Small is a relative term - we have now got to the level where software is written by individuals... I said people, as in everyone.
You're trying to have it both ways. You're stating that patents are a threat to an individual because they are a threat to a group of people working for a company. Then, you argue that patents should not be a threat to companies, because they can be a single individual working on his own.
It's pretty simple:
Software created for private use, either by an individual or by a company, is effectively untouchable by patents.
Software created as a product and sold for profit is, and should be, touchable by patents from other inventors.
You can argue these issues individually, but you're trying to conflate them. You're combining the drama of the poor individual coder oppressed by patents with the visibility of a company offering a commercial product. Pick one.
However, if nothing is done to head it off, we will see patent attacks on Linux soon, as one example of patent attacks on open-source software.
Ah, the looming spectre of the Linux patent assault. Why do you suppose this hasn't occurred yet, despite chicken-little prophesies by minor Linux advocates for at least five years? Why does Linus Torvalds not seem worried?
Answer: It would be a terrible business decision. IBM needs Linux as a free design platform. Microsoft needs Linux to dodge accusations of monopolization, and attacking it would damage the its reputation (which it's striving to salvage.) In short, no one who owns such an arsenal can derive a business benefit by attacking Linux.
You've got to face facts here. Patents are not weapons, threats, or land grabs. They are business tools. Businesses acquire and use them solely for business reasons. Any outcry over perceived use in a non-business sense is out of sync with reality.
Hardly - the reproducibility of software is key to the fact that it can be created by people for themselves.
You continue to miss the point of the analogy. The point is freedom to expand into new uses. This has nothing to do with physicality, or reproducibility. You're trying to avoid the analogy on an unrelated basis.
(And you're obviously not reading closely at all; I've made no reference to the windshield analogy raised elsewhere in these post comments. If you're not willing to read before responding, then I'm going to stop responding to you.)
This bears no resemblence to the software world I know. All the commercial projects I have worked on continued to release updates. Far more than free software, the commercial model depends on continuing to sell newer versions of their software, or else periodic licencing fees, or else both.
What are you talking about?
What software are you buying that requires mandatory upgrades or periodic licenses? Aside from the two niches I mentioned before - antivirus software and mapping software - not a single software product I own requires paid-for upgrades. Almost all of them offer patches for quite some time to fix bugs, and even add new features, for free. And I own a lot of software.
As to the cusomising - my whole point is that we have now reached the point where people do it themselves.
That would be a disaster for the many free software companies. Since they can make very little money on the software product under the GPL, the service aspect is their primary source of income. If we've reached the point where users can customize their own software, then you can kiss many of the free Linux distros goodbye - their business models are in tatters.
The simple fact of the matter is that technology, the availability of it, and the education necessary to use it has reached the point where it is no longer the preserve of big companies with heavy R&D budgets.
What do you mean, "no longer?" I remember the software market of 10 or 20 years ago as very heavily populated by a whole lot of small companies. Even Microsoft was a niche player in the OS market, and just starting to branch into applications. If you equate software patents with big companies, then you should be happy to see these trends heading in the same direction.
All the big companies are doing with their patent efforts are trying to force people to depend on them for things they can produce themself.
Show me an instance where a patent was asserted against a person. Go ahead - I'll wait.
Patents are business tools that companies use against each other. What you do for your own purposes with a copy of Visual Studio is largely irrelevant to patentees, unless you turn the result into some kind of product. Why would they waste their time and money stopping you from doing something in private that doesn't impact their business?
All this talk of cars is meaningless and misleading. If a few hundred engineers get together to build one, then only one of them can drive away, but a few hundred get together and write an OS, or a web-browser, and everyone gets a copy.
The purpose of my analogy had nothing to do with the physicality of the auto vs. the intangibility of the software. That is irrelevant.
My point was that you can choose between two product models. One product costs more up front, but does everything you might want to do with it (within reason.) The other is free, but every time you want to adapt it to a new use, you have to pay someone to customize it for you. I prefer the former model; the ongoing need for more service is a debilitating hassle.
So what if we have reached the stage where people can produce sophisticated software as a pasttime?
Hey, I do that now. We're already there. And guess what - I can do that as much as I want, for my own purposes, without getting sued for patent infringement.
Then you have a big problem. In fact, it's intractible.
Let's say you're the USPTO, and someone applies for a patent for a new process of drug identification. It involves finding proteins that inhibit a target protein and stop cancer. Presuming it's novel and nonobvious, do you allow it?
Now here's the crux of the problem: The patent validity and enforceability have nothing to do with how the applicant intends to carry out the process. Maybe he'll carry it out in a test tube, or maybe as a simulation. Maybe his competitors will carry it out in a test tube, or as a simulation. What matters is whether someone is performing the process, which is the invention.
I've had this thrown at me before, and it's silly. No, I haven't studied the grammar of US patents, but I'm a computer programmer and patents read exactly like programs.
What? Really? I've never seen a program that reads like this:
"What is claimed is: A medical diagnostic x-ray system, comprising: an x-ray generating device operative to produce medical diagnostic x-rays and thermal energy; a hydrogenated polyaromatic fluid operative to circulate about said device to absorb at least a portion of said thermal energy and electrically insulate said device, said fluid comprising a hydrogenated compound that upon exposure to said x-rays forms an unsaturated hydrocarbon and hydrogen atoms; and an effective amount of a hydrogenation catalyst for interaction with said fluid, said catalyst reactable with said hydrogen atoms and said unsaturated hydrocarbon to recombine said hydrogen atoms with said unsaturated hydrocarbon."
Seriously, patents and code couldn't be more different. A program contains the actual instructions, in precise detail and chronological order, for carrying out an algorithm. A patent broadly describes an invention, focusing on the key pieces/advantages and omitting much detail.
Maybe an analogy would work for you -
Computer code : Patent:: Recipe : Photograph of Food
I'm not quite sure what your position is - are you arguing for or against software-as-a-product?
I'm arguing for it. I believe in promoting the interests of customers over the interests of business. Moreover, I believe that businesses often work best when they also promote the interests of customers, especially compared with an alternative business model that costs customers more and gives them less.
For instance, I hate the idea of DRM - I hate it to its core. It's a movement to sell people music that's less useful than music on a CD, and probably for more money (additional hardware, re-purchases, etc.) I froth with irritation when companies pitch this inanity to customers as "trusted computing" or similar nonsense. It's terrible business - it alienates customers, and probably won't work - and I suspect that the RIAA will regret it in the long run.
I feel similarly about continuous-payment software models. Hope that's obvious in retrospect.
Of course, those industries would want ways to make money by creating something once, then getting paid for it over and over. If the market weren't encumbered by IP laws...
What in the world do IP laws have to do with rental business models? You're lashing the spectre of technology stagnation to your disfavor of software patents with little more than rhetoric and chewing gum.
Even software has to evolve to adapt to changes in circumstance.
A few software products, sure. I totally get why Symantec sells subscriptions to Norton Antivirus: it's expected to respond immediately to every new virus. Same thing with mapping software, since the maps need frequent updating. Those things make sense, and so their customers are grateful to pay for them.
But very little software fits that need. Look at the leading software apps - Windows, Office, Photoshop, file compression, email and instant messaging clients, web browsers, media players. I just want them to work, forever. The company is expected to patch product defects - not because I pay them to, but because their product shouldn't be defective. But I don't need updates; if I really want new features, I'll buy the next version.
You might be able to write a perfect, bulletproof program that satisfies every current customer need - but the customer WILL have new requirements, the platforms that the software needs to run on WILL change, the customer might have new projects (and would prefer to contract with you again if you did a good job the first time), etc.
Fine. So you go write a new version, and I'll think about buying it. If it's not worth it to me, I'll stick with the software that I bought and own. What's wrong with that model?
That'd be a lot easier & cheaper if they didn't have to worry about violating IP laws.
There you go again, with superfluous use of "IP laws" as some kind of talisman. And in this instance, you're completely wrong. What prevents companies from tweaking purchased? It's not IP law of any kind. It's the license agreement with the vendor - the private contract under which the software was sold. "IP laws" have nothing to do with it.
It seems to me that a "process" involves physical phenomena and a "computer-implemented process" involves, for starters, a computer, running some kind of digital representation of something that may appear to correspond to reality, or may be totally imaginary.
You need to read more patents. Patent practitioners go for breadth: if the inventor has invented a process, the patent usually claims a "process," regardless of how it will be carried out. Check out this claim from U.S. Patent No. 4,405,829:
A method for establishing cryptographic communications comprising the step of: encoding a digital message word signal M to a ciphertext word signal C, where M corresponds to a number representative of a message and 0.ltoreq.M.ltoreq.n-1 where n is a composite number of the form n=p.multidot.q where p and q are prime numbers, and where C is a number representative of an encoded form of message word M, wherein said encoding step comprises the step of: transforming said message word signal M to said ciphertext word signal C whereby C.ident.M.sup.e (mod n) where e is a number relatively prime to (p-1).multidot.(q-1).
Do you see the word "computer" in there? No, it's just a process. It happens to be the process of RSA encryption and decryption, which, in virtually every instance, is carried out by a computer processor. But on the off-chance that you want to decrypt a message with a slide rule or an abacus, you would still have violated the patent (which is now expired.)
And don't try focusing on the mathematical nature of the process as a means of determining bad "software" vs. good "non-software" patents. You can easily embed this same process in a novel circuit, and patent that. So are you also opposed to patents for circuits?
The delineation gets even more gray when using high-level object-oriented programming. Such a "process" may be a control mechanism for an automobile assembly line, involving cars, paint, welding tools, etc. Does it matter if the process is a hardware controller, or if it's a purely sofware simulation? Should it?
Or, how about drug identification: should the process of finding a protein target for a cancer drug be patentably different if carried out in a test tube or a software simulation?
I've read patents. Some patents are simply a long list of circular definitions, and do not deserve to be patents because they don't refer to anything real or unique.
Sounds to me like you haven't studied patents enough to understand their style. Simple words like "means" and "comprising" and "for" have some special meanings that you wouldn't grasp from just reading patents. On the other hand, patents that are obfuscated - by intention or by incompetence - are not valid and not enforceable; the patentee has wasted its own time and money to get a worthless patent.
Unless you can explain how to patent a dictionary, a newspaper, a magazine or a novel...
You can't, under the "printed matter" doctrine. Computers are different: they don't just tell you how to do something; they can actually do it for you. Hence the distinction.
Red Hat sells suppport services for their linux distro. Microsoft sells support services for their OS.
Never mind the total revenue - let me ask this: How often does the average user have to pay Microsoft to fix a problem with their OS?
Certainly there are companies that would like to pay to have their software heavily and exactingly customized, or to have extremely rare, highly case-specific bugs fixed immediately. It's good that those services are available. The question is how frequently those calls have to be made. If it's so often that it's your primary source of income, then your software is probably broken beyond repair.
Frankly, I can't believe we're debating this. Imagine a line of dirt-cheap automobiles that required mechanical adjustments every time you drove to a new city. If you plan to take many road trips to various unknown destinations, why would you ever put up with this hassle? Why not pay more for a car that goes everywhere for free?
Pet peeve time. Exceptions don't make rules. they weaken/break them.
The phrase means: "If you have to go to that extreme to find a counterexample, then that shows the strength of the rule in the typical case." This exchange was a perfect example of that. Consider our dialogue, condensed:
A: "Companies can't survive by selling GPLed software."
B: "Red Hat sells GPLed software, and they survive fine."
A: "Red Hat survives despite selling GPLed software - its distro is a large net loss. It socks up the losses by deriving income from an ancillary service. Your example just strengthens my point."
Sounds a lot like 'But if your business plan involves selling Microsoft-dependent software,(and you are not Microsoft) you'd better keep your resume close at hand."
Well, yeah, because your business model is illegal and you're going to jail.
If you are Microsoft, though, you do very well selling Microsoft software. Similarly, if you are Adobe, then you do well selling Photoshop. Show me a company that has made enterprise-scale profits by selling GPLed software - not an ancillary service, but the actual software.
AFAIK, MySQL is doing just fine as well.
1) They do well because they sell ancillary support. In fact, a whole lot of other companies also profit from MySQL support services.
2) MySQL does well because it doesn't have any competitors in its niche: a small, lightweight, scalable DMBS. It succeeds for the same reason WinZip succeeds: no one else really wants its space at present.
That is going to change. Microsoft's future OSes will increasingly have native DBMS capabilities. It starts with WinFS, a file system based on a relational database, and ADO.NET, which provides (clumsy) format-independent data access. Let's return to this conversation in, let's say, 2008.
That's because you can't get the idea of software-as-product out of your head. If you want a solid business model that doesn't depend on copyright law, then you make a living by charging for SERVICES.
No, see, I completely understand why a software company would want that model to succeed. I similarly understand why the media industry wants DRM.
In both cases, I can't see why customers would want it. That creates a ripe opportunity for a competitor to give them a more favorable one. Examples: DivX died a painful but much-deserved death to DVD circa 1999; long-term cellphone contracts had their market shares stolen by no-contract businesses.
Does a plumber charge you every time you use something they fixed?... Why should programmers be treated any differently?
Pipes carry water. We expect them to rust, seals to fail, toilets to clog. Physical apparatuses experience wear and tear - it's unavoidable. Customers accept that.
Software has no such inherent limitation. If your company thrives on having people pay for software fixes, then face it - your software sucks.
Somebody wants software created or modified, software developer does the work, gets paid, goes on to do the next job.
So instead of buying software that works as you expect right out of the box, you get some raw materials for free, and then you have to spend time and money paying someone to make it work for you? Is that how it works?
Software businesses can survive just fine by providing services.
That's an oft-cited model. It's also baffling to me.
A software product that requires you to call Technical Support, even once, has failed you as a customer. It needs better documentation, or a user interface overhaul, or a security or stability upgrade, or a more coordinated online community. (More likely, a combination of these factors.) Taken the next step: A software product that requires you to keep paid company support permanently on hand has cataclysmic failed you as a customer.
More simply put: Well-designed software, de facto, does not require you to pay someone else to help you use it.
Red Hat seems to get away with it because we expect OSes to be flaky and complex and in need of support. We see it as the open source equivalent of a patching service. We don't see this as an asset; we see it as an annoyance that's tolerable because the TCO is still better than Microsoft's. That doesn't mean it's a good thing.
Here's what I just don't get about advocates of this concept. What do you think of the "software rental model," where you don't own the software, but you have a limited-term lease and have to keep paying? Bogus, right? Why? Because we inherently feel that software sales should be a one-shot deal: you pay the price, and you use it indefinitely.
If you agree with this, then what's the difference between paying continuously for use, and paying continuously for access to a user manual? Is it just the fact that one initiative is sponsored by a company we like, and the other by a company we hate? Is that remotely logical?
Yup, that is why Red Hat went bankrupt so quickly.
Oh, wait a minute..
Red Hat's primary product is not its Linux distro. That may be its most visible feature to most consumers, but that's not really its business. The distro is essentially advertising for its core product, which is technical support.
Google operates the same way: the search engine is its most visible facet, but its core business is selling sponsored links. In this way, it's backwards: its search engine is advertising for its core product, which is an advertising conduit.
But, in fact, this is the exception that proves the rule. The rule is: GPL-derived software is not commercially viable as a product. You may be able to exploit it, e.g., as leverage for a related service or a related (non-GPLed) product. But if your business plan involves selling GPL-dependent software, you'd better keep your resume close at hand.
My suspicion is that, if IBM holds patents that threaten Free Software, RMS doesnt care if they're a strong Linux advocate -- he would rather force them to grant a license to all their software patents to any and all software released under the GPL.
That would be a disaster for IBM and any other GPL-relying company.
Why has IBM built up a massive patent arsenal but rarely used it? It's a defensive measure. If anyone sues IBM for patent infringement, it can threaten to counter-assert its 40,456 patents.
Now imagine if IBM had to grant that attacking company free licenses under all software (and patents) developed using GPLed code. Doesn't that vastly undermine IBM's defensive patent strategy?
Broadly extrapolating: Doesn't this render any company utilizing GPLed software a magnet for infringement lawsuits from non-GPL-reliant companies?
However, the meaning of the word corporate here has almost exactly the opposite of the way we think of the word today.
May those with mod points reward you. This is an extraordinarily interesting point, even if off-topic.
You seem like the right person to answer this question: Is there a term for the kind of corruption we have today - corporations fully co-opting government?
On the one hand, I suspect there is, since this isn't a new concept. Even Adam Smith was terrified that laissez-faire economics would cause corporations to amass enough money to do so.
On the other hand, maybe not. I think that such amazing power to control the masses via advertising/savvy media skills is a recent invention.
If no term is available, can we coin one? Foxism, maybe?
Remember, that the GPL doesn't primarily protect the developer's Freedom, it protects the USERS!
Well, to be clear - it primarily protects users who are also developers. The defining trait of the GPL is that your users may modify your code and pass along the modification on similar terms. If you just want to give rights to non-developer users, then much more restrictive licenses are just as good.
Being a long-time observer of GNU and the GPL I'm confident in going out on a limb and predicting what such penalties would be. They would be identical to the penalities for using proprietary software - i.e. you can't use GPL'd code in a program that contains implementations of software patents nor implementations of DRM.
That raises a very interesting question: Who enforces this rule?
Let me illustrate a typical case where someone would want to trigger this rule. Let's say you're Microsoft. You take some GPLed code - e.g., an algorithm for converting binary-coded-decimal numbers to pure binary representations (see Gottschalk v. Benson) - and then you patent an improvement to this process, and you embody the patented improved GPL code in a product, say, Clippy's Wild Advntures.
A competitor takes your improved BCD idea and embeds it in his product - let's call it's Eidos's new game, Tomb Raider: BCD Adventures. You sue for patent infringement; they want to raise this novel GPL defense against you.
Now, how would that play out?
The dispute over the patent does not involve your embodiment of the patented algorithm, i.e., Clippy's Wild Adventures. It involves your patent and Eidos's game. Eidos would have to go fishing through your products, including totally unrelated ones, for an embodiment of the patent. Thus, if Eidos asks you if you had an embodiment of the software, you could refuse to answer the question as irrelevant to the dispute over your patent and their product. And it can't just go reverse-engineering all of your products: in addition to be wildly expensive, this would probably violate their software licenses to each of the products.
Worse: Even if Eidos proved that you had violated the GPL, it doesn't help them. It merely creates a new, totally unrelated dispute between you and the Free Software Foundation. Eidos is still on the hook for infringement of your patent.
So why would Eidos go fishing? Answer: It wouldn't. The only party with an interest in enforcing the GNU license violation is the FSF - which doesn't have the resources to do so.
and how does crippling the software with DRM or patents...
Software is "crippled" by patents? How so?
If you mean "software" as in a software product: Patents don't cover the patentee's particular software product. You're thinking of copyright. When someone misuses a company's software, e.g. off-license distribution (I refuse to use the stupid term "piracy"), the company usually asserts its copyright. If you want to rant about copyright, that's a whole different issue.
If you mean "software" as in the inventive software concept: Most companies that patent software choose this as an alternative to a much older tactic - trade secrets. And while software binaries can be reverse engineered, most companies prohibit reverse engineering in the license agreement. So it's still illegal, just in a context not involving patents.
So what's the net difference between a software patent and a trade-secret/reverse-engineering-prohibiting-licen se arrangement? The primary difference is that a software patentee discloses his entire invention to the public, rather than keeps it secret forever. If either option "cripples" software, it's the latter.
I started out with a Garmin eTrex Legend (cheap $125, durable and simple) and just recently upgraded to a Garmin GPS Map 60CS. I absolutely love this thing!!
Just a comment about this:
Higher-end devices are often more accurate, more detailed, more full-featured, and equipped with brighter, more vibrant displays. They also have a drawback: You won't want to use them for geocaching. While geocaching, you will climb up and down steep embankments, pick through dirty areas, and ford streams. Are you comfortable risking damage to a $100 device? How about a $500 device?
(Also note: That bright, color display will eat up batteries fast. I've had the misfortune of being near a geocache and having my GPS device die. Not fun.)
What you need for geocaching are reasonably accurate lat/long coordinates and a compass. Most bottom-of-the-line devices do both fine. The rest is costly window dressing and a distraction.
The new licence does not say that you cannot charge for your work. It is designed to be incompatible with Patents and DRM. This is distinct to copyright.
The GPL is rather silly in this regard. Under Section 1, you may charge a fee for sending your program derived from GPLed code to a third party. But consider the combined effect of these clauses:
Section 2 requires mandatory and free licensing of that derivative.
Section 3 requires you to let the recipient distribute your derivative product at will under the GPL.
Section 6 bars you from restricting the recipients' rights to anything narrower than the GPL.
Thus - you may charge a fee for sending the program to anyone else, but you can't stop them from distributing it for free to anyone else. Even worse, they can resell your software and give you nothing.
This is not how software businesses survive. And this does not even approximate the ordinary workings of copyright.
My guess is that the clause will be triggered by any company attempting to sue over infringing GPLed (or possibly other free) software.
Patent assertion rarely involves lawsuits. Because patent litigation is hideously expensive, so sending a shake-down letter to an accused infringer often prompts a settlement. It need not even be a "we own all your code, pay us all your money" style letter. Many are simply worded: "please review your products in light of U.S. patent no. 6,543,210." Is merely referencing a patent sufficient?
Foregoing that - what if a competitor files for declaratory judgment that your patents are invalid or not infringed by its software? Does your company violate the GPL by defending the validity and potential infringement of its patents? If so, might a competitor file this action for the express purpose of causing this result?
How about countersuing? Does that trigger an en masse revocation of the GPL for your company? Aren't some Linux companies asserting some software patents in an attempt to ward off SCO?
And how about asking a court for a temporary restraining order (a/k/a preliminary injunction)? That's an ex parte determination, so it's not really a lawsuit. Same with a request for reexamination of a patent by the U.S. Patent & Trademark Office.
Etc.
Law is complex. You can't finesse a solution to a complex problem with an elephant gun. Richard Stallman should stick to software, and leave the nasty business law stuff to the (potentially nasty) business lawyers.
You don't appear to be annoyed. You appear to be a techno-snob.
I posit that pastime names are like trademarks: priority is won by predominant use, not the earliest use. The predominant use of "geocaching" is for hiding and finding kits of stuff via latitude/longitude coordinates - not a technique in a niche techology. This predominance is clearly indicated by Google, by wikipedia, and by DNS entries for geocaching.
Given these circumstances, your response does not constitute an interesting riposte, but merely portrays you in an unfavorable light.
This means that people who were using older GPL'd software are free from obligations of the upcoming license.
Correct. Section 9 of the GPL reads:
9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.
Each versio nis given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by hte Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.
So, this threat only has teeth if the latest version of the GPL does away with this retroactivity. It would have to allow users to elect only a version including the "patent punishment" clause.
But that's not the biggest issue. Here's the biggest issue: Does the "patent punishment" clause trigger if the company patents (A) software related to the GPLed software, or (B) any kind of software?
Either option has issues:
(A) If the clause only triggers for software derived from the GPLed code, then that's fair and straightforward. It's also completely redundant with Section 6 of the current GPL:
6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein.
Thus, this option is non-newsworthy.
(B) If it's a threat to punish companies that seek any software patent, then this raises a host of untenable issues. Obvious problems:
It's really difficult to characterize "software patents." Some patents clearly claim a computer-implemented algorithm; most don't (in part, because the USPTO was rejecting such claimed inventions out of hand as "nonstatutory" until about 1998.) The difference between a "process" and a "computer-implemented process" is infinitely gray.
Virtually every large software company has patents on software. The world's largest software patentee, by far, is IBM - a strong Linux advocate. Patents are acquired for a variety of reasons, including defensive purposes. The critical issue isn't whether or not a company patents software, but what it does with those software patents. Simply punishing all "software patentees" is an untenable oversimplification.
I recognize that the Slashdot view of software patents is that Microsoft wants them in order to protect its fat, bloated monopoly. That's not untrue. But another beneficiary of software patents - arguably the most important one - is the small tech startup. And since cash-starved startups are more likely to use GPLed software than established firms, this penalty clause disproportionately impacts startups. In effect, this would discourage innovation and promote tech stagnation.
This penalty clause would be a huge blow to the free software movement. Remember when BillG began touting the GPL as a "virus" that threatened to eviscerate companies' IP? This would lend much credence to that argument, and would send tech companies running away in droves.
Thus, this option is extremely problematic. It probably poses a much greater threat to small companies, and GPLed software itself, than to the stereotypical "bad guys" in the software patent biz. You can almost hear the industry fat cats rubbing their hands in glee, muttering, "Yes, please go ahead and kill the concept of GPL."
The point is that certification has become dissociated from the actual abilities, and hence, the evaluation becomes useless.
It's not nearly as dissociated as the OP suggested. You can't pass an XML web services exam unless you know something about XML web services. A passing score doesn't guarantee expertise, but it does indicate a minimum level of competence.
Generally: Certifications are hoops. They have the same purpose as any other career hoop: jumping through them demonstrates that you're more motivated than your competitors who don't jump through them. As a job candidate, you can consider it a safety measure: you might get rejected for not having it, but you probably won't get rejected for having it.
IT has become too broad to be evaluated in a single test.
It's rarely a single test. Exams are pretty specific to a particular topic: web services, domain administration, configuring BizTalk server, etc. Even the A+ cert, which is pretty superficial, has one part for "hardware" and one for "software."
It's not hard to understand, and if [infringement] were the only possible use for the software, I'd say "yep, court's right. Don't let them distribute it". Since that's not the only possible (or active) use for it, (and here I go sounding like a broken record), the illegal activity is the inducement, not the distribution of the software, in my personal view.
Dunno if you're still pressing this argument (Slashdot conversations typically die within nanoseconds of the post slipping off the end of the page), but in case you are:
You're arguing that the inducement is the criminal act, not the distribution. I counter that the inducement is a lesser included offense. I'll give you three compare/contrast scenarios that point this out - and bear with me until you read this through:
If you solicit a prostitute, but she turns you down, you are guilty of soliciting a prostitute.
If you solicit a prostitute, and she accepts, then you are guilty of engaging in prostitution. You're also guilty of solicitation the prostitute, but it's absorbed into the greater offense of engaging in prostitution (and the justice system can't punish you separately for both crimes.)
If you unwittingly have sex with a prostitute - if you just think you're picking up a regular unpaid whore, whereas she thinks you're soliciting her services and will pay her afterwards - you actually haven't engaged in prostitution. And you haven't solicited one, either.
Here's my point: As situation #3 shows, the intent is clearly necessary for any crime to be committed. But if you have the intent and succeed in committing the crime, the intent part (albeit necessary) is not the primary crime you've committed.
Similarly, Grokster. It is guilty of soliciting copyright infringement. But more importantly, it is guilty of contributory infringement, by distributing a product with the advertised intent that the recipients infringe.
You're doing the same thing I was complaining about--requiring an answer at one extreme or the other.
I don't need a "yes" or "no" answer - a one-word answer would have improperly criminalized or exculpated a number of situations.
A sensible set of guidelines would have been most appreciated. The courts are good at that: they could have cited the primary factors in the analysis, or set up a balancing test, or at least provided pro and con examples of infringement.
They did no such thing. They simply stated, "Grokster is culpable for this specific reason (and a couple of secondary ones.)" That's why I'm underwhelmed.
I don't have time to respond in detail, and I don't like responding superficially, but I do want to acknowledge that your points are fair and insightful. Thanks for the interesting response.
They were asked the question, "should Grokster be liable for promoting copyright-infringing acts by their users". The question, then, isn't "Is Grokster liable for copyright-infringing acts", nor is it "Is P2P software promoting copyright-infringement". The question, more specifically, is:
"Did the makers of Grokster actively promote the copyright infringement uses of it's software, and - if it did - should it be liable to the infringements that occur".
Isn't that question worthless, outside of this single MGM v. Grokster lawsuit and any follow-up litigation by MGM against Grokster? Wouldn't the question of Grokster's individual, specific liability be completely irrelevant against Kazaa, or LimeWire, or any other company? If the Court is ruling strictly in the context of Grokster, wouldn't MGM have to re-litigate the issue almost de novo, potentially before the Supreme Court again, against Kazaa or LimeWire?
The Supreme Court is not supposed to dispose of individual cases. Frankly, the question of Grokster's individual liability is not worthy of the attention of the Supreme Court. It is supposed to create precedent that allows lower courts to dispose of the whole class of similar cases. It violated this concept in deciding Bush v. Gore, and continues to do so.
Slashdot Mirror
I said people, as in everyone.
You're trying to have it both ways. You're stating that patents are a threat to an individual because they are a threat to a group of people working for a company. Then, you argue that patents should not be a threat to companies, because they can be a single individual working on his own.
It's pretty simple:
- Software created for private use, either by an individual or by a company, is effectively untouchable by patents.
- Software created as a product and sold for profit is, and should be, touchable by patents from other inventors.
You can argue these issues individually, but you're trying to conflate them. You're combining the drama of the poor individual coder oppressed by patents with the visibility of a company offering a commercial product. Pick one.However, if nothing is done to head it off, we will see patent attacks on Linux soon, as one example of patent attacks on open-source software.
Ah, the looming spectre of the Linux patent assault. Why do you suppose this hasn't occurred yet, despite chicken-little prophesies by minor Linux advocates for at least five years? Why does Linus Torvalds not seem worried?
Answer: It would be a terrible business decision. IBM needs Linux as a free design platform. Microsoft needs Linux to dodge accusations of monopolization, and attacking it would damage the its reputation (which it's striving to salvage.) In short, no one who owns such an arsenal can derive a business benefit by attacking Linux.
You've got to face facts here. Patents are not weapons, threats, or land grabs. They are business tools. Businesses acquire and use them solely for business reasons. Any outcry over perceived use in a non-business sense is out of sync with reality.
Hardly - the reproducibility of software is key to the fact that it can be created by people for themselves.
You continue to miss the point of the analogy. The point is freedom to expand into new uses. This has nothing to do with physicality, or reproducibility. You're trying to avoid the analogy on an unrelated basis.
(And you're obviously not reading closely at all; I've made no reference to the windshield analogy raised elsewhere in these post comments. If you're not willing to read before responding, then I'm going to stop responding to you.)
This bears no resemblence to the software world I know. All the commercial projects I have worked on continued to release updates. Far more than free software, the commercial model depends on continuing to sell newer versions of their software, or else periodic licencing fees, or else both.
What are you talking about?
What software are you buying that requires mandatory upgrades or periodic licenses? Aside from the two niches I mentioned before - antivirus software and mapping software - not a single software product I own requires paid-for upgrades. Almost all of them offer patches for quite some time to fix bugs, and even add new features, for free. And I own a lot of software.
As to the cusomising - my whole point is that we have now reached the point where people do it themselves.
That would be a disaster for the many free software companies. Since they can make very little money on the software product under the GPL, the service aspect is their primary source of income. If we've reached the point where users can customize their own software, then you can kiss many of the free Linux distros goodbye - their business models are in tatters.
- David Stein
What do you mean, "no longer?" I remember the software market of 10 or 20 years ago as very heavily populated by a whole lot of small companies. Even Microsoft was a niche player in the OS market, and just starting to branch into applications. If you equate software patents with big companies, then you should be happy to see these trends heading in the same direction.
All the big companies are doing with their patent efforts are trying to force people to depend on them for things they can produce themself.
Show me an instance where a patent was asserted against a person. Go ahead - I'll wait.
Patents are business tools that companies use against each other. What you do for your own purposes with a copy of Visual Studio is largely irrelevant to patentees, unless you turn the result into some kind of product. Why would they waste their time and money stopping you from doing something in private that doesn't impact their business?
All this talk of cars is meaningless and misleading. If a few hundred engineers get together to build one, then only one of them can drive away, but a few hundred get together and write an OS, or a web-browser, and everyone gets a copy.
The purpose of my analogy had nothing to do with the physicality of the auto vs. the intangibility of the software. That is irrelevant.
My point was that you can choose between two product models. One product costs more up front, but does everything you might want to do with it (within reason.) The other is free, but every time you want to adapt it to a new use, you have to pay someone to customize it for you. I prefer the former model; the ongoing need for more service is a debilitating hassle.
So what if we have reached the stage where people can produce sophisticated software as a pasttime?
Hey, I do that now. We're already there. And guess what - I can do that as much as I want, for my own purposes, without getting sued for patent infringement.
- David Stein
Then you have a big problem. In fact, it's intractible.
Let's say you're the USPTO, and someone applies for a patent for a new process of drug identification. It involves finding proteins that inhibit a target protein and stop cancer. Presuming it's novel and nonobvious, do you allow it?
Now here's the crux of the problem: The patent validity and enforceability have nothing to do with how the applicant intends to carry out the process. Maybe he'll carry it out in a test tube, or maybe as a simulation. Maybe his competitors will carry it out in a test tube, or as a simulation. What matters is whether someone is performing the process, which is the invention.
I've had this thrown at me before, and it's silly. No, I haven't studied the grammar of US patents, but I'm a computer programmer and patents read exactly like programs.
What? Really? I've never seen a program that reads like this:
"What is claimed is: A medical diagnostic x-ray system, comprising: an x-ray generating device operative to produce medical diagnostic x-rays and thermal energy; a hydrogenated polyaromatic fluid operative to circulate about said device to absorb at least a portion of said thermal energy and electrically insulate said device, said fluid comprising a hydrogenated compound that upon exposure to said x-rays forms an unsaturated hydrocarbon and hydrogen atoms; and an effective amount of a hydrogenation catalyst for interaction with said fluid, said catalyst reactable with said hydrogen atoms and said unsaturated hydrocarbon to recombine said hydrogen atoms with said unsaturated hydrocarbon."
Seriously, patents and code couldn't be more different. A program contains the actual instructions, in precise detail and chronological order, for carrying out an algorithm. A patent broadly describes an invention, focusing on the key pieces/advantages and omitting much detail.
Maybe an analogy would work for you -
Computer code : Patent :: Recipe : Photograph of Food
- David Stein
I'm arguing for it. I believe in promoting the interests of customers over the interests of business. Moreover, I believe that businesses often work best when they also promote the interests of customers, especially compared with an alternative business model that costs customers more and gives them less.
For instance, I hate the idea of DRM - I hate it to its core. It's a movement to sell people music that's less useful than music on a CD, and probably for more money (additional hardware, re-purchases, etc.) I froth with irritation when companies pitch this inanity to customers as "trusted computing" or similar nonsense. It's terrible business - it alienates customers, and probably won't work - and I suspect that the RIAA will regret it in the long run.
I feel similarly about continuous-payment software models. Hope that's obvious in retrospect.
Of course, those industries would want ways to make money by creating something once, then getting paid for it over and over. If the market weren't encumbered by IP laws...
What in the world do IP laws have to do with rental business models? You're lashing the spectre of technology stagnation to your disfavor of software patents with little more than rhetoric and chewing gum.
Even software has to evolve to adapt to changes in circumstance.
A few software products, sure. I totally get why Symantec sells subscriptions to Norton Antivirus: it's expected to respond immediately to every new virus. Same thing with mapping software, since the maps need frequent updating. Those things make sense, and so their customers are grateful to pay for them.
But very little software fits that need. Look at the leading software apps - Windows, Office, Photoshop, file compression, email and instant messaging clients, web browsers, media players. I just want them to work, forever. The company is expected to patch product defects - not because I pay them to, but because their product shouldn't be defective. But I don't need updates; if I really want new features, I'll buy the next version.
You might be able to write a perfect, bulletproof program that satisfies every current customer need - but the customer WILL have new requirements, the platforms that the software needs to run on WILL change, the customer might have new projects (and would prefer to contract with you again if you did a good job the first time), etc.
Fine. So you go write a new version, and I'll think about buying it. If it's not worth it to me, I'll stick with the software that I bought and own. What's wrong with that model?
That'd be a lot easier & cheaper if they didn't have to worry about violating IP laws.
There you go again, with superfluous use of "IP laws" as some kind of talisman. And in this instance, you're completely wrong. What prevents companies from tweaking purchased? It's not IP law of any kind. It's the license agreement with the vendor - the private contract under which the software was sold. "IP laws" have nothing to do with it.
- David Stein
You need to read more patents. Patent practitioners go for breadth: if the inventor has invented a process, the patent usually claims a "process," regardless of how it will be carried out. Check out this claim from U.S. Patent No. 4,405,829:
A method for establishing cryptographic communications comprising the step of: encoding a digital message word signal M to a ciphertext word signal C, where M corresponds to a number representative of a message and 0.ltoreq.M.ltoreq.n-1 where n is a composite number of the form n=p.multidot.q where p and q are prime numbers, and where C is a number representative of an encoded form of message word M, wherein said encoding step comprises the step of: transforming said message word signal M to said ciphertext word signal C whereby C.ident.M.sup.e (mod n) where e is a number relatively prime to (p-1).multidot.(q-1).
Do you see the word "computer" in there? No, it's just a process. It happens to be the process of RSA encryption and decryption, which, in virtually every instance, is carried out by a computer processor. But on the off-chance that you want to decrypt a message with a slide rule or an abacus, you would still have violated the patent (which is now expired.)
And don't try focusing on the mathematical nature of the process as a means of determining bad "software" vs. good "non-software" patents. You can easily embed this same process in a novel circuit, and patent that. So are you also opposed to patents for circuits?
The delineation gets even more gray when using high-level object-oriented programming. Such a "process" may be a control mechanism for an automobile assembly line, involving cars, paint, welding tools, etc. Does it matter if the process is a hardware controller, or if it's a purely sofware simulation? Should it?
Or, how about drug identification: should the process of finding a protein target for a cancer drug be patentably different if carried out in a test tube or a software simulation?
I've read patents. Some patents are simply a long list of circular definitions, and do not deserve to be patents because they don't refer to anything real or unique.
Sounds to me like you haven't studied patents enough to understand their style. Simple words like "means" and "comprising" and "for" have some special meanings that you wouldn't grasp from just reading patents. On the other hand, patents that are obfuscated - by intention or by incompetence - are not valid and not enforceable; the patentee has wasted its own time and money to get a worthless patent.
Unless you can explain how to patent a dictionary, a newspaper, a magazine or a novel...
You can't, under the "printed matter" doctrine. Computers are different: they don't just tell you how to do something; they can actually do it for you. Hence the distinction.
- David Stein
Never mind the total revenue - let me ask this: How often does the average user have to pay Microsoft to fix a problem with their OS?
Certainly there are companies that would like to pay to have their software heavily and exactingly customized, or to have extremely rare, highly case-specific bugs fixed immediately. It's good that those services are available. The question is how frequently those calls have to be made. If it's so often that it's your primary source of income, then your software is probably broken beyond repair.
Frankly, I can't believe we're debating this. Imagine a line of dirt-cheap automobiles that required mechanical adjustments every time you drove to a new city. If you plan to take many road trips to various unknown destinations, why would you ever put up with this hassle? Why not pay more for a car that goes everywhere for free?
Pet peeve time. Exceptions don't make rules. they weaken/break them.
The phrase means: "If you have to go to that extreme to find a counterexample, then that shows the strength of the rule in the typical case." This exchange was a perfect example of that. Consider our dialogue, condensed:
A: "Companies can't survive by selling GPLed software."
B: "Red Hat sells GPLed software, and they survive fine."
A: "Red Hat survives despite selling GPLed software - its distro is a large net loss. It socks up the losses by deriving income from an ancillary service. Your example just strengthens my point."
Sounds a lot like 'But if your business plan involves selling Microsoft-dependent software,(and you are not Microsoft) you'd better keep your resume close at hand."
Well, yeah, because your business model is illegal and you're going to jail.
If you are Microsoft, though, you do very well selling Microsoft software. Similarly, if you are Adobe, then you do well selling Photoshop. Show me a company that has made enterprise-scale profits by selling GPLed software - not an ancillary service, but the actual software.
AFAIK, MySQL is doing just fine as well.
1) They do well because they sell ancillary support. In fact, a whole lot of other companies also profit from MySQL support services.
2) MySQL does well because it doesn't have any competitors in its niche: a small, lightweight, scalable DMBS. It succeeds for the same reason WinZip succeeds: no one else really wants its space at present.
That is going to change. Microsoft's future OSes will increasingly have native DBMS capabilities. It starts with WinFS, a file system based on a relational database, and ADO.NET, which provides (clumsy) format-independent data access. Let's return to this conversation in, let's say, 2008.
- David Stein
No, see, I completely understand why a software company would want that model to succeed. I similarly understand why the media industry wants DRM.
In both cases, I can't see why customers would want it. That creates a ripe opportunity for a competitor to give them a more favorable one. Examples: DivX died a painful but much-deserved death to DVD circa 1999; long-term cellphone contracts had their market shares stolen by no-contract businesses.
Does a plumber charge you every time you use something they fixed? ... Why should programmers be treated any differently?
Pipes carry water. We expect them to rust, seals to fail, toilets to clog. Physical apparatuses experience wear and tear - it's unavoidable. Customers accept that.
Software has no such inherent limitation. If your company thrives on having people pay for software fixes, then face it - your software sucks.
Somebody wants software created or modified, software developer does the work, gets paid, goes on to do the next job.
So instead of buying software that works as you expect right out of the box, you get some raw materials for free, and then you have to spend time and money paying someone to make it work for you? Is that how it works?
- David Stein
That's an oft-cited model. It's also baffling to me.
A software product that requires you to call Technical Support, even once, has failed you as a customer. It needs better documentation, or a user interface overhaul, or a security or stability upgrade, or a more coordinated online community. (More likely, a combination of these factors.) Taken the next step: A software product that requires you to keep paid company support permanently on hand has cataclysmic failed you as a customer.
More simply put: Well-designed software, de facto, does not require you to pay someone else to help you use it.
Red Hat seems to get away with it because we expect OSes to be flaky and complex and in need of support. We see it as the open source equivalent of a patching service. We don't see this as an asset; we see it as an annoyance that's tolerable because the TCO is still better than Microsoft's. That doesn't mean it's a good thing.
Here's what I just don't get about advocates of this concept. What do you think of the "software rental model," where you don't own the software, but you have a limited-term lease and have to keep paying? Bogus, right? Why? Because we inherently feel that software sales should be a one-shot deal: you pay the price, and you use it indefinitely.
If you agree with this, then what's the difference between paying continuously for use, and paying continuously for access to a user manual? Is it just the fact that one initiative is sponsored by a company we like, and the other by a company we hate? Is that remotely logical?
- David Stein
Red Hat's primary product is not its Linux distro. That may be its most visible feature to most consumers, but that's not really its business. The distro is essentially advertising for its core product, which is technical support.
Google operates the same way: the search engine is its most visible facet, but its core business is selling sponsored links. In this way, it's backwards: its search engine is advertising for its core product, which is an advertising conduit.
But, in fact, this is the exception that proves the rule. The rule is: GPL-derived software is not commercially viable as a product. You may be able to exploit it, e.g., as leverage for a related service or a related (non-GPLed) product. But if your business plan involves selling GPL-dependent software, you'd better keep your resume close at hand.
- David Stein
That would be a disaster for IBM and any other GPL-relying company.
Why has IBM built up a massive patent arsenal but rarely used it? It's a defensive measure. If anyone sues IBM for patent infringement, it can threaten to counter-assert its 40,456 patents.
Now imagine if IBM had to grant that attacking company free licenses under all software (and patents) developed using GPLed code. Doesn't that vastly undermine IBM's defensive patent strategy? Broadly extrapolating: Doesn't this render any company utilizing GPLed software a magnet for infringement lawsuits from non-GPL-reliant companies?
- David Stein
May those with mod points reward you. This is an extraordinarily interesting point, even if off-topic.
You seem like the right person to answer this question: Is there a term for the kind of corruption we have today - corporations fully co-opting government?
On the one hand, I suspect there is, since this isn't a new concept. Even Adam Smith was terrified that laissez-faire economics would cause corporations to amass enough money to do so.
On the other hand, maybe not. I think that such amazing power to control the masses via advertising/savvy media skills is a recent invention.
If no term is available, can we coin one? Foxism, maybe?
- David Stein
Well, to be clear - it primarily protects users who are also developers. The defining trait of the GPL is that your users may modify your code and pass along the modification on similar terms. If you just want to give rights to non-developer users, then much more restrictive licenses are just as good.
- David Stein
That raises a very interesting question: Who enforces this rule?
Let me illustrate a typical case where someone would want to trigger this rule. Let's say you're Microsoft. You take some GPLed code - e.g., an algorithm for converting binary-coded-decimal numbers to pure binary representations (see Gottschalk v. Benson) - and then you patent an improvement to this process, and you embody the patented improved GPL code in a product, say, Clippy's Wild Advntures.
A competitor takes your improved BCD idea and embeds it in his product - let's call it's Eidos's new game, Tomb Raider: BCD Adventures. You sue for patent infringement; they want to raise this novel GPL defense against you.
Now, how would that play out?
The dispute over the patent does not involve your embodiment of the patented algorithm, i.e., Clippy's Wild Adventures. It involves your patent and Eidos's game. Eidos would have to go fishing through your products, including totally unrelated ones, for an embodiment of the patent. Thus, if Eidos asks you if you had an embodiment of the software, you could refuse to answer the question as irrelevant to the dispute over your patent and their product. And it can't just go reverse-engineering all of your products: in addition to be wildly expensive, this would probably violate their software licenses to each of the products.
Worse: Even if Eidos proved that you had violated the GPL, it doesn't help them. It merely creates a new, totally unrelated dispute between you and the Free Software Foundation. Eidos is still on the hook for infringement of your patent.
So why would Eidos go fishing? Answer: It wouldn't. The only party with an interest in enforcing the GNU license violation is the FSF - which doesn't have the resources to do so.
In other words: This is totally pointless.
- David Stein
Software is "crippled" by patents? How so?
If you mean "software" as in a software product: Patents don't cover the patentee's particular software product. You're thinking of copyright. When someone misuses a company's software, e.g. off-license distribution (I refuse to use the stupid term "piracy"), the company usually asserts its copyright. If you want to rant about copyright, that's a whole different issue.
If you mean "software" as in the inventive software concept: Most companies that patent software choose this as an alternative to a much older tactic - trade secrets. And while software binaries can be reverse engineered, most companies prohibit reverse engineering in the license agreement. So it's still illegal, just in a context not involving patents.
So what's the net difference between a software patent and a trade-secret/reverse-engineering-prohibiting-licen se arrangement? The primary difference is that a software patentee discloses his entire invention to the public, rather than keeps it secret forever. If either option "cripples" software, it's the latter.
- David Stein
You have an unfettered right to act as you wish. I was merely advising you how you appeared to some of us.
- David Stein
Just a comment about this:
Higher-end devices are often more accurate, more detailed, more full-featured, and equipped with brighter, more vibrant displays. They also have a drawback: You won't want to use them for geocaching. While geocaching, you will climb up and down steep embankments, pick through dirty areas, and ford streams. Are you comfortable risking damage to a $100 device? How about a $500 device?
(Also note: That bright, color display will eat up batteries fast. I've had the misfortune of being near a geocache and having my GPS device die. Not fun.)
What you need for geocaching are reasonably accurate lat/long coordinates and a compass. Most bottom-of-the-line devices do both fine. The rest is costly window dressing and a distraction.
- David Stein
The GPL is rather silly in this regard. Under Section 1, you may charge a fee for sending your program derived from GPLed code to a third party. But consider the combined effect of these clauses:
Thus - you may charge a fee for sending the program to anyone else, but you can't stop them from distributing it for free to anyone else. Even worse, they can resell your software and give you nothing.
This is not how software businesses survive. And this does not even approximate the ordinary workings of copyright.
- David Stein
Patent assertion rarely involves lawsuits. Because patent litigation is hideously expensive, so sending a shake-down letter to an accused infringer often prompts a settlement. It need not even be a "we own all your code, pay us all your money" style letter. Many are simply worded: "please review your products in light of U.S. patent no. 6,543,210." Is merely referencing a patent sufficient?
Foregoing that - what if a competitor files for declaratory judgment that your patents are invalid or not infringed by its software? Does your company violate the GPL by defending the validity and potential infringement of its patents? If so, might a competitor file this action for the express purpose of causing this result?
How about countersuing? Does that trigger an en masse revocation of the GPL for your company? Aren't some Linux companies asserting some software patents in an attempt to ward off SCO?
And how about asking a court for a temporary restraining order (a/k/a preliminary injunction)? That's an ex parte determination, so it's not really a lawsuit. Same with a request for reexamination of a patent by the U.S. Patent & Trademark Office.
Etc.
Law is complex. You can't finesse a solution to a complex problem with an elephant gun. Richard Stallman should stick to software, and leave the nasty business law stuff to the (potentially nasty) business lawyers.
- David Stein
I posit that pastime names are like trademarks: priority is won by predominant use, not the earliest use. The predominant use of "geocaching" is for hiding and finding kits of stuff via latitude/longitude coordinates - not a technique in a niche techology. This predominance is clearly indicated by Google, by wikipedia, and by DNS entries for geocaching.
Given these circumstances, your response does not constitute an interesting riposte, but merely portrays you in an unfavorable light.
- David Stein
Correct. Section 9 of the GPL reads:
9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.
Each versio nis given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by hte Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.
So, this threat only has teeth if the latest version of the GPL does away with this retroactivity. It would have to allow users to elect only a version including the "patent punishment" clause.
But that's not the biggest issue. Here's the biggest issue: Does the "patent punishment" clause trigger if the company patents (A) software related to the GPLed software, or (B) any kind of software?
Either option has issues:
(A) If the clause only triggers for software derived from the GPLed code, then that's fair and straightforward. It's also completely redundant with Section 6 of the current GPL:
6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein.
Thus, this option is non-newsworthy.
(B) If it's a threat to punish companies that seek any software patent, then this raises a host of untenable issues. Obvious problems:
Thus, this option is extremely problematic. It probably poses a much greater threat to small companies, and GPLed software itself, than to the stereotypical "bad guys" in the software patent biz. You can almost hear the industry fat cats rubbing their hands in glee, muttering, "Yes, please go ahead and kill the concept of GPL."
- David Stein
It's not nearly as dissociated as the OP suggested. You can't pass an XML web services exam unless you know something about XML web services. A passing score doesn't guarantee expertise, but it does indicate a minimum level of competence.
Generally: Certifications are hoops. They have the same purpose as any other career hoop: jumping through them demonstrates that you're more motivated than your competitors who don't jump through them. As a job candidate, you can consider it a safety measure: you might get rejected for not having it, but you probably won't get rejected for having it.
IT has become too broad to be evaluated in a single test.
It's rarely a single test. Exams are pretty specific to a particular topic: web services, domain administration, configuring BizTalk server, etc. Even the A+ cert, which is pretty superficial, has one part for "hardware" and one for "software."
- David Stein
Dunno if you're still pressing this argument (Slashdot conversations typically die within nanoseconds of the post slipping off the end of the page), but in case you are:
You're arguing that the inducement is the criminal act, not the distribution. I counter that the inducement is a lesser included offense. I'll give you three compare/contrast scenarios that point this out - and bear with me until you read this through:
If you solicit a prostitute, but she turns you down, you are guilty of soliciting a prostitute.
If you solicit a prostitute, and she accepts, then you are guilty of engaging in prostitution. You're also guilty of solicitation the prostitute, but it's absorbed into the greater offense of engaging in prostitution (and the justice system can't punish you separately for both crimes.)
If you unwittingly have sex with a prostitute - if you just think you're picking up a regular unpaid whore, whereas she thinks you're soliciting her services and will pay her afterwards - you actually haven't engaged in prostitution. And you haven't solicited one, either.
Here's my point: As situation #3 shows, the intent is clearly necessary for any crime to be committed. But if you have the intent and succeed in committing the crime, the intent part (albeit necessary) is not the primary crime you've committed.
Similarly, Grokster. It is guilty of soliciting copyright infringement. But more importantly, it is guilty of contributory infringement, by distributing a product with the advertised intent that the recipients infringe.
- David Stein
I don't need a "yes" or "no" answer - a one-word answer would have improperly criminalized or exculpated a number of situations.
A sensible set of guidelines would have been most appreciated. The courts are good at that: they could have cited the primary factors in the analysis, or set up a balancing test, or at least provided pro and con examples of infringement.
They did no such thing. They simply stated, "Grokster is culpable for this specific reason (and a couple of secondary ones.)" That's why I'm underwhelmed.
- David Stein
- David Stein
"Did the makers of Grokster actively promote the copyright infringement uses of it's software, and - if it did - should it be liable to the infringements that occur".
Isn't that question worthless, outside of this single MGM v. Grokster lawsuit and any follow-up litigation by MGM against Grokster? Wouldn't the question of Grokster's individual, specific liability be completely irrelevant against Kazaa, or LimeWire, or any other company? If the Court is ruling strictly in the context of Grokster, wouldn't MGM have to re-litigate the issue almost de novo, potentially before the Supreme Court again, against Kazaa or LimeWire?
The Supreme Court is not supposed to dispose of individual cases. Frankly, the question of Grokster's individual liability is not worthy of the attention of the Supreme Court. It is supposed to create precedent that allows lower courts to dispose of the whole class of similar cases. It violated this concept in deciding Bush v. Gore, and continues to do so.
- David Stein