red Hat's going to be the next microsoft at this rate.. right? I mean.. patents.. didn't they have odd licensing and odd software in there somewhere? A big company for the whole desktop thing and such.. least, I could see it going there. That's right. Competitive upgrades from other linux distros. That was bashed a bit on the register, I believe..
But now all are complaining about how it's patenting things.. Well, that may be, but going by the title of the patents, they look like they could be actual legitimate patents. Nothing like the whole wheel thing that went on.
So.. they might be on the way to proprietary code, closed source, and all that.. Well, GNU will keep it open. and we all like open because, unlike how Micro$oft does it, it allows people to fix bugs and work with the code they're given. So.. it's not bad there, either, imo. Or at least it can't be.
So.. give Redhat a break. They have to make money some way, and this may be what works on their end. Even doing what it's doing, it seems to be being nice.
Just my few pennies. But what do I know.
-DrkShadow
I'm not going to defend these patents, but keep in mind that the claims are ANDed together, not ORed. Don't read the first claim and exclaim that Red Hat patented reports. It patented a static HTTP server that uses an object cache in an O/S kernel and meets the characteristics of all twelve claims.
The question is, as with all patents: is this a novel, non-obvious (to one skilled in the art) leap from the existing prior art? I doubt it.
Why would a Gilmore patent alienate IBM and HP? They just have to pay license fees, like with a regular patent. And since they are a commercial entity, they can afford to. The Gilmore Patent does not put any more restrictions on its use that a regular patent, and it is most definitely not "religious."
It has been discussed on slashdot before that if Open Source advocates held software patents that other software patent holders want to use then cross licensing agreements might be set up allowing Open Sourcerers to duplicate the functionality of those programs that you say are forever off limits. Well I say we need more software patents to be held by free software friendly parties as well as a legal mechanism to pool those patents and use them as bargaining chips.
What would the ramafications be of requiring the use of technology described by a patent to be open source? Microsoft would have no way to incorporate a kernel-based web server into their IIS product without opening significant amounts of source code to the public.
Oh shit! I forgot to click "Post Anonymously"...
"Screw" other Linux distributions? Wow.
Of course, RedHat has to *compete* with other Linux distributions, but "screw" them? I seriously doubt it.
Red Hat is not my favorite distro (although it is one of several that I use), but It's a pisser to see all the irrational, unfounded RedHat bashing that goes on.
Red Hat is one of the biggest Linux distributors (probably *the* biggest here in the US), and many people seem to feel that just because they are big, they are also evil.
First of all, TUX (and most everything else that Red Hat has added to its Linux distro) is licensed under the GPL, and others have pointed out that the GPL provides that a free license effectively be granted for any patented part of the code. (The method and apparatus are still protected, which gives them the protections they are trying to get against MS and others using the technology in a proprietary product). On this point alone, any fears of Red Hat screwing other Linux distributions seems little more than paranoia.
Red Hat is by no means a perfect company (is there such a thing in the real world?), but they have gone out of their way many times to help and assist the goals of Linux and open source in general. Of course doing something for Linux as a whole also benefits Red Hat, but because of the nature of open source it also benefits everyone as well.
Many people don't seem to realize that because they are bigger than most Linux distributors, they have some extra reasources that others don't have to apply to general causes. For example, pushing Linux in education, lobbying to fight really bad laws like the DMCA and the Hollings Disney protection act, providing the credibility and support to get Linux into large corporations, and many other things that in the end will benefit everyone in the OSS world.
Moreover, Microsoft has shown the ability to steal ideas from others. Software patents may be bad in general, but Red Hat is actually acting responsibly to protect IP that they've licensed under the GPL. Assuming the patents are valid, which I'm admittedly not in a position to evaluate, this will give them the ability to further protect the ideas in the GPL'd code from abuses by MS and others while still making the technology transparently and freely available to the open source community. Because the code is GPL's, the patents are actually a benefit rather than a liability.
The USA is not a good example, because in many ways, it is the Microsoft of the world (pushy/bullying -- even to friends... but I guess US products don't suck as much as MS products. :)
It seems to me like the "method and apparatus for atomic file look-up" is an old technique. See, for example, Stanford's Cache Kernel, which is entirely built around the idea of the kernel keeping caches (address space translation, file name translation, etc.) and faulting to user mode processes.
First off: I trust you, Alan, if for no other reason than you still have the bullocks to post here. Call me a lemming, but if you're down with the patent thing, then I'm fine too. I don't like patents, yet I have my name on 17 pending patent applications. I'm not sure they'll pass muster, and I'm not sure I'll care either way. They were filed a while ago, when I worked for a very pro-IP company.
Having said that, it *is* a little weird that Red Hat is patenting stuff. The suits want CYA, the lawyers want accountability, but the guys that wrote the stuff you distribute don't like patents. I didn't either, until amazon.com decided they could patent my finger acting upon a microswitch in the mouse on my desktop, then I realized that everyone had better patent whatever they can before the unscrupulous money weasels got their act together. We're well into the 3rd act.
IM(V)HO, patents aren't that great unless you're on the right end of the licensing agreement and I wish we didn't have to have them all. But we need to have them, if only to keep the bottom feeders of the world from reinventing the tux wheel and then patenting it out of existence. If I invent some novel "Method and Apparatus for the Extraction of Novel Nutrients from Common Playground Sand..." and then GPL/BSD/give-it-to-charity, can someone then patent a slightly reworked version and sue me for doing OSS work? Is that the precedence under which the license I pick wants to be tested? So, completely non-hypothetically, is Red Hat taking the so-called pre-emptive strike here?
I'm hoping -- as a very long time RHAT user, and a stockholder as long as there's been stock to hold -- that you guys will use some clout and cash to take out a patent or fourteen for the "good guys", as nasty as the conept sounds. If you have a claim to something novel, than have at it, I say, as long as you give back the key bits to the world. If you don't, then someone else will just patent it and hold onto/license it. So what sort of use to regular humans will be granted with the rights given RHAT by the USPTO? Does your company have a policy for this sort of thing?
I don't mean to grill you, but it *is* weird. Maybe I should think about patenting the process by which an OSS, for-profit company patents new ideas in order to give back to the world by protecting those ideas... :-)
-B
Ash and Hickory, straight-grained and true, make excellent bludgeons, dandy for the cudgeling of vegetarians.
The problem is not that Red Hat filed for the patents - the problem is that they could file for the patents.
Software patents are ridiculous - they are the work of ONE MAN who both pushed the idea as a practice and later became a judge and ruled on the legality of his own creation! . Can you say "conflict of interest"? Historically software was properly ruled out as being unpatenable.
RedHat is a tinv company compared to Microsoft, so I think the screaming, "RedHat goes to Redmond" is a bit premature. They are of course trying to compete with other distros or else they would not have offered those competitive upgrades for Mandrake and SuSE users. I don't think they are either as bad as some claim (rpm's etc) or as user friendly as SuSE or Mandrake. It's a company trying to make some money in fairly difficult times.
I really would wait and see what happens. If they start throwing cease and desists and legal suites around, that would be a better time to start an outcry.
And so it begins, the road to hell is paved with good intentions.
I can't blame RedHat, they are in this for the business, you can't give away everything if you are going to make a profit.
But I can't say that I like these kinds of patents, the are not good for a free market, for healthy competition, nor for non-profit software or the users.
Something needs to be done.
There has been no official statement on this yet, and probably the final wording of the license isn't done yet, but in short, this will not be used against any open source projects or companies.
Software patents are evil, yes. But if you can't get rid of them, you unfortunately have to play the dirty game if you don't want to be sued for infringing on patents like "text in a window".
There are several things I totally dislike about Red Hat (such as their ultimately stupid choice of default desktops), but there are plenty of other things that keep me here, like the strong commitment to Open Source.
If that were to change, I'd be out of here the same day.
As long as you see my posting from a *@*redhat* address, don't worry about things like this.
This message is provided under the terms outlined at http://www.bero.org/terms.html
TUX is the main performance competitor to IIS.
Uh, no. That would be iPlanet, thttpd and Zeus.. Especially Zeus.
TUX is a cute proof-of-concept, and if RH can get some leverage over MS by patenting it, I'm all for it. But let's not kid ourselves about who IIS' competitors are.
News for Nerds. Stuff that Matters? Like hell.
Folks... RedHat will stay a supported of the open and free just as long as it's profitable for them to do so.
They are a publicly traded company with a duty to maximize shareholder value. If you think that means giving things away they could be charging money for, think again.
Yes, they give away some stuff for free. Why? Because they want to? No, because they have to, or because it's in their best interests to do so (ie: Driver development for linux, etc)
As long as you see my posting from a *@*redhat* address, don't worry about things like this.
Yup. And Alan Cox. You guys are my barometer.
You'll find that, in general, you'll make a complete fool of yourself if you actually read the posts that you're responding to.
The thread was about "performance competitors" to IIS. As lovely as apache is for many uses, nobody (including the apache authors) would pretend that it's anywhere near in the same performance league as IIS, Zeus or iPlanet.
And I didn't mention Apache at all in the first place.
News for Nerds. Stuff that Matters? Like hell.
The idea is very similar to the GPL. Maybe we need a general "patent GPL" - one which is not a "policy", which can be changed later, but a stronger assignment of patent rights to a GPL'ish foundation in defense.
Maybe it's time to revive the League for Programming Freedom, but along these lines.
Sig: What Happened To The Censorware Project (censorware.org)
Besides, I seem to recall the GPL protects us against anyone integrating incompatibly licensed code into GPL software.
Except that Linus has given a special exemption to kernel modules. Hence, no dice.