Patent Concerns Unlikely To Nix Munich Linux Plan

MonkeyDev writes "Yahoo is reporting that Munich is ready to move forward with plans to 'abandon Microsoft Windows in favor of upstart rival Linux. The council is expected to take a calculated risk and vote through the move, despite concerns about possible software patent infringements in the face of coming European Union legislation that caused months of delay.' Not everyone is excited about it. A software developer at MySQL claims 'Linux violates 283 U.S. software patents.' How does the Linux community respond to these claims?" (Florian Mueller, the MySQL developer mentioned, isn't opposed to Munich using Linux, though -- just the opposite.) Update: 09/29 02:22 GMT by T : Marten Mickos of MySQL AB writes with a correction: "Florian Müller is an independent software developer and entrepreneur. He is ALSO an advisor to MySQL AB but he does not work for the company. He is presently engaged in coordinating opposition against software patents in EU, and thereby doing all of us within free software and open source a great favour."

Comments

Wrong Job

[mod_critical]

A MySQL software developer? Hmm, it would seem that it dosen't bother MySQL AB too much, so why should it bother Munich.

Seriously, if a MySQL developer is worried about the legality of running Linux then maybe he has the wrong job ;)

Re:Wrong Job

[bumski]

Seriously, if a MySQL developer is worried about the legality of running Linux then maybe he has the wrong job ;)

More like a MySQL developer worried about the promulgation of bad patent law.

Re:Wrong Job

[jdhutchins]

Linux using patented ideas may or may be a great problem. Some of these patents are probably in code contributed by companies. When you open up code under the GPL, you are also saying you won't use those patents against derivatives of that code (that's part of the GPL). So 'violating' may be the wrong word in this case, becuase is (hopefully all) of the cases, they've been given to the Linux movement.

Besides, I'm sure Microsoft has looked into patents that Linux might violate, becuase they're looking to destroy Linux, and a legimate patent claim would certaintly help. If they had found something, I'm sure they would have used it by now.

Re:Wrong Job

[NialScorva]

Or perhaps there's enough companies with other patents that MS can't be sure that they can enforce one of theirs without starting a patent war with the likes of IBM or HP. Patents work on MAD principles among big businesses.

Re:Wrong Job

[AstroDrabb]

Besides, I'm sure Microsoft has looked into patents that Linux might violate, becuase they're looking to destroy Linux, and a legimate patent claim would certaintly help. If they had found something, I'm sure they would have used it by now.

Well, the only thing MS could sue over would be patent violations of MS patents. Florian Müller who claims to have examined _all_ of the Linux kernel and found 283 patent violation, might not have found any MS held patents. Howver, I doubt Florian Müller's claims are correct on the number of patent violations. It would take tons of research not to mention complete knowlege of all US software patents to know what to even look for. That is why I doubt Florian Müller's claims.

Headline: Munich EXPECTED to approve...

[Infonaut]

Sheesh.

This is not news until the vote actually occurs.

Re:Headline: Munich EXPECTED to approve...

[freakmn]

They changed the headline, so nobody will get your post... It used to be: "Munich Approves Linux Despite Patent Concerns." It is still listed on the front page as this, but the headline has been changed, without any mention in the summary. Soon they will probably add that to the summary, making my post obsolete...

claims ?

[Anonymous Coward]

A software developer at MySQL claims 'Linux violates 283 U.S. software patents.' How does the Linux community respond to these claims?"

Hmmm, easy answer : back up your claims, show us the list.

Re:claims ?

[zyridium]

Bad bad idea... If they are publicly identified then they have to be defended....

Re:claims ?

[Geek of Tech]

Not really sure whether it would be a surprise or not, but I can just see SCO suing said developer for defamation against their IP. (Hey they think it's theirs anyway.)

Re:claims ?

[Anonymous Coward]

Not making this public is somewhat analog to security through obscurity, and we all know how good this kind of strategy is.
Give us the list, so we can find the code which need to be fixed/changed.

Re:claims ?

[TedCheshireAcad]

A software developer at MySQL claims 'Linux violates 283 U.S. software patents.' How does the Linux community respond to these claims?"

The same way we respond to any other question about our beloved operating system: RTFM noob.

Re:claims ?

[TWX]

you don't want to see a computer/technology patent war. everyone would be guilty of something. Heck, IBM would have almost everyone by the balls if they chose.

Since patents expire after seventeen years it's wise to just let them go. It's definitely safer.

Re:claims ?

[flacco]

you don't want to see a computer/technology patent war. everyone would be guilty of something. Heck, IBM would have almost everyone by the balls if they chose.

luckily for open source / free software advocates, ibm is currently fondling, not squeezing. :-)

Re:claims ?

[mmde]

The number 283 presumably refers to the 283 patents identified by Open Source Risk Management (OSRM) as potentially being infringed by Linux. Then again, I don't believe they have revealed which 283 patents they are talking about either.

For more details, see http://news.com.com/Group:+Linux+potentially+infri nges+283+patents/2100-7344_3-5291403.html

Re:claims ?

[suckmysav]

Nevertheless, it costs money to defend yourself from the demands of the patent holder, irrespective of whether their patent is bogus or not.

Can someone explain software patents?

[QuantumG]

If you ship software that has code in it that is covered by a patent what does that mean? Can the owner of the patent hit the author up for money? Can they hit the users of the code up for money? Can the author say "you, the user, are responsible for getting licenses for any patents that cover this code" and pass the buck?

Re:Can someone explain software patents?

[evslin]

I doubt it. The thing is ... how many lines of code go into a Linux distribution, and can you expect that the general public is gonna go through all of the source on their own and try to determine if any of it was stolen? Even if you looked through the kaleidescope long enough and tried to make it look like they were receiving stolen property, that'd be a tough mountain for the prosecution to climb if something like this went to court.

Re:Can someone explain software patents?

[QuantumG]

Who said anything about stealing? Check out this patent. Every operating system on the planet does that. It's a required part of loading every executable file format since the early 50s. It's really really easy for someone to violate a patent without even trying. You write code, you make up all your own ideas, but because someone had that idea 5 years before you and hired a lawyer he's gunna sue you.

Re:Can someone explain software patents?

[KillerCow]

IANAL.

If you ship software that has code in it that is covered by a patent what does that mean?

Patents cover the manufacture, sale, and use of an item.

Can the owner of the patent hit the author up for money?

Yes, the author manufactured and sold it without a license to the patent. This is what patents mainly protect. What will be more likely that the patent holder will get an injunction saying that the author can no longer sell his product, and then seek damages.

Can they hit the users of the code up for money?

Yes, the user used it without a license. See cases involving geneticaly modified seeds.

Can the author say "you, the user, are responsible for getting licenses for any patents that cover this code" and pass the buck?

Yes and no. If they the author doesn't have permision to grant further licenses for use of the patent, then the users must get one themselves from the patent holder. But the author needs a license to produce and sell the item in the first place too.

Re:Can someone explain software patents?

[TheHonestTruth]

AFAIK, this is all accurrate, but the superset of being right. Patent claims are analyzed on a claim by claim basis. It all depends on the claim language

For example: I patent a method of doing foo. You build an apparatus for doing foo, but you never really do foo, the customer at the end of the line practices the method. You do not infringe, the customer does. Similar scenario: I claim a system for doing foo. You sell "f" and "oo" but you do not sell "foo." Your customer buys each and assembles them into "foo." They infringe, you do not.

This is why a lot of patents have a method claim or two and an apparatus claim and maybe a "means for". The idea is to cover all possible infringers (though you never go after the customer because they are poor). That all being said, my case law is spotty, and IANAL, but that is how the initial analysis goes down. I actually think there is a case stating the opposite of of my combination example, i.e., if you make the parts and ship them into the US, you infringer, but I can't be sure.

-truth

Re:Can someone explain software patents?

[belmolis]

If I understand it correctly, this patent is much more specific than parent suggests. The abstract and initial claims make it sound like the standard general memory allocation system, but if you read further it turns out to be intended for parallelized FORTRAN scientific computing. The purpose is to ensure that all of the necessary data is actually in memory and prevent swapping.

Re:Can someone explain software patents?

[TheHonestTruth]

The software developer and IBM/Redhat/bigLinuxVendor. Why IBM/Redhat? Deep pockets. Why name the developer just because? To scare people away. Adding another name to the Defendant list is no big deal if you're taking on Big Blue. The scare tactic is the one used by the RIAA. The RIAA is going after the little guy because the deterrent factor is much more valuable than the monetary settlement.

-truth

Re:Can someone explain software patents?

[jonwil]

People suing over patent infringement are usually after 2 things. 1 is money (especially if the infringer has deep pockets) and the second is to ensure that is the only one.

For example, if a big company makes money of a piece of software covered by a software patent, its in the interests of said big company to sue a little guy who makes free software that infringes on the patent. Even if the little guy has absolutly no money, it doesnt matter. The aim of the lawsuit is to ensure that distrbution of the software is stopped and that it can no longer hurt sales of the big companys software.

Although as has been proven many times over, stoppnig distrbution of any piece of software is truely impossible (just look at DeCSS)

Its the same with other patent infringements. For example, a big drug company will sue to make sure that cheaper versions of its patented drugs wont be sold. And a big biotech will sue to make sure that farmers wanting to use their genericly modified crops (and as we have seen, quite a few who dont want their GM crap but have become "infected" by mistake) pay up big.

Re:Can someone explain software patents?

[Halo1]

The fact is however that the abstract doesn't matter, and that each claim on its own is a separate monopoly. It doesn't matter much if you do not infringe on e.g. claims 10-14, because you still can be sued for infringing on claims 1-9 (and if you want to overturn the patent, you have to overturn each claim individually).

Re:Can someone explain software patents?

[lachlan76]

Claims one and two alone would cover every OS in existance:

1. A static memory allocation system for use in a computer system allocating memory for executing programs, said static memory allocation system comprising:
   
   linkage editor means for storing information about an amount of memory required at initialization of an executable program, in a management information section of a file containing the executable program by obtaining the amount of memory required, when each program is translated, assembled or compiled to produce a corresponding executable program;
   
   loader means for automatically loading the executable program into memory, for reading the information about the amount of memory required when the executable program is initialized, stored in the management information section of the file, when the executable program is loaded; and
   
   memory management section means for statically allocating the amount of memory indicated in the information read by said loader means is statically allocated memory for the executable program to be loaded into by said loader means.
2. 
   
   Okay, that covers working out how much stack space you need at compile time
   
   
3. The static memory allocation system according to claim 1, further comprising:
   program executor means for dynamically allocating additional memory, when the statically allocated memory is insufficient during program execution
   
   And that covers malloc()

Well, it seems like we have a patent against memory allocation.

These are US software patents

[virgil_attack]

Why would these guys in Munich be concerned with violating US software patents? Just as long as they don't become European software patents (although that doesn't look like happening).

Re:These are US software patents

[moonbender]

Actually, the German government was one of the few who originally took a stance against the proposed EU patent legislation. However, after a couple of (key?) changes to the proposition, they agreed with it.

Re:These are US software patents

[BlueWonder]

Actually, the German government was one of the few who originally took a stance against the proposed EU patent legislation. However, after a couple of (key?) changes to the proposition, they agreed with it.

The German Department of Justice seems to be in favor of software patents, but at same time it carefully avoids admitting so publically. From the Department's point of view, the latter makes sense, since almost all (all but a small number of very large) companies are strongly opposed to software patents here in Germany.

A day before the vote in the EU Counsil (May 17th, 2004), there was a protest in Berlin, and a speaker of Department of Justice told the protesters that Germany would abstain in the vote.

On May 18th, the following "compromise" was reached. The original text of article 2b

A technical contribution means a contribution to the state of the art in a field of technology which is not obvious to a person skilled in the art.

was changed to

A technical contribution means a contribution to the state of the art in a field of technology which is new and not obvious to a person skilled in the art.

(emphasis mine). That's right, the "couple of key changes" was to insert the words "new and"!

Of course, the law already states that a patent cannot be granted if there is prior art, so the effect of the change is exactly zero. How the change is supposed to prevent software patents is honestly beyond me, but nevertheless it caused Germany to vote in favor instead of abstaining.

re violation of n US patents

[pfriedma]

I wish that developers would, instead of noting that such violations exist, correct them. Now... this is not always possible... for instance, I'm sure that a patent or copyright exists for "displaying multiple pages through the use of a clickable scroll bar" and undoubtedly, more than one OS has this functionality. Perhaps the issue will boil down to not whether or not parts of Linux violate copyrights but rather, whether or not said copyrights are even enforceable in the first place?

Re:re violation of n US patents

[virgil_attack]

I think that the patent office should get some intellegent people in there and stop issuing these ridiculous patents we keep on seeing.

Re:re violation of n US patents

[ThogScully]

That's the point... largely, software patents don't deal with truly clever ways of accomplishing things. They deal with obvious things that are practically accepted as standard methods of doing things. They are written to prevent other projects from achieving interoperation without stepping on patents.

If software patents only covered truly novel ideas like the patent system was initially designed to do, then no one would have a problem with them.

I look at it like this... good patents cover the way something is accomplished and bad patents cover the accomplishment. A good patent something like a particular method of preparing a chemical that is particuarly difficult normally to prepare. A bad patent is patenting that chemical, regardless of how it's prepared, as if that chemical's existence is owed to the patent holder.

It's hard to make up a good example of this with software, because any program you use, you're only seeing the output, but that may well be the patented methods, like a scrollbar in your example, which could be implemented many different ways of course.

-N

Re:re violation of n US patents

[torokun]

Well, in fact many chemicals exist solely because someone created them artificially. There are many modern chemicals that do not occur in nature. For that sort of substance, the simple existence of the substance itself is often a great contribution to technological advancement.

There are four types of patentable subject matter under 35 U.S.C. section 101: processes, machines, manufactures, and compositions of matter.

Re:re violation of n US patents

[jonwil]

Basicly, the biggest problems with the patent system are:
1.not enough examination of a patent before its rubber stamped (if what some people here say about prior art for and how there exists great easy to find prior art for a bunch of these patents, then the PTO needs people who can find that prior art)
2.futher to 1, the next problem is that the whole system is geared towards approving as many patents as possible. It is more fanantially benificial to the PTO to approve a patent than to reject it. This has to change so that the PTO gets the same finantial outcome regardless of if the patent is passed.
3.It should be easier to get a patent overturned if proof of prior art is discovered. Also, the method of dealing with patent infringements should be changed.
There should be a single "Patent Infringements Tribunal" or something. All patent infringements and all cases where someone has claimed to find prior art would be heard there. There would be some kind of steps taken to ensure that friviolous prior art claims are not made.
For infringements, if the patent is found not to infringe in the case in question (or is overturned), the patent holder will be penalized. Ditto if prior art is found that makes the patent invalid. (although if that was introduced, things like the eolas vs microsoft case may not have happened because the MS lawyers are 800kg gorillas and the eolas lawyers are little tiny ants)

With the right changes, it is possible to allow genunely inovative software patents (although I dont like it, something like RSA does represent the investment of a fair bit of time Ron Rivest, Adi Shamir and Leonard Adleman and does deserve some level of protection) to exist but the crappy patents (software and otherwise) to go away.

So what?

[Threni]

> Linux violates 283 U.S. software patents.' How does the Linux community respond
> to these claims?

I just don't care.

Re:So what?

[lphuberdeau]

Well, 238 patents isn't much if you consider the amount of stupid patents out there. You can barely cross a street without violating one. In fact, I quite glad to see it's not that much of a disaster. Now if you remove things like using an image as an icon, drag and drop of files, pressing tab to change links in a browser, memory allocation, displaying characters while you type, ... you're probably down to 2.

Re:So what?

[benjamindees]

The conduct of the [freedom seekers] is a threat to the authority of the [arbitrary authority], and a threat to peace. [freedom seekers] has answered a decade of [arbitrary authority] demands with a decade of defiance. All the world now faces a test, and the [arbitrary authority] a difficult and defining moment. Are [arbitrary authority] resolutions to be honored and enforced, or cast aside without consequence? Will the [arbitrary authority] serve the purpose of its founding, or will it be irrelevant?

Re:So what?

[taj]

I agree, 183 patents isn't much if you consider the amount of stupid patents out there.

In fact at this rate there wont be any patents to worry about.

Upstart?

[Slapdash X. Hashbang]

when does Linux stop being an "upstart" in the popular press? It's getting on to 15 years old, and it's quite prevalent already.

Re:Upstart?

[pHatidic]

An operating system is upstart when someone you know is using it, it becomes prevalent when you yourself start to use it.

NOT a MySQL developer

[martenmickos]

Please read the source text carefully!

Florian Müller is NOT a MySQL developer. He is an independent software developer who ALSO is an advisor to MySQL.

And when Florian mentiones the patents, he is only quoting another source.

Florian Müller is engaged (successfully, I might add) in opposing the legalisation of software patents in EU. By doing this, he is doing all of us within the free software and open source world an enormous favour.

I am afraid that many of the postings on this topic are based on erroneous input data. Hope this helps to set things straight.

Marten Mickos, CEO, MySQL AB

Talk about posting too quickly

[Infonaut]

I'm thinking now I should just hit refresh for three minutes before posting, just to be on the safe side. ;-)

Worry much?

[Cyco(k)]

Linux has been around for 13 years, almost 14 years, and now they are just starting to worry about software patents. What is next copyright then the RIAA of Software gets into the mess. This is linux, not winux ( my new imaginary marketable overpriced OS, for point making purposes ). - Cyco(k)

Yeah...

[Audacious]

And how many of those 283 patents are based upon other patents which have already expired or are really not unique? (Many of the patents being issued today are only extensions of pre-existing patents which is why there are these long lists of other patents being referenced.) This is not to dump on those truly unique patents - it is to dump on those (like the usage of a laser light as a cat toy) patents which, to programmers, are so obvious as to make you sick that the Patent Office could actually issue a patent on the invention. As per this other /. article - there are a lot of people saying the Patent System is broken and needs to be fixed.

And How Many Patents Does MySQL Infringe?

[John Hasler]

> A software developer at MySQL claims 'Linux
> violates 283 U.S. software patents.

Linux _may_ infringe some of 283 U.S. software patents. And MySQL? Are they willing to tell us how many they may be infringing? Do they know? You can be damn sure they are infringing some.

Re:And How Many Patents Does MySQL Infringe?

[martenmickos]

John, this is a misquote from the start. MySQL has not claimed ANYTHING, nor has any MySQL developer. See my posting elsewhere under this topic.

A person who is only an advisor to MySQL has simply repeated something that someone else has previously said about Linux and patents.

I am sorry for the confusion, but we did not create it.

Marten Mickos, MySQL

Re:And How Many Patents Does MySQL Infringe?

[John Hasler]

I'm aware that he was quoting another source: OSRM to be exact. My point is that _any_ software project of any substantial size is certain to be infringing many patents. Linux differs in that someone actually went through and systematically identified all the likely candidates (I'm sure the missed some, of course.)

Re:And How Many Patents Does MySQL Infringe?

[MrLint]

Perhaps it would be wise to keep irrelevant mentions of associations out article copy.

linux patent violation #1:

[deathcloset]

How about Math.

Please explain to me why a computer program is not simply a gigantic math problem?

Can it's processing not be broken down into nothing more than binary operations of a function. A formula that some determined individual may write out longhand?

Sure, that blackboard may stretch to the moon (and be made of carbon nanotubes), but it is an equation nonetheless.

is it not?

I mean, it takes input values, and returns output values.

It's just a really useful math problem.

When did we suddenly become able to patent Mathematics?

Re:linux patent violation #1:

[bigberk]

When did we suddenly become able to patent Mathematics?

Simple; ever since business people and lawyers decided they were almighty God.

Re:linux patent violation #1:

[KillerCow]

When did we suddenly become able to patent Mathematics?

See the RSA algorithm. It is non-obvious, and deserving of a patent.

Re:linux patent violation #1:

[tool462]

How about Physics?

Please explain to me why a machine is not simply a gigantic physical reaction?

Can its function not be broken down into nothing more than charged particles exerting force on each other? A system that some determined individual may explain in detail?

Sure, that explanation may take longer than the lifetime of the universe, but it is a fundamental reaction, nonetheless

Is it not?

I mean, it takes in energy from some source, and converts that energy to work.

It's just a really useful physical reaction.

When did we suddenly become able to patent Physics?

-- End Tongue-In-Cheek --

Don't misunderstand my meaning--I do not support software patents. I just don't think your argument is a correct way to invalidate them, unless you intend to throw out the entire patent system all together (which is certainly a plausible option).

Re:linux patent violation #1:

[torokun]

If you reduce any machine or process to inputs, *a function* and outputs, you could say they're all mathematics.

But the difference in function has meaning to us, practical meaning, as humans. Software that can reliably pinpoint tumors in medical images is not "just mathematics." It has meaning and it has social ramifications.

You are trying to look at the issue, as many of the people here on slashdot try to do, completely objectively. That is, you are tending in the direction of trying to see the universe as it is without any subjective human categorizations. But human life and human society do not and cannot function this way.

There are distinctions that we can make between software and mathematics. The fact that you can generalize and generalize until everything is mathematics says nothing about what the practical attributes of software are.

In other words, you are looking at the uber-parent class's properties and methods instead of the ultimate derived class's properties and methods.

According to this logic, you might argue that all english textual trademarks are just letters, and letters are just information which can be represented in binary, which means they are all just numbers, so there's no reason to prefer any one number over another. To paraphrase, "Since when did we get to register numbers as trademarks?" Clearly, by this point, you've lost all concept of the actual properties of various trademarks in the human environment.

I think I'll just stop here. ;)

Re:linux patent violation #1:

[Monkelectric]

All logic functions are decomposable into smaller functions. However, if you look at things with that fine a microscope, then you are really just a collection of atoms. Obviosuly you're much more then atoms, and a program is much more then logic. Here's a joke Consultants tell themselves (I believe there is a version for mechanics):

A network consultant gets an urgent call from a company manager, their network is down and they need him *NOW*. He walks into the company, looks everything over, says "hmmm" strokes his chin, pulls a patch cable out of his bag and makes a connection. Instantly everything starts working. The frantic manager is relieved, and the consultant writes out his bill, for "network consultation, $500". The manager says, "this is outrageous, I won't pay this bill! I demand you rewrite it!" So the contractor writes another one -- "$5 patch cable .... $495 Knowing where to plug it in".

The lesson is this really... That although programs are only logic, they aren't just logic :) Just like plugging a patch cable in, isn't just plugging a patch cable in. Now, I happen to agree with you actually. Software patents are rediculous, and I think a better argument would go something like this:

Programs are expressible in *languages*. To a programmer, these languages are every bit as real as spoken languages. If a statement in a language coresponds to a word in human languages, then a paragraph might corespond to a function, and a chapter a module, a book a program perhaps?

So basically, we have people patenting the plots to short stories. And they're saying, *NOBODY* can use this plot, its ours, we invented it. Well really a plot is just linguistic expression, and frankly there aren't that many.

Re:linux patent violation #1:

[Brandybuck]

Software that can reliably pinpoint tumors in medical images is not "just mathematics."

As someone who works in the medical imaging industry, I can't argue with that. There is some amazing work being done in this field that truly deserves patents.

But at the same time, I have a hard time swallowing the idea that all 283 of those patents in Linux are of that quality. I suspect that they're all of the "so obvious no one ever bothered to file a patent until we came along" class of patent.

Re:linux patent violation #1:

[Halo1]

If you reduce any machine or process to inputs, *a function* and outputs, you could say they're all mathematics.

You cannot reduce everything to mathematics. You can approximate a lot of stuff, but not accurately describe it. The fact is that in software, we are actually living in an idealised world. As far as your software itself is concerned, you do not have to care about all those exceptions and annoying side effects from the real world (even not when your software is used to search for brain tumors in images).

Of course, if you use your software executed by a computer to steer an external device, then you may have to take into account the physical properties of this device. But that is completely independent from the fact that you use software. You'd have to take into account exactly the same things if you steered it mechanically or manually.

My point is that whether or not you use software should be completely irrelevant as far as patentability is concerned. If people think that great mathematical algorithms should be patentable because they can be used to find tumors in medical images, then why on earth only make them patentable when implemented in software? Where is the difference in investment or the beneficial effect to the economy that suddenly warrants the patent monopoly?

Either you think they should be patentable, or you don't, but the fact is that software automagically makes a lot of otherwise unpatentable stuff (such as business methods and mathematics) suddenly patentable.

But the difference in function has meaning to us, practical meaning, as humans. Software that can reliably pinpoint tumors in medical images is not "just mathematics." It has meaning and it has social ramifications.

When I write a technical manual about how to perform a chemical reaction, then if people follow that manual, this also has meaning and social/real world ramifications. That does not mean my manual is patentable, nor that the potential owner of a patent on the reaction can forbid me to publish and sell my book. The chemical reaction itself could be patentable, but that's completely separate from whether or not it's written down in a manual or described using software to steer a machine.

Software is also merely a description of something, just like the technical manual. It does not make sense to allow patents on descriptions (although software patents do forbid certain forms of descriptions, which is probably one of the reasons software patents do not include source code: otherwise, a patent itself could infringe on other patents!). It also does not make sense to change the patentability of something depending on whether it's described in non-technical English or in C. There are no economical arguments for such a differentiation.

According to this logic, you might argue that all english textual trademarks are just letters, and letters are just information which can be represented in binary, which means they are all just numbers, so there's no reason to prefer any one number over another.

That's a false analogy. A computer program nothing but a description in another (mathematical) language. The translation investments are covered by copyright. What you describe, may or may not be patentable. Some things are on purpose not patentable, because such patents are considered to do more harm than good (there is no economical law that states that patents per definition have a positive overall effect).

I think the only argument one could possibly make to justify software patents, would be in case they demonstrably would have lead to more innovation and a healthier economical situation in the software sector (or even in the economy as a whole, since software is not just used by software developers). They don't.

"Linux violates 283 U.S. software patents"

[bani]

How many patents does microsoft windows violate? How about osx? solaris? aix? hp/ux? Probably tens of thousands.

We only know about the linux 'violations' because the code is open. I'm sure if someone were to evaluate "those other operating systems" we'd find far more -- because there is no open public oversight of their code. They operate in secret, who's the wiser if they were deliberately violating patents?

Also, do any of "those other companies" provide indemnification to end users? No, in fact microsoft's license is almost exclusively to provide microsoft with indemnification from end users.

Using microsoft or any other OS isn't likely any safer than using linux, when it comes to patent violations.

The Linux Community's Answer

[Featureless]

'Linux violates 283 U.S. software patents.' How does the Linux community respond to these claims?'

You pulled 283 out of your ass. Those are just the ones you know about. You know why you pulled that number out of your ass? It's impossible to review the whole patent database and screen it against the whole of Linux. Humanly impossible. 283 is a made up number. May as well say 2,830 or 283,000.

An exact count of how many software patents are violated by Microsoft Windows, for instance, would be equally impossible - nay, more, because they keep their source code a secret; however, it is incontravertibly a similar if not higher number.

If you try to follow the U.S. software patent system you won't even be able to power a pocket calculator without a half-million dollars for attorneys and payoffs. Yes, you think I'm exaggerating, don't you.

That's why 100% of Americans ignore their own system. Everybody knows its ridiculous. Even SP's major proponents are afraid to use them because they fear the whole system will unravel if they test it, so all they do is occasionally shake people down, hit and run once and a while. They want to sue Linux over patents; they've been desperate to do it for years, and they're too scared of how badly it will backfire. They're probably right. So they're reduced to backdoors like SCO. And we see how well that cleverness works for them.

Re:Two Hundred and Eighty-Three?

[martenmickos]

Slashdotters, this is a very important discussion (the one on software patents), but let's start with accurate facts.

The 283 thing is old news and was just repeated by Mr Florian Müller (who is NOT a MySQL developer). See here:

http://news.com.com/Group:+Linux+potentially+infri nges+283+patents/2100-7344_3-5291403.html

I quote from that article:

- - -

Linux potentially infringes 283 patents, including 27 held by Microsoft but none that have been validated by court judgments, according to a group that sells insurance to protect those using or selling Linux against intellectual-property litigation.

Dan Ravicher, founder and executive director of the Public Patent Foundation, conducted the analysis for Open Source Risk Management. OSRM is like an insurance company, selling legal protection against Linux copyright-infringement claims. It plans to expand the program to patent protections.

- - -

So it seems that an important discussion has got onto the wrong track due to incorrect input information.

But let us discuss software patents! MySQL's official position can be found here:
http://www.mysql.com/company/legal/patents.html

Marten Mickos, CEO, MySQL AB

Linux and patents..

[euxneks]

'Linux violates 283 U.S. software patents.' How does the Linux community respond to these claims?"

Maybe it's not Linux which is what's wrong/broken. Fill in the rest.

No so fast

[deadline]

Linux violates 283 U.S. software patents

If some one said this to me, I would take the "Groklaw Approach". I would ask them first, what are the specific patents and second, how can he be sure that Linux violates all these patents. It would seem to me that to do a fair assessment of 283 patents would take a fair amount of time. So let's see the details.

My guess is that his answers would consist of words like "I", "don't", and "know".

Re:No so fast

[BeeRockxs]

You do realize that the study that said was made by OSRM, a company/foundation/whatever, who PJ, the woman that set up Groklaw works for?

Re:You could pull it out your ass as you suggest,

[fatwreckfan]

A working link:

http://www.internetnews.com/ent-news/article.php/3 389071

User input in a formal description

[DarkMan]

I can think of two ways, of the top of my head. Note that the 'equation' is, as with mathematical desrciptions of relationships, just a method of formal statment, and need not 'look like' a typical mathematical equation.

The first, and probably the simplest, is to use a formal description of a state machine. Thus, the system is normally in the 'wait for input state', and then it branches to a state determined by the type of user input. This encapsulates the interactive element in a single part, and closely resembles the typical structure of most GUI systems (with callbacks etc).

The other method is to treat the system as a function of infinite arguments and use combinator logic. Curry all the arguments into the function, up to the latest existsing, then as each piece of new input occurs, add another argument, and curry it out. Something like an RDBMS is probably better suited to this sort of desription.

This is, I think, an impure calculus - currying is normally invoked to allow description of multiargument functions withing lambda calculus which requires single argument functions. Nevertheless, I can see no major hurdles to describing a program in such a manner.

Granted, I've not had the chance to express any non-trivial programs in either form as yet (lack of time), but I think that some programs are better suited to one representation that the other.

Worth noting that there is no change in the description if the input is known all a priori, and processed from a file, or if it's all garnered piecemeal. Consider a shell script for an example where this is obvious.

A mouse action is probably best represented by a tuple of tuples, giving button down and button up coordinates - thus ((10,10),(20,10)) represnts a mouse drag action, and clicks fall out as degenerate coordinates.

Alternativly, for another option, look at how a pure functional language does a GUI. For example, the wxHaskell bindings for, (unsurprisingly) Haskell. Haskell (a pure functional language with lazy evaluation) programs are just an implementation of a mathematical model of computation. Here, the description is effectivly as a set of functions, where the 'user input' determines which function to run [0]. Techincally, this is no longer a set of equations, however I can prove that this is equivelent to the state machine description above.

Any errors, feel free to correct - this is all a bit rough and ready. I just felt that it was worth pointing out that just because it might be absurb, doesn't make it impossible. For example, the halting problem seems absurb to everyone whose not studied it (because it's obvious if a program will finish).

[0] Or something close to that. I'm not intimatly familer with either the language or the bindings, so I might be a little of with the description.

Re:User input in a formal description

[BarryNorton]

You're quite right that interaction is one of the reasons that classic ('functional') computation theory (with Turing machines being non-interactive, merely providing a computable function between input tape and output tape) is not the end of the debate. You have not taken the consequences far enough though.

Process algebras neatly take this concern together with the need to cover non-determinism due to concurrency, hence making the Pi-calculus a real rival to the Lambda-calculus as a fundamental theory of computation.

(Yes, I know, not even Robin Milner thinks it's actually the final word along this dimension and is looking for something more fundamental...)

U.S. enforces patents...

[the-build-chicken]

...rest of the world innovates on linux and open source to the benefit of their economies...

hey, that's your foot america...what are you doing with that gun??!?

Re:Two Hundred and Eighty-Three?

[martenmickos]

Unfortunately I don't think there is any mechanism for challenging *all* software patents in one go, although I would welcome one.

But there is plenty of good work done behind the scenes (and on the scenes) with the purpose of limiting or completely abolishing the patentability of software.

These things change slowly, and all of us need to be prepared to live WITH software patents until the present malfunctioning software patent system has been changed.

But everyone doesn't have patents

[MCRocker]

That's how the (software) patent game works. It's virtually impossible to avoid violating patents due to the very nature of software. When challenged, the big guns go through their patent portfolio and can usually find something that the challenger violates in some way. Then, a cross licensing agreement is hammered out. Thus, the big fish are protected, but any small fry can't compete.

So, yes, everyone is guilty, but the IBM's of this world can weather the war just fine, while everyone else gets wiped out. In fact, the war is already in full swing and has simply become part of doing business.

I think RMS covers this in his speech The dangers of software patents.

If patents applied to law practices...

[MCRocker]

It's interesting to note that patents apply to all sorts of unlikely things like software and business processes, but not to legal strategies, practices or processes. It seems as if the lawyers realized how badly that would muck them up and haven't applied the patent pricipals to their own field. I guess that there's lots of money to be made from messing up everyone else's business, but not their own.

It would be very bizarre to hear an objection to a legal argument because someone else owned the right to make arguments of they style that the motion used. Perhaps only Johny Cochraine could "play the race card" or something like that. Every other law firm might have to pay him a royalty to use the argument.

If someone could come up with a clear demonstration showing that software patents are as sensible as legal strategy patents, then I'll bet that the supreme court would overturn the current incarnation of the patent laws in a heart beat.

Re:If patents applied to law practices...

[dghcasp]

Actually, business methods are patentable.

Historically (Before 1998,) business methods were not patentable for the following reasons:

1. Business methods are not "technical"
2. A market monopoly might result

However, this was questioned in State Street Bank & Trust v. Signature Financial Group Inc., 49 F. 3d 1369 (Fed Cir. 1998).

Signature was assigned a patent, "Data processing system for Hub and Spoke Financial Services Configuration," which described a computer system for asset management, where mutual funds ("Spokes") pooled their assets in an investment portfolio ("The Hub,") which was organized as a partnership to offer tax advantages and economies of scale.

State Street had been negotiating for use of the patent. When talks broke down, they brought legal action saying that the patent was invalid.

The lower court found for State Street, saying business methods are not patentable. The US Federal Court of Appeals overturned this decision, saying that considering business methods unpatentable was ill conceived and unsupported by the Patent Act. In other words, business methods should be treated as any other patent claim.

Because of this ruling, business methods are now patentable as long as they can be implemented in software. This has been one of the drivers on the rush for software patents.

Sources:

• Duplessis et. al, Canadian Business and the Law, 2001 (background; most text adapted form there)
• Internet search on "state street v. signature" (verification of no supreme court appeal; current impact assessment)

  Disclaimer:
  IANAL.

Re:Two Hundred and Eighty-Three?

[Yenin]

Linux potentially infringes 283 patents, including 27 held by Microsoft but none that have been validated by court judgments, according to a group that sells insurance to protect those using or selling Linux against intellectual-property litigation.

Re:Algorithms

[ThogScully]

I considered adding a paragraph about algorithms being the appropriate parallel in software, but pulled it out just because of what you said here...

I guess at that point, you'd have to also confirm it's a non-obvious method that is being patented. If someone invents a truly novel algorithm of encrypting communication that doesn't involve really big factorizations or something, that might warrant a patent. If someone develops an encryption algorithm though that just does different factorizations or longer keys than normal methods, while different, that algorithm isn't really novel.

So essentially, I guess algorithms could potentially be patented ethically, but it would more or less be a math patent, not a software patent. And i would take a good mathemetician to identify if it was obvious or not, because I know I certainly couldn't decide...

And even with all that, this is a gray area, which is largely why patent reform is necessary in the first place. Innovation is being hindered by these questions since some of the most innovation right now is coming out of software development.
-N

How many patents does any software violate?

[Moderation abuser]

I'd put money on it that virtually any piece of software that anyone writes now violates one or more US patents.

It seems that the goal of introducing software patents was to give all software development to the big guys with patent portfolios of their own they can use for defence.

Why they are waiting...

[johannesg]

They are waiting until Europe has approved software patents. It is no use to them to only destroy Linux in the USA, they need to destroy it wherever they can. If they were to do so now, the opponents of software patents in Europe would gain a powerful weapon. If they wait a while, until software patents have become a reality, they can strike and destroy Linux in two major markets.

Expect them to strike as soon as the future of software patents in Europe is known.