Hell, the UK is the only place I ever heard of where those wrongfully imprisoned are then forced to reimburse the government for the cost of their imprisonment.
In Norway, Per Liland, wrongfully convicted for murder and jailed for 14 years, had the cost of living he would have had if not jailed, deducted from his compensation. (Until there was an uproar and the deduction was cancelled.)
The logic was that the compensation was compensation for loss due to imprisionment. Without imprisionment he would have had living expenses. The compensation as granted by the Parliament was for lost income. Now they adjusted for lost expenses too. The logic is flawless: it had two part, the amount he would presumably have earned in a job, and a compensation for reduced quality of life. But they failed to compute a compensation for the insult of doing such calculations.
David Chaum's original system, Secret-ballot Receipt, does not have completely centralized decryption. There is a number of trustees, each with a private decryption key. If e1, e2,..., en are the encryption functions, e1(e2(...en(ballot)...)) is sent to trustee 1, who decrypts and sends e2(...en(ballot)...) to trustee 2. Eventually trustee n receives en(ballot) and publishes ballot. But each trustee receives a batch of N encrypted ballots, and emits another batch of N ballots in a different order. Only the trustee itself knows which ballot in the emitted batch corresponds to which element of the received batch. Since the trustees can include political parties, human rights organizations, churches, etc, it is unlikely that all would collude to trace a decrypted ballot all the way back to the original encrypted ballot of which the voter has a receipt. There are additional mechanisms in place to prevent other kinds of abuses.
However, Chaum's original scheme was deemed to have usability problems for the voter (which I think are overblown), and its mathematical virtues are hard to explain to ordinary people beyond stating "it's sooo safe! Mathematicians say so!".
Homomorphic encryption has the property that e(n)+e(m)=e(n+m) for some meaning of = and +.
If 'n' and 'm' are answer vectors of 0 (no) and 1 (yes) answers to a vector of questions (Mickey Mouse for president? Donald Duck for president? etc), the sum m+n is the tally of those two ballots. Now I presume that with fully homomorphic encryption it is also possible to (e.g.) weed out overvotes and undervotes before adding up the results.
Even if it may be beyond most people how an encryption scheme can be homomorphic, it may be explainable to a somewhat larger section of the public how a homomorphic encryption can be used to tally the ballots publicly while maintaining the secrecy.
More about Chaum's scheme:
All inputs and outputs of all trustees are posted on a web site where anybody can download them, with some additional data that allows everyone to check 1) my encrypted ballot (as shown in my receipt) is there in the input to trustee 1, and 2) if a trustee has cheated with n ballots, anyone who runs the checking algorithm on the data will flag at least some of them, with probability (1 - 2^-n). Since usually more than 100 ballots must be stolen to steal an election, the chances of doing so undetected is less than one in 2^100. An detection will not just be "detection", it will be a mathematical fact that every math teacher in the world can verify (and many others). Additionally, there are similar guarantees that the encrypted data on the receipt really represents the votes that the voter saw and accepted in the booth.
If the ayatollahs had had such a system in place in Iran, they would probably have been able to convince the public that Ahmadinejad won - provided he did. So, even supposing that there was no fraud, the ayatollah's are to blame for the doubt and the protests. They could have avoided that making the elections more transparent.
The 2000 Florida fraud would still not have been prevented. More than 60000 voters were turned away because the shared names with alleged felons. All the recount discussions failed to consider them, discussing far smaller numbers of hanging chads and military absentee votes. No mathematical voting scheme could have prevented that. In Florida, those turned away should have been offered the opportunity to cast a provisional ballot that could have been counted after they had challenged their "felon" status.
If the aim of negotiations is a binding treaty, this is a back door to legislation without democratic control.
Has the general public ever had the use of reason? Secrecy is justified in two kinds cases. Protecting the privacy of individuals, and protecting military installations, weapon systems, response plans, etc. But none of these cases are of legislative nature. Why is it at all possible to pretend legislative or quasi-legislative preparations need secrecy, and get away with it? It is a blatant inconsistency in terms. It ought to completely discredit whoever forwards it. It should have elicited unanimous uproar from the congress...
I guess the poster has misunderstood. I also guess the text invites misunderstandings.
Usually, there are two parties to an agreement, and third parties are all others. But when an agreement between A and B is changed to an agreement between A and C, becase B sold its part to C, then there are suddenly three parties, until everyone forgets B ever was part of the deal.
To complicate matters even more, there are multiple agreements here, one between the user (A) and Verizon/Fairpoint (B/C), and another between the user and (say) Yahoo. With respect to the second deal, Verizon and Fairpoint are third parties. With respect to the first deal, Yahoo, MSN, etc are third parties.
NOW tell me what "third party web site" the users will no longer have access to, or will no longer access their emails through.
I guess the real meaning is that Verizon had a web portal in which Yahoo and some other webmail providers was integrated, so that users could have the impression that Verizon was doing an email service. Now Verizon is closing its site, and Fairpoint is taking over, providing a new portal with similar integrated access to yahoo etc.
Agreed, but for the record: The link has a typoo, try to make the last part ?lnk=http# instead of ?lnk=hptt#.
Of course that does not change the point, a faulty link on the front page is discouraging.
Nearest cell phone... Is there no routing? What prevents a message from being propagated all the way to Tierra del Fuego and back, and down there again, and...
If sombody has a pointer to a proper answer, I'd like to have a look.
Novelty and non-obviousness are distinct criteria for patentability. However, novelty is easier to define, and so people and examiners prefer to make the non-obviousness an automatic corollary to novelty. I suppose this is part of the problem.
Please fix the typography; it is 1.783 x (ten-to-the-power-of-minus-twenty-seven), or if it works this time, 1.783 x 10^-27. A small amount by everday measures. (But surprisingly much for a moron like me. How can a particle this heavy be responsible for endowing electrons and neutrinos with their respective masses, that are mch smaller?)
We have been raised to believe ideas seem obvious only in hindsight. That is not the case.
The reason nobody did this before is that the technologies were not mature, the hard disks were too small (or too expensive), etc, etc. Yet this patent is not about anything that enabled this.
This patent monopolizes the problem, not the solution. It describes a product, not its implementation.
Superficially it is about an improved way of providing time shifting, which allows simultaneous recording and playback at a reasonable price. However, everybody at the time knew that harddisks offered this kind of advantages. The patent says use MPEG (to reduce harddisk capacity need), and everybody knew that also, but it was very hard to do the compression sufficiently fast with cheap processors and cheap memory. However, TiVo has not invented faster CPUs, better compression schemes or anything enabling.
I guess that when TiVo started, to get the whole thing working with the parts available at the time and at a reasonable price _was_ a challenge, and probably required some investments. However, this patent discloses nothing but a sketch, a plan that somebody skilled in the art would have written down in ten minutes. The tricky part is not disclosed. However, a couple of years later, with even faster processors etc available off the shelf, then also the implementation of the sketch becomes increasingly something that anyone skilled in the art can do without further tips and without much research. This is ironical: The product plan as described in then patent was obvious at the time, what was not obvious was not disclosed in the patent. Yet the continued evolution has made also that obvious so that it appears like the patent did disclose enough. However, here is where the hindsight is.
We might perhaps want a new kind of protection (other than patents) for small firms that do heavy research at the point in time when an inherently obvious principle becomes possible because of developments in other fields (ram capacities, CPU frequencies, etc) but then that should be formulated and the conditions, limits to applicability, etc. should be worked out. The patent system had a different (but related) purpose, but has fallen into such abuse that it is mostly by accident that it now protects a company that once did some hard work, ironically, through a patent the company acquired from somebody else long after they did their work.
I used to work for a company that made computers with a version of Unix on them. We would mostly manage to get a fix out in a week, but I managed to reduce that to about two days, in seldom cases one day. I was much of the bottleneck - I did the quality assurance and the packaging. This was in the early 90s. I improved the process by creating something that resembles (remotely) a redhat package system, that made it a snap to revert patches if anything went wrong. The effect was, not foreseen by me, to streamline the process so that we could focus much more on the contents of the fix. I saw in the beginning that also the packaging process could introduce errors and mistakes. Packetizing also helped testing, making it easier to install versions of the software, or just patches, test it, and then revert the configuration to what was needed for that computer's ordinary mission.
Fasthost is pretending to pursue "Godfrey v. Demon Internet", which is highly dubious.
The case Godfrey v. Demon Internet is itself a case that reeks strongly of stupidity. In that case, an ISP refused to remove usenet postings falsely pretending to be from Mr. Godfrey. The ISP pretended they had no responsibility, because the defamation act does not apply to those who merely relay information like e.g., newspaper vendors. When this was struck down, the ISP found a batch of other messages with forged sender ids, pretending they were from Mr. Godfrey and alleged these messages were provoking. They wanted to have their responsibility reduced because of the alleged provocation. Through this step, they made the process much, much more costly, and in the end they had to pay for it all.
The ISP of course also claimed freedom of speech. They forgot to say who's speech. If the articles i question were actually from Mr. Godfrey, he has the the right to cancel them. The Usenet software has automated mechanisms he could have used. In order to claim freedom of speech for anybody else, they automatically were claiming the right to falsely write under the name of somebody else.
I think the legal systems in nearly all countries could be said to have several layers. In some countries it is pretended that every decision is a carefull but strict application of the statutes. However, any half-witted person can see that there is a great selectivity in how the statutes are applied and how the circumstances of a case are found to fit descriptions, or what labels apply to them. Therefore there is a second layer which could be said to consist of mostly "common sense" and "public interest", although this second layer is also often abused, perhaps out of stupidity, producing decisions that sometimes more than border to the mock trials.
However, a reasonable application of common sense and public interest should make it very clear that in absense of any editor, as the Usenet is designed to run entirely without any top-down control, then the human owners of the participating computers must be required to assist in limiting the damage. Society cannot, or at least should not, allow any important damage to persist just because of a technicality not foreseen by the lawmakers. I think that is why the judge made the ISP a "publisher" in the Godfree v Demon case.
This question should be entirely different in the present case, where there is a clearly identifiable and named responsible person behind the allegedly defamating articles. In the present case the articles are entirely under control of Craig Murray, and Fasthost's decision is interfering with his freedom of speech. I guess that if a storehouse chain decided not to carry The Guardian on a particular day because of potential liability for an alledgedly defamating article, the chain could be sued and would be likely to have to pay damages to The Guardian.
What remains disturbing is, that if Fasthost had decided to ask Usmanov's people to turn to Murray himself with their request, there is no guarantee that all judges would side with Fasthost. While I am damn sure that my analysis is correct (even I don't know the law), I am also damn sure that only a small majority of the judges have the brain to analyse the case correctly. (That is, a large minority does not.) Even the little I have seen quoted from the British defamation law, makes it appear like the law carries a list of circumstances when contributors to the publishing of a defamation are not to be held liable, while it fails to say explicitly why. It seems to happen over and over again, that because judges cannot be relied on to interpret overly general principles, the laws are written with lists of more specific nature, and the principle behind the list is not mentioned at all. I think this way of writing laws could be improved.
I have not read the actual text of the Godfrey v Demon decision, and I do not know if the decision makes it very clear why it found that the ISP is a publisher. In
And when it comes down to it, why didn't she merely turn off the monitor?
Likely because it did not occur to her that she could do that. That depends on mindsetting, and on mental flexibility, also called "intelligence". She probably thought that she needed to have the computer turned on to use it for whatever she had planned for the class, and struggled to control it so she could do just that.
What was the alleged crime, really? Was it that she had intentionally made the children see the stuff? Or was it that she had irresponsibly visited improper sites before the class, and not been able to control the resulting malware infection?
The first alternative sounds unlikely because so few would have suficiently strong motive. The second is much more credible, since so many are capable of being excited by porn. But in the second case, not having the idea to turn off the computer is not in itself a crime, and not evidence of what she did to have the malware infection occur.
Only sworn testimony and physical evidence is supposed to be considered by the jury.
The article says
Smith countered Horner's testimony with that of Norwich Police Detective Mark Lounsbury, a computer crimes investigator. On a projected image of the list of Web sites visited while Amero was working, Lounsbury pointed out several highlighted links.
I suppose Horner's testimony was a sworn one.
"You have to physically click on it to get to those sites," Smith said. "I think the evidence is overwhelming that she did intend to access those Web sites."
I can all to easily imagine a jury that fails to distinguish what they are supposed to believe, what the "computer crime expert" showed them, and what they are supposed to disregard, the prosecution then saying this is strong evidence.
Anyway, I suppose the jury is not instructed to think that the prosecution is a gang of asses that will gladly twist things and (mis)interpret things to fool the jury in insiduous ways. Neither are they instructed to see the defence in this way, but for many, if not most, this goes without saying. Criminals have always defended themselves as best they could. Still most people prefer to believe that dishonest policemen is exceptional outside the crim novels. I am absolutely convinced that my sister, for one, would believe a policeman if he "looks like an authority".
If the defense attorney wasn't utterly incompetent
How can we be sure the defense attorney is so utterly incompetent? The quotation from the article,
Computer expert W. Herbert Horner, testifying in Amero's defense, said he found spyware on the computer and an innocent hair styling Web site "that led to this pornographic loop that was out of control."
seems to indicate adequate competence, at least on the technical side. On the other hand, I think
"You have to physically click on it to get to those sites," Smith said. "I think the evidence is overwhelming that she did intend to access those Web sites."
indicates the prosecution is dishonestly making up "facts" and playing on their credibility as "voice of the authority", and a stupid jury or judge believes the accusation because "the authorities say so".
In court each side has a limited number of opportunities to speak. If the prosecution made this claim in a response to the defense witness, the defense probably (I don't know the actual protocol in this court) had limited opportunity to start a new debate, and find expert witnesses, etc. to counter this statement.
Consider the Linux distribution like a communcations network, and a mechanism is used to make a copy of a copyrighted work reach a select group of justified receivers. The blob is encrypted with the vendor ID, and the driver, besides, of course, not registering a device unless it supports it - and that is determined based on the vendor and model IDs - if it accepts to drive the device, decrypts the blob with the vendor ID of the device before loading it. 16-bit encryption with known key is not a big deal to crack, but for a non-intended receiver to do so manually would be an extraordinary step. It clearly goes beyond ordinary usage, and constitutes a kind of break-in. Since the driver and the blob are available on the net unencrypted, there is no point in actually having a strong encryption, but rather to have a reasonable token of directing the communication to the justified recipients, and not to everybody.
At some point copyright laws were amended so that "copies" that exist in wires and routers during transmission would not be seen as copies in the sense of copyright laws. It is almost unbelievable that anybody thought such amendments were needed. Perhaps they were, back then. Judges and lawmakers need time to get their heads around new fields. Some of them need, anyway. But in spite of impressions one sometimes get, laws are not usually interpreted so blindly and mechanically. Imagine, people could have been sued for opening a book in a room with a mirror!
Concepts in the laws, like "copying" and "distribution" must have a sensible reality, e.g of producing another copy in a way that has any real consequences, like somebody enjoying a copy by reading it, listening to it, etc, or at least having a possibility thereof. I do not pretend that judges always observe this requirement, but in principle it is required, and that is why you cannot be sued for opening a book in a room with a mirror. There is no exception for mirrors in the law.
Are we there today, that we can hope judges will accept that a binary blob in a linux distribution is not being "distributed" if the communications channel does not "unpack" it, or "deliver" it in some sense?
Still I feel that it would be risky to include it in a Linux distribution. If somebody sues, there is no guarantee that the judge will be reasonable. In the long run there is little doubt, pargmatism will creep in. In the short term, it is risky.
Off the top of my head: A garage door vendor claimed some years ago that a remote control vendor was infringing on the former's IP rights by selling devices that sent the specific codes to control the garage door. The judge said the garage door owners had paid for, and the garage door vendor had sold the permission to use that IP material. The garage door vendor was not entitled to require the door owners to use the door vendor's equipment to send those codes. Garage door vendors were not allowed to unilaterally withdraw their permission to use the codes after the permission was sold.
Similarly, owners of wireless devices have paid for, and the wireless device vendors have sold and received money for a permission to download the firmware to the device. I doubt the wireless vendors have any right to force users to use Windows or the Windows driver to send that firmware file to the device.
In many countries, specifically in Norway, where I live, there are laws to protect the right to do reverse engineering. Any license agreement not to do reverse engineering is invalid here. The laws say explicitly that these provisions cannot be overruled by agreeements between the parties. It should be possible to extract the firware from the driver in a country that allows this.
I strongly doubt that distributing the firmware code to owners of the corresponding devices can properly be seen as distribution against the permission of the copyright holder. It may even not be considered to be "distribution" in the sense of the IP laws. While companies may try to convince judges about anything they see convenient, especially if they have secret arrangements with a monopolistic software vendor, it is not too likely that the courts will believe them.
I am more uncertain about the inclusion of the firmware blob in e.g. Linux distributions. Instead linux distributions could be equipped with software that automatically searches the net for the binary blob. Such software would only work if the computer has another way of getting connected. On the other hand, the software could be made to do this under Windows too.
Here I am talking about the firmware, not the driver. As more functionality is moved into the driver, it probably becomes more difficult to reverse engineer the device driver even if the firmware blob can be successfully loaded into the device. Still it should be doable, while reverse engineering the firmware is next to impossible.
Do Autodesk products refuse to work with files not containing the TrustedDWG codes? If they do, the Open Design Alliance has a good reason to put the code in the files.
All the TrustedDWG thing does is basically tell the person opening a dwg in autocad that the file was last saved by a trusted dwg product.
The above statement seems to contradict any such interoperablility issue. Is the statement misleading, or are all the others who talk about this aspect completely off the mark?
By the logic of this patent, I should be able to patent nuclear plant refurbishing wearing pink boots. While most likely somebody has worn pink boots while working on some nuclear plant contsruction or repair or refurbishing project, I bet it has not been described in the litterature.
Unfortunately, the competition will get around it by wearing brown or black boots... Perhaps I can patent that too?
But the interesting thing is that I don't need to know anything about nuclear physics to patent it. I just patent refurbishing, said refurbishing comprising putting the boots on to protect from nuclear spill, enter the premises showing badges, doing the modification, cleaning up the mess, leaving the premises, and disposing of the boots as hazardous waste.
Never mind that patent applications should describe the invention so that a person with ordinary skills in the art can complete it by following the instructions in the patent description. Anyone who has ordinary skills in the art of nuclear plant refurbishing already knows how to do that, so why must I describe it?
Just look at this patent.
2. method for users to create and maintain a rich-media application on said host website via the Internet comprising:
creating a user account; accessing a user account; and viewing available options for creating rich-media applications,
wherein said accessing a user account comprises one or more of the following: accessing account information; creating a new rich-media application; modifying an existing rich-media application; and accessing statistics from an existing rich-media application;
There is not a single word about how to create this program or computer system that allows the user to create a user account, etc. Why bother? Anyone having ordinary skill in the art of computer programming already knows how to do that.
But the law also has a requirement that the invention be non-obvious. What happened to that? Well, they found it so difficult to determine reliably if something is obvious, that the reformulated it to mean, if you have to combine the teachings from multiple pieces of litterature, then the combination is non-obvious even if each part has been described in the trade litterature.
In other words, let's go ahead and patent things, until nuclear plants are finally forced to close. One patent for the sake of the environment every day!
We should rather "DDOS" the politicians who let us down. For each spam you receive, send a copy to each member of the congress/parliament asking them to provide you with a legal way of defending yourself against this, or to enforce any solution they have already legislated.
If a politician acts credibly against spammers, delete him from your list.
In a country with 25 million voters, if 10 percent do this, and they receive on average 20 spams per day, this is 50 million mails to each politician.
Bind code to do this to the "spam" button on your email client. Then every mail will have been sent manually and for a good reason. There is no need to coordinate the timing, the point is not to bring down the MEPs servers but to simply remind them that the spam receivers are voters too.
I think I have been opposed to software patents all the time. But I do not think all the arguments are as convincing as "we" believe. Especially, I am not certain that software is that fundamentally different from many other fields.
Why are digital watches fundamentally different from analog ones? Or are they not since both deal with a "force of nature", time?
If we for the time being skip "computer-implemented inventions" like cd/dvd music/video players that could be analog or digital, and instead consider the computer as an information processing device, a communication device, etc, what then about a pair of glasses? Is that patentable? You use them to read! What about a desk calculator? A rotary press? A photocopier?
I guess you will rather compare all these examples to the physical aspects and workins of a CRT or LCD monitor, a hard disk, etc, which most anti-software-patent people think should be patentable.
Before the advent of the modern transistors and microchips, there was a branch of the "computing" industry that created "analog computers". In an analog computer, a capacitor could be used to integrate a function represented as an analog current. They had a rich library of devices that could represent various mathematical operations. What do you think? Shoud none of that be patentable? Do they not express human ideas?
I absolutely think we should do everything possible to stop software patents, and you have mentioned a couple of reasons that I do not doubt at all: They are not necessary, and they do a lot of harm, they prevent rather than further innovation, they increase cost for everybody, but the mayor corporations still want them because they know the smaller companies will feel the burden much, much harder, since they must pay those expenses from incomes on fewer sold units.
I do not think the anti-software-patent movement should stop saying all this, even there are reasons why it takes time before such statements sink in and are believed. (Some of that time has already passed, and quite a few have been convinced.) Yet if we give more focus to the obviousness problem, then many of the other statements become easier to understand.
Why are we not all equally against patents on clever ways of making aluminium car bumpers? Such patents restrict the competition. They increase cost for everybody and the smaller companies feel it harder. Large metalurgical and chemical conglomerates acquire patent portfolios and cross license them. That makes it hard for smaller companies to enter the field.
There is a logic behind the patent system, and this logic is not consistent with the realities. The logic is that
invention is hard and expensive and risky
ripoff is easy after the invention has been published or the product is on the marked and can be studied.
Inventions are ideas that the society would not have had, had the inventors not published them. Therefore it does not cost society much to give patents. Only occasionally are some inventions made independently by multiple inventors. Giving a monopoly for ca. 20 year only costs something the last approx. 10-5 years, that is when less gifted people would have found them independently.
Those few times patents prevent genuinely independent developments are a price worth paying for the benefits of a healthy network of scientists and inventors that are able to live off their creations.
Those few times when inventors would have published the invention even in absence of patents, the society pays a monopoly premium and those who would have ripped off the invention loose business. But society should give moral priority to those who made it all possible, and not defend the rippers.
For most people this logic is about as certain as that the world is round, or they believe it as firmly as many believed the earth is flat in the middle ages. Tell them that it is all wrong, and they w
I fully agree. The problem is that people, including the legislators, believe software embodies "working principles", i.e., methods, that need protection not provided by copyright.
But surely that's an easy one to disprove. Microsoft have forged a multi billion dollar company, with all their revenues coming from copyright. There is no need of software patents, else Mighty Microsoft would be a hole-in-the-wall operation and IBM still unmoved from its heyday.
Easy? How?
Politicians are quite used to all sorts of lobby and pressure groups using all sorts of emotional, or untrue, or half true, yet sometimes quite persuasive, arguments. How do you convince them that Microsoft's wealth is not dependent on software patents? Yes, I know the statistics, the late pickup of patenting from Microsoft. But Microsoft says otherwise. They could say that things are different now because the field is more mature, or any other b******t.
I briefly considered if we could show that virtually no patent discribes the innards of programs. All patents only describe what programs do, i.e., the business idea behind them, or the problem that they solve. Show me a single patent that addresses the innards of the programs! Then I remebered that Microsoft recently got a patent on modeling something as an object, I just can't remember what it was... a mail address or contact information? They will pretend it is a very valuable idea, that there is a lot of research behind it, and they need patent protection to prevent others from exploiting their research.
Would it be possible to say that computer programming is more of a craft than an advanced science? Why is copyright enough protection?
I think the answer to that question is because the difficulty in computer programming is to get all the details to fit together. If someone wants to create an independent implementation of a program, even if he uses the same broad "methods" as the original, the cost is about the same. There is no substantial saving from copying the methods from the earlier implementation.
What statements are likely to gain almost-universal support, enough to convince politicians that are not themselves programmers?
My last statement above is not 100% true, because the process of constructing the requirements and working through the use cases of a large system is not always a small part of the development cost. Still it is true to a high enough degree to give investors a reasonable protection. If a market is large, there will be competitors attacking it, with considerable startup costs and considerable first-mover advantage to the initial investor. If the market is small there will not be very many competitors for a long time, because of the risks involved.
heh... picture the scene
Do you realise ma'am, the if little Johnny here were to design a jet turbine engine, spend 200 million dollars to set up a production line, and then market it internationally, he could be sued by Boeing and Lockheed and for No Good Reason!?!"
It seems we are approaching (yet another) significant difference between software and other fields: Software enterprises can be micro-scale, with zero or near-zero startup costs.
Or: Software production attracts kids.
Or... What is really the essence here? I am not kidding, I think there is something definite to be said here that you can convince any reasonable person about, and that has a bearing on the software patents issue. Something that does not so easily get dismissed as "just a pressgroup litany or religion". But I am not yet sure what that is.
On the other hand, little Johnny programmer is not that likely to get sued, because it costs mony to sue and the patent holders are not that likely to see any benefit in suing him. And people are surprisingly willing to dismiss principles on grounds of pragmatism, meaning they do no
If we are talking about an idea that is truly non-obvious, it does not help to formally own it. We could need to convince the bright-heads to publish it.
Is that the old "without due recompense, researchers will not share their ideas with the world" argument? If so, it doesn't apply to software patents, since the copyright mechanism is more than adequate for recompense.
I fully agree. The problem is that people, including the legislators, believe software embodies "working principles", i.e., methods, that need protection not provided by copyright.
This is one of the points where I feel we have failed to analyze the situation correctly, and therefore we have failied to address an important underlying misconception.
our efforts would be better focussed upon the legislators
The legislators are not stupid, but they do share all the popular misconceptions, at least until we address them properly.
tell them how little Johnny, if he shares that computer program he wrote and of which he is so proud, is going to get sued if he shares it with his friends. Tell them how if writes something useful and publishes it on the internet, someone is going to sue his ass off in case he damages their profits.
Because the patented subject matter is so perfectly obvious, any little Johnny that strays near the subject will hit upon the same idea.
Had patents been restricted to truly non-obvious inventions, the danger would have been much, much smaller. This is not to say the danger would have been non-existent, among other things, because as time goes by, the once novel and non-obvious idea becomes common good. This is one reason why software patents are evil beyond the evil of obvious patents.
But notice how this argument could be applied to any field. What makes software special? You say because software is expression of ideas, but so are watches and kitchen appliances. What matters here is that little Johnny is unlikely to engage in making watches or kitchen appliances, while the field of programming attracts a very large mass of practitioners, and so software writing becomes a far more common form of expression. Weighting the goods and evils of patents, the evils weights heavier when it restricts the expressions of more people.
I know the "goods of patents" is a dubious thing as there are grounds to doubt that patents actually work the way people think even in fields like pharmaceutics.
The extreme difficulty in coming up with a working definition of "obvious"
... the much more easily determinable question of obviousness
I'm not clear on how you intend "easily determinable" to be interpeted here.
You are right, I am not fully consistent with myself. The extreme difficulty is with defining it exhaustively in its essence. The easy part is realizing that someting really obvious is indeed obvious, once you have internally a notion of obviousness.
On the political side, I am referring to how easy it is to convince the legislators about the obviousness of specific examples.
But consider that the politicians will then think that those examples are perhaps not typical, they are perhaps more like accidents of the process. They will also think about the dangers of declaring ideas to be obvious in hindsight. Against this, it help a lot to point out the rules adopted by USA Federal Circuit, where they expressly demand that for any combination of previously known methods, the combination as such be deemed obvious only if the prior art litterature contains a suggestion or motivation to combine the two. Given such a rule, it immediately becomes far more plausible that those examples are no accident. Unfortunately I do not know how this translates to EPO settings, how they defend, if at all, the obviousness of their patents. It would b
we do not need to buy these ideas. We already own them.
If we are talking about an idea that is truly non-obvious, it does not help to formally own it. We could need to convince the bright-heads to publish it. Imagine we were living in 1905, and the theories of relativity and quantum mechanics were not yet published. Suppose these scientists met and decided to keep everything secret, and rather user their knowledge to develop processes to produce goods in ways that were a complete mystery to the rest of the world.
There is a measure of absurdity to this discussion, because the vast majority, perhaps the complete totality, of the software patents are not truly non-obvious. The patent system trades something for something, and that is part of the mindsetting of the politicians as well as that of the public. They both believe that there is a lot of research effort behind most software products, and that each such product is based on a host of very clever methods. To tell the people this is wrong is doable, but it has to be done. When doing this, you not only have to enlighten them about something new, you also have to deconstruct all the misconceptions. It is not as difficult as telling them the world is not flat (if that were their deeply entrenched and well connected world outlook) but still you have to overcome some obstackles. Most of the arguments I have seen presupposes that the public and the politicians share much more of the arguer's outlook than they actually do.
While I do think that it is far better for society to treat software exactly the same way as mathematics, I think that at this time it is far the safest way forward to focus much more on those aspects of the patent issues that are really easy for the general public to grasp and relate to.
you seem to be arguing that we accept unreservedly those arguments that best support the pro-patent lobby
I cannot see we need to accept anything. We need to see where the major openings are. I do not argue that we need to accept software patents. I think that countering obvious patents of all fields, for instance patents in bio-chemistry that are run-of-the-mill applications of previously known methods in new areas, will help society more because it will do more to stem the flood of harmfull patents. I also believe that this stategy will scare the patent lobby more. We can still tell the world that software patents is a bad idea, and explain or point out the damage they do.
Still it remains important to understand what kind of concepts and misconceptions our audience harbor, and target the misconceptions properly. I am quite convinced that unnecessarily many decision-makers are giving scant attention to our arguments because they miss the points and suspect us of having incoherent views, not because they are evil-wanters.
presuming that MEPs are as ignorant and confused as you paint them, a meme we'd be foolish to adopt
Fortunately, you are right. Had MEPs been just ignorant and lowbrow, everything would have been much worse. Still our cause is a very small part of all the subjects the MEPs relate to, and they are often quick to switch attention away from us once they (incorrectly) decide "that is a hotblooded group with some rather incoherent views".
I guess that the size of the demonstartions, etc. has taken quite a few of them by surprise and made them take a second look at the cause. Still I think it helps winning credibility by winning a battle that is easier to win, over the much more easily determinable question of obviousness. My main point is perhaps that the obviousness of today's patents is an affront to the ideas that most people already have.
As it is I also believe we are given a second chance with some decistion-makers thanks to the sharpened attention the issue attracts because of the unusual combination of undemocratic tricks the patent
It presupposes that software is in fact the same as hardware
I am afraid that the general public, and the MEPs as well, do presuppose this. They will simply assume that software is something technical, and all things technical are the same. If they think any more about it, the first thing they see is that it is being sold on the merits of what it does, which the buyer presumably finds usefull.
The general public has no idea what program development is like. They do not see it as a form of expression any more than making kitchen appliances would be a form of expression. If our society can live with the restrictions on the kitchen appliance maker's rights to expressions, so it presumably can with restrictions on programmer's rights.
The general public also believes that program development is similar to hardware development they have learned about, like stores about Thomas Edison's countless failed efforts to make a working light bulb. They believe it when Microsoft pretends that without patent "protection" anyone can rip off the results of their research and make competing products at a fraction of the cost. They believe it when the patent lobby says that copyright does not cut it because they need protection of the method, not just a particular implementation.
All these sort of misconceptions make the battle against patents on software much more uphill than it would otherwise be. The burden of proof falls squarly on those who need the public to know better.
I can even imagine campaigning on emotional points that the public can understand, like saying that every time the patent office grants an obvious patent, a tool flies out of my toolchest. Stop the plundering of my toolchest!
We could emphasize that patents properly assigned should not limit others in any substantial degree. They should protect ideas that others would otherwise not have, except after seeing the patent disclosure. Those are the ideas that the society wants to buy from the inventors at an offered price of granting them a time-limited monopoly.
We should explain that many of the so-called novel ideas that today receive patents are novel only because they were not needed before. These ideas have become usefull because of developments external to the idea, like the advent of the computer or of the internet. The reason nobody has implemented the idea before is that it was not practical before, not that it required any intelligence or effort to find it. Such ideas are similar to making a closet that fits in a corner where the walls meet in a 105 degree angle. You don't need to make it until you are called by a customer who lives in a house with such a corner. That should not make the 105 degree closet patentable, and the same applies to all other obvious patents.
Slashdot Mirror
Hell, the UK is the only place I ever heard of where those wrongfully imprisoned are then forced to reimburse the government for the cost of their imprisonment.
In Norway, Per Liland, wrongfully convicted for murder and jailed for 14 years, had the cost of living he would have had if not jailed, deducted from his compensation. (Until there was an uproar and the deduction was cancelled.)
The logic was that the compensation was compensation for loss due to imprisionment. Without imprisionment he would have had living expenses. The compensation as granted by the Parliament was for lost income. Now they adjusted for lost expenses too. The logic is flawless: it had two part, the amount he would presumably have earned in a job, and a compensation for reduced quality of life. But they failed to compute a compensation for the insult of doing such calculations.
David Chaum's original system, Secret-ballot Receipt, does not have completely centralized decryption. There is a number of trustees, each with a private decryption key. If e1, e2, ..., en are the encryption functions, e1(e2(...en(ballot)...)) is sent to trustee 1, who decrypts and sends e2(...en(ballot)...) to trustee 2. Eventually trustee n receives en(ballot) and publishes ballot. But each trustee receives a batch of N encrypted ballots, and emits another batch of N ballots in a different order. Only the trustee itself knows which ballot in the emitted batch corresponds to which element of the received batch. Since the trustees can include political parties, human rights organizations, churches, etc, it is unlikely that all would collude to trace a decrypted ballot all the way back to the original encrypted ballot of which the voter has a receipt. There are additional mechanisms in place to prevent other kinds of abuses.
However, Chaum's original scheme was deemed to have usability problems for the voter (which I think are overblown), and its mathematical virtues are hard to explain to ordinary people beyond stating "it's sooo safe! Mathematicians say so!".
Homomorphic encryption has the property that e(n)+e(m)=e(n+m) for some meaning of = and +.
If 'n' and 'm' are answer vectors of 0 (no) and 1 (yes) answers to a vector of questions (Mickey Mouse for president? Donald Duck for president? etc), the sum m+n is the tally of those two ballots. Now I presume that with fully homomorphic encryption it is also possible to (e.g.) weed out overvotes and undervotes before adding up the results.
Even if it may be beyond most people how an encryption scheme can be homomorphic, it may be explainable to a somewhat larger section of the public how a homomorphic encryption can be used to tally the ballots publicly while maintaining the secrecy.
More about Chaum's scheme:
All inputs and outputs of all trustees are posted on a web site where anybody can download them, with some additional data that allows everyone to check 1) my encrypted ballot (as shown in my receipt) is there in the input to trustee 1, and 2) if a trustee has cheated with n ballots, anyone who runs the checking algorithm on the data will flag at least some of them, with probability (1 - 2^-n). Since usually more than 100 ballots must be stolen to steal an election, the chances of doing so undetected is less than one in 2^100. An detection will not just be "detection", it will be a mathematical fact that every math teacher in the world can verify (and many others). Additionally, there are similar guarantees that the encrypted data on the receipt really represents the votes that the voter saw and accepted in the booth.
If the ayatollahs had had such a system in place in Iran, they would probably have been able to convince the public that Ahmadinejad won - provided he did. So, even supposing that there was no fraud, the ayatollah's are to blame for the doubt and the protests. They could have avoided that making the elections more transparent.
The 2000 Florida fraud would still not have been prevented. More than 60000 voters were turned away because the shared names with alleged felons. All the recount discussions failed to consider them, discussing far smaller numbers of hanging chads and military absentee votes. No mathematical voting scheme could have prevented that. In Florida, those turned away should have been offered the opportunity to cast a provisional ballot that could have been counted after they had challenged their "felon" status.
If the aim of negotiations is a binding treaty, this is a back door to legislation without democratic control.
Has the general public ever had the use of reason? Secrecy is justified in two kinds cases. Protecting the privacy of individuals, and protecting military installations, weapon systems, response plans, etc. But none of these cases are of legislative nature. Why is it at all possible to pretend legislative or quasi-legislative preparations need secrecy, and get away with it? It is a blatant inconsistency in terms. It ought to completely discredit whoever forwards it. It should have elicited unanimous uproar from the congress...
Good grief.
Usually, there are two parties to an agreement, and third parties are all others. But when an agreement between A and B is changed to an agreement between A and C, becase B sold its part to C, then there are suddenly three parties, until everyone forgets B ever was part of the deal.
To complicate matters even more, there are multiple agreements here, one between the user (A) and Verizon/Fairpoint (B/C), and another between the user and (say) Yahoo. With respect to the second deal, Verizon and Fairpoint are third parties. With respect to the first deal, Yahoo, MSN, etc are third parties.
NOW tell me what "third party web site" the users will no longer have access to, or will no longer access their emails through.
I guess the real meaning is that Verizon had a web portal in which Yahoo and some other webmail providers was integrated, so that users could have the impression that Verizon was doing an email service. Now Verizon is closing its site, and Fairpoint is taking over, providing a new portal with similar integrated access to yahoo etc.
Agreed, but for the record: The link has a typoo, try to make the last part ?lnk=http# instead of ?lnk=hptt#. Of course that does not change the point, a faulty link on the front page is discouraging.
Nearest cell phone... Is there no routing? What prevents a message from being propagated all the way to Tierra del Fuego and back, and down there again, and... If sombody has a pointer to a proper answer, I'd like to have a look.
Could people tell us if they are being used in their precints?
Novelty and non-obviousness are distinct criteria for patentability.
However, novelty is easier to define, and so people and examiners prefer to make the non-obviousness an automatic corollary to novelty. I suppose this is part of the problem.
Please fix the typography; it is 1.783 x (ten-to-the-power-of-minus-twenty-seven), or if it works this time, 1.783 x 10^-27. A small amount by everday measures. (But surprisingly much for a moron like me. How can a particle this heavy be responsible for endowing electrons and neutrinos with their respective masses, that are mch smaller?)
We have been raised to believe ideas seem obvious only in hindsight. That is not the case.
The reason nobody did this before is that the technologies were not mature, the hard disks were too small (or too expensive), etc, etc. Yet this patent is not about anything that enabled this.
This patent monopolizes the problem, not the solution. It describes a product, not its implementation.
Superficially it is about an improved way of providing time shifting, which allows simultaneous recording and playback at a reasonable price. However, everybody at the time knew that harddisks offered this kind of advantages. The patent says use MPEG (to reduce harddisk capacity need), and everybody knew that also, but it was very hard to do the compression sufficiently fast with cheap processors and cheap memory. However, TiVo has not invented faster CPUs, better compression schemes or anything enabling.
I guess that when TiVo started, to get the whole thing working with the parts available at the time and at a reasonable price _was_ a challenge, and probably required some investments. However, this patent discloses nothing but a sketch, a plan that somebody skilled in the art would have written down in ten minutes. The tricky part is not disclosed. However, a couple of years later, with even faster processors etc available off the shelf, then also the implementation of the sketch becomes increasingly something that anyone skilled in the art can do without further tips and without much research. This is ironical: The product plan as described in then patent was obvious at the time, what was not obvious was not disclosed in the patent. Yet the continued evolution has made also that obvious so that it appears like the patent did disclose enough. However, here is where the hindsight is.
We might perhaps want a new kind of protection (other than patents) for small firms that do heavy research at the point in time when an inherently obvious principle becomes possible because of developments in other fields (ram capacities, CPU frequencies, etc) but then that should be formulated and the conditions, limits to applicability, etc. should be worked out. The patent system had a different (but related) purpose, but has fallen into such abuse that it is mostly by accident that it now protects a company that once did some hard work, ironically, through a patent the company acquired from somebody else long after they did their work.
I used to work for a company that made computers with a version of Unix on them. We would mostly manage to get a fix out in a week, but I managed to reduce that to about two days, in seldom cases one day. I was much of the bottleneck - I did the quality assurance and the packaging. This was in the early 90s. I improved the process by creating something that resembles (remotely) a redhat package system, that made it a snap to revert patches if anything went wrong. The effect was, not foreseen by me, to streamline the process so that we could focus much more on the contents of the fix. I saw in the beginning that also the packaging process could introduce errors and mistakes. Packetizing also helped testing, making it easier to install versions of the software, or just patches, test it, and then revert the configuration to what was needed for that computer's ordinary mission.
The case Godfrey v. Demon Internet is itself a case that reeks strongly of stupidity. In that case, an ISP refused to remove usenet postings falsely pretending to be from Mr. Godfrey. The ISP pretended they had no responsibility, because the defamation act does not apply to those who merely relay information like e.g., newspaper vendors. When this was struck down, the ISP found a batch of other messages with forged sender ids, pretending they were from Mr. Godfrey and alleged these messages were provoking. They wanted to have their responsibility reduced because of the alleged provocation. Through this step, they made the process much, much more costly, and in the end they had to pay for it all.
The ISP of course also claimed freedom of speech. They forgot to say who's speech. If the articles i question were actually from Mr. Godfrey, he has the the right to cancel them. The Usenet software has automated mechanisms he could have used. In order to claim freedom of speech for anybody else, they automatically were claiming the right to falsely write under the name of somebody else.
I think the legal systems in nearly all countries could be said to have several layers. In some countries it is pretended that every decision is a carefull but strict application of the statutes. However, any half-witted person can see that there is a great selectivity in how the statutes are applied and how the circumstances of a case are found to fit descriptions, or what labels apply to them. Therefore there is a second layer which could be said to consist of mostly "common sense" and "public interest", although this second layer is also often abused, perhaps out of stupidity, producing decisions that sometimes more than border to the mock trials.
However, a reasonable application of common sense and public interest should make it very clear that in absense of any editor, as the Usenet is designed to run entirely without any top-down control, then the human owners of the participating computers must be required to assist in limiting the damage. Society cannot, or at least should not, allow any important damage to persist just because of a technicality not foreseen by the lawmakers. I think that is why the judge made the ISP a "publisher" in the Godfree v Demon case.
This question should be entirely different in the present case, where there is a clearly identifiable and named responsible person behind the allegedly defamating articles. In the present case the articles are entirely under control of Craig Murray, and Fasthost's decision is interfering with his freedom of speech. I guess that if a storehouse chain decided not to carry The Guardian on a particular day because of potential liability for an alledgedly defamating article, the chain could be sued and would be likely to have to pay damages to The Guardian.
What remains disturbing is, that if Fasthost had decided to ask Usmanov's people to turn to Murray himself with their request, there is no guarantee that all judges would side with Fasthost. While I am damn sure that my analysis is correct (even I don't know the law), I am also damn sure that only a small majority of the judges have the brain to analyse the case correctly. (That is, a large minority does not.) Even the little I have seen quoted from the British defamation law, makes it appear like the law carries a list of circumstances when contributors to the publishing of a defamation are not to be held liable, while it fails to say explicitly why. It seems to happen over and over again, that because judges cannot be relied on to interpret overly general principles, the laws are written with lists of more specific nature, and the principle behind the list is not mentioned at all. I think this way of writing laws could be improved.
I have not read the actual text of the Godfrey v Demon decision, and I do not know if the decision makes it very clear why it found that the ISP is a publisher. In
What was the alleged crime, really? Was it that she had intentionally made the children see the stuff? Or was it that she had irresponsibly visited improper sites before the class, and not been able to control the resulting malware infection?
The first alternative sounds unlikely because so few would have suficiently strong motive. The second is much more credible, since so many are capable of being excited by porn. But in the second case, not having the idea to turn off the computer is not in itself a crime, and not evidence of what she did to have the malware infection occur.
The article says I suppose Horner's testimony was a sworn one. I can all to easily imagine a jury that fails to distinguish what they are supposed to believe, what the "computer crime expert" showed them, and what they are supposed to disregard, the prosecution then saying this is strong evidence.Anyway, I suppose the jury is not instructed to think that the prosecution is a gang of asses that will gladly twist things and (mis)interpret things to fool the jury in insiduous ways. Neither are they instructed to see the defence in this way, but for many, if not most, this goes without saying. Criminals have always defended themselves as best they could. Still most people prefer to believe that dishonest policemen is exceptional outside the crim novels. I am absolutely convinced that my sister, for one, would believe a policeman if he "looks like an authority".
How can we be sure the defense attorney is so utterly incompetent? The quotation from the article,
seems to indicate adequate competence, at least on the technical side. On the other hand, I think
indicates the prosecution is dishonestly making up "facts" and playing on their credibility as "voice of the authority", and a stupid jury or judge believes the accusation because "the authorities say so".
In court each side has a limited number of opportunities to speak. If the prosecution made this claim in a response to the defense witness, the defense probably (I don't know the actual protocol in this court) had limited opportunity to start a new debate, and find expert witnesses, etc. to counter this statement.
What about encrypting it too?
Consider the Linux distribution like a communcations network, and a mechanism is used to make a copy of a copyrighted work reach a select group of justified receivers. The blob is encrypted with the vendor ID, and the driver, besides, of course, not registering a device unless it supports it - and that is determined based on the vendor and model IDs - if it accepts to drive the device, decrypts the blob with the vendor ID of the device before loading it. 16-bit encryption with known key is not a big deal to crack, but for a non-intended receiver to do so manually would be an extraordinary step. It clearly goes beyond ordinary usage, and constitutes a kind of break-in. Since the driver and the blob are available on the net unencrypted, there is no point in actually having a strong encryption, but rather to have a reasonable token of directing the communication to the justified recipients, and not to everybody.
At some point copyright laws were amended so that "copies" that exist in wires and routers during transmission would not be seen as copies in the sense of copyright laws. It is almost unbelievable that anybody thought such amendments were needed. Perhaps they were, back then. Judges and lawmakers need time to get their heads around new fields. Some of them need, anyway. But in spite of impressions one sometimes get, laws are not usually interpreted so blindly and mechanically. Imagine, people could have been sued for opening a book in a room with a mirror!
Concepts in the laws, like "copying" and "distribution" must have a sensible reality, e.g of producing another copy in a way that has any real consequences, like somebody enjoying a copy by reading it, listening to it, etc, or at least having a possibility thereof. I do not pretend that judges always observe this requirement, but in principle it is required, and that is why you cannot be sued for opening a book in a room with a mirror. There is no exception for mirrors in the law.
Are we there today, that we can hope judges will accept that a binary blob in a linux distribution is not being "distributed" if the communications channel does not "unpack" it, or "deliver" it in some sense?
Still I feel that it would be risky to include it in a Linux distribution. If somebody sues, there is no guarantee that the judge will be reasonable. In the long run there is little doubt, pargmatism will creep in. In the short term, it is risky.
Similarly, owners of wireless devices have paid for, and the wireless device vendors have sold and received money for a permission to download the firmware to the device. I doubt the wireless vendors have any right to force users to use Windows or the Windows driver to send that firmware file to the device.
In many countries, specifically in Norway, where I live, there are laws to protect the right to do reverse engineering. Any license agreement not to do reverse engineering is invalid here. The laws say explicitly that these provisions cannot be overruled by agreeements between the parties. It should be possible to extract the firware from the driver in a country that allows this.
I strongly doubt that distributing the firmware code to owners of the corresponding devices can properly be seen as distribution against the permission of the copyright holder. It may even not be considered to be "distribution" in the sense of the IP laws. While companies may try to convince judges about anything they see convenient, especially if they have secret arrangements with a monopolistic software vendor, it is not too likely that the courts will believe them.
I am more uncertain about the inclusion of the firmware blob in e.g. Linux distributions. Instead linux distributions could be equipped with software that automatically searches the net for the binary blob. Such software would only work if the computer has another way of getting connected. On the other hand, the software could be made to do this under Windows too.
Here I am talking about the firmware, not the driver. As more functionality is moved into the driver, it probably becomes more difficult to reverse engineer the device driver even if the firmware blob can be successfully loaded into the device. Still it should be doable, while reverse engineering the firmware is next to impossible.
The above statement seems to contradict any such interoperablility issue. Is the statement misleading, or are all the others who talk about this aspect completely off the mark?
Unfortunately, the competition will get around it by wearing brown or black boots... Perhaps I can patent that too?
But the interesting thing is that I don't need to know anything about nuclear physics to patent it. I just patent refurbishing, said refurbishing comprising putting the boots on to protect from nuclear spill, enter the premises showing badges, doing the modification, cleaning up the mess, leaving the premises, and disposing of the boots as hazardous waste.
Never mind that patent applications should describe the invention so that a person with ordinary skills in the art can complete it by following the instructions in the patent description. Anyone who has ordinary skills in the art of nuclear plant refurbishing already knows how to do that, so why must I describe it?
Just look at this patent.
There is not a single word about how to create this program or computer system that allows the user to create a user account, etc. Why bother? Anyone having ordinary skill in the art of computer programming already knows how to do that.
But the law also has a requirement that the invention be non-obvious. What happened to that? Well, they found it so difficult to determine reliably if something is obvious, that the reformulated it to mean, if you have to combine the teachings from multiple pieces of litterature, then the combination is non-obvious even if each part has been described in the trade litterature.
In other words, let's go ahead and patent things, until nuclear plants are finally forced to close. One patent for the sake of the environment every day!
We should rather "DDOS" the politicians who let us down. For each spam you receive, send a copy to each member of the congress/parliament asking them to provide you with a legal way of defending yourself against this, or to enforce any solution they have already legislated.
If a politician acts credibly against spammers, delete him from your list.
In a country with 25 million voters, if 10 percent do this, and they receive on average 20 spams per day, this is 50 million mails to each politician.
Bind code to do this to the "spam" button on your email client. Then every mail will have been sent manually and for a good reason. There is no need to coordinate the timing, the point is not to bring down the MEPs servers but to simply remind them that the spam receivers are voters too.
Why are digital watches fundamentally different from analog ones? Or are they not since both deal with a "force of nature", time?
If we for the time being skip "computer-implemented inventions" like cd/dvd music/video players that could be analog or digital, and instead consider the computer as an information processing device, a communication device, etc, what then about a pair of glasses? Is that patentable? You use them to read! What about a desk calculator? A rotary press? A photocopier?
I guess you will rather compare all these examples to the physical aspects and workins of a CRT or LCD monitor, a hard disk, etc, which most anti-software-patent people think should be patentable.
Before the advent of the modern transistors and microchips, there was a branch of the "computing" industry that created "analog computers". In an analog computer, a capacitor could be used to integrate a function represented as an analog current. They had a rich library of devices that could represent various mathematical operations. What do you think? Shoud none of that be patentable? Do they not express human ideas?
I absolutely think we should do everything possible to stop software patents, and you have mentioned a couple of reasons that I do not doubt at all: They are not necessary, and they do a lot of harm, they prevent rather than further innovation, they increase cost for everybody, but the mayor corporations still want them because they know the smaller companies will feel the burden much, much harder, since they must pay those expenses from incomes on fewer sold units.
I do not think the anti-software-patent movement should stop saying all this, even there are reasons why it takes time before such statements sink in and are believed. (Some of that time has already passed, and quite a few have been convinced.) Yet if we give more focus to the obviousness problem, then many of the other statements become easier to understand.
Why are we not all equally against patents on clever ways of making aluminium car bumpers? Such patents restrict the competition. They increase cost for everybody and the smaller companies feel it harder. Large metalurgical and chemical conglomerates acquire patent portfolios and cross license them. That makes it hard for smaller companies to enter the field.
There is a logic behind the patent system, and this logic is not consistent with the realities. The logic is that
invention is hard and expensive and risky
ripoff is easy after the invention has been published or the product is on the marked and can be studied.
Inventions are ideas that the society would not have had, had the inventors not published them. Therefore it does not cost society much to give patents. Only occasionally are some inventions made independently by multiple inventors. Giving a monopoly for ca. 20 year only costs something the last approx. 10-5 years, that is when less gifted people would have found them independently.
Those few times patents prevent genuinely independent developments are a price worth paying for the benefits of a healthy network of scientists and inventors that are able to live off their creations.
Those few times when inventors would have published the invention even in absence of patents, the society pays a monopoly premium and those who would have ripped off the invention loose business. But society should give moral priority to those who made it all possible, and not defend the rippers.
For most people this logic is about as certain as that the world is round, or they believe it as firmly as many believed the earth is flat in the middle ages. Tell them that it is all wrong, and they w
Easy? How?
Politicians are quite used to all sorts of lobby and pressure groups using all sorts of emotional, or untrue, or half true, yet sometimes quite persuasive, arguments. How do you convince them that Microsoft's wealth is not dependent on software patents? Yes, I know the statistics, the late pickup of patenting from Microsoft. But Microsoft says otherwise. They could say that things are different now because the field is more mature, or any other b******t.
I briefly considered if we could show that virtually no patent discribes the innards of programs. All patents only describe what programs do, i.e., the business idea behind them, or the problem that they solve. Show me a single patent that addresses the innards of the programs! Then I remebered that Microsoft recently got a patent on modeling something as an object, I just can't remember what it was... a mail address or contact information? They will pretend it is a very valuable idea, that there is a lot of research behind it, and they need patent protection to prevent others from exploiting their research.
Would it be possible to say that computer programming is more of a craft than an advanced science? Why is copyright enough protection?
I think the answer to that question is because the difficulty in computer programming is to get all the details to fit together. If someone wants to create an independent implementation of a program, even if he uses the same broad "methods" as the original, the cost is about the same. There is no substantial saving from copying the methods from the earlier implementation.
What statements are likely to gain almost-universal support, enough to convince politicians that are not themselves programmers?
My last statement above is not 100% true, because the process of constructing the requirements and working through the use cases of a large system is not always a small part of the development cost. Still it is true to a high enough degree to give investors a reasonable protection. If a market is large, there will be competitors attacking it, with considerable startup costs and considerable first-mover advantage to the initial investor. If the market is small there will not be very many competitors for a long time, because of the risks involved.
It seems we are approaching (yet another) significant difference between software and other fields: Software enterprises can be micro-scale, with zero or near-zero startup costs.
Or: Software production attracts kids.
Or... What is really the essence here? I am not kidding, I think there is something definite to be said here that you can convince any reasonable person about, and that has a bearing on the software patents issue. Something that does not so easily get dismissed as "just a pressgroup litany or religion". But I am not yet sure what that is.
On the other hand, little Johnny programmer is not that likely to get sued, because it costs mony to sue and the patent holders are not that likely to see any benefit in suing him. And people are surprisingly willing to dismiss principles on grounds of pragmatism, meaning they do no
I fully agree. The problem is that people, including the legislators, believe software embodies "working principles", i.e., methods, that need protection not provided by copyright.
This is one of the points where I feel we have failed to analyze the situation correctly, and therefore we have failied to address an important underlying misconception.
The legislators are not stupid, but they do share all the popular misconceptions, at least until we address them properly.
Because the patented subject matter is so perfectly obvious, any little Johnny that strays near the subject will hit upon the same idea.
Had patents been restricted to truly non-obvious inventions, the danger would have been much, much smaller. This is not to say the danger would have been non-existent, among other things, because as time goes by, the once novel and non-obvious idea becomes common good. This is one reason why software patents are evil beyond the evil of obvious patents.
But notice how this argument could be applied to any field. What makes software special? You say because software is expression of ideas, but so are watches and kitchen appliances. What matters here is that little Johnny is unlikely to engage in making watches or kitchen appliances, while the field of programming attracts a very large mass of practitioners, and so software writing becomes a far more common form of expression. Weighting the goods and evils of patents, the evils weights heavier when it restricts the expressions of more people.
I know the "goods of patents" is a dubious thing as there are grounds to doubt that patents actually work the way people think even in fields like pharmaceutics.
You are right, I am not fully consistent with myself. The extreme difficulty is with defining it exhaustively in its essence. The easy part is realizing that someting really obvious is indeed obvious, once you have internally a notion of obviousness.
On the political side, I am referring to how easy it is to convince the legislators about the obviousness of specific examples.
But consider that the politicians will then think that those examples are perhaps not typical, they are perhaps more like accidents of the process. They will also think about the dangers of declaring ideas to be obvious in hindsight. Against this, it help a lot to point out the rules adopted by USA Federal Circuit, where they expressly demand that for any combination of previously known methods, the combination as such be deemed obvious only if the prior art litterature contains a suggestion or motivation to combine the two. Given such a rule, it immediately becomes far more plausible that those examples are no accident. Unfortunately I do not know how this translates to EPO settings, how they defend, if at all, the obviousness of their patents. It would b
If we are talking about an idea that is truly non-obvious, it does not help to formally own it. We could need to convince the bright-heads to publish it. Imagine we were living in 1905, and the theories of relativity and quantum mechanics were not yet published. Suppose these scientists met and decided to keep everything secret, and rather user their knowledge to develop processes to produce goods in ways that were a complete mystery to the rest of the world.
There is a measure of absurdity to this discussion, because the vast majority, perhaps the complete totality, of the software patents are not truly non-obvious. The patent system trades something for something, and that is part of the mindsetting of the politicians as well as that of the public. They both believe that there is a lot of research effort behind most software products, and that each such product is based on a host of very clever methods. To tell the people this is wrong is doable, but it has to be done. When doing this, you not only have to enlighten them about something new, you also have to deconstruct all the misconceptions. It is not as difficult as telling them the world is not flat (if that were their deeply entrenched and well connected world outlook) but still you have to overcome some obstackles. Most of the arguments I have seen presupposes that the public and the politicians share much more of the arguer's outlook than they actually do.
While I do think that it is far better for society to treat software exactly the same way as mathematics, I think that at this time it is far the safest way forward to focus much more on those aspects of the patent issues that are really easy for the general public to grasp and relate to.
I cannot see we need to accept anything. We need to see where the major openings are. I do not argue that we need to accept software patents. I think that countering obvious patents of all fields, for instance patents in bio-chemistry that are run-of-the-mill applications of previously known methods in new areas, will help society more because it will do more to stem the flood of harmfull patents. I also believe that this stategy will scare the patent lobby more. We can still tell the world that software patents is a bad idea, and explain or point out the damage they do.
Still it remains important to understand what kind of concepts and misconceptions our audience harbor, and target the misconceptions properly. I am quite convinced that unnecessarily many decision-makers are giving scant attention to our arguments because they miss the points and suspect us of having incoherent views, not because they are evil-wanters.
Fortunately, you are right. Had MEPs been just ignorant and lowbrow, everything would have been much worse. Still our cause is a very small part of all the subjects the MEPs relate to, and they are often quick to switch attention away from us once they (incorrectly) decide "that is a hotblooded group with some rather incoherent views".
I guess that the size of the demonstartions, etc. has taken quite a few of them by surprise and made them take a second look at the cause. Still I think it helps winning credibility by winning a battle that is easier to win, over the much more easily determinable question of obviousness. My main point is perhaps that the obviousness of today's patents is an affront to the ideas that most people already have.
As it is I also believe we are given a second chance with some decistion-makers thanks to the sharpened attention the issue attracts because of the unusual combination of undemocratic tricks the patent
The general public has no idea what program development is like. They do not see it as a form of expression any more than making kitchen appliances would be a form of expression. If our society can live with the restrictions on the kitchen appliance maker's rights to expressions, so it presumably can with restrictions on programmer's rights.
The general public also believes that program development is similar to hardware development they have learned about, like stores about Thomas Edison's countless failed efforts to make a working light bulb. They believe it when Microsoft pretends that without patent "protection" anyone can rip off the results of their research and make competing products at a fraction of the cost. They believe it when the patent lobby says that copyright does not cut it because they need protection of the method, not just a particular implementation.
All these sort of misconceptions make the battle against patents on software much more uphill than it would otherwise be. The burden of proof falls squarly on those who need the public to know better.
I can even imagine campaigning on emotional points that the public can understand, like saying that every time the patent office grants an obvious patent, a tool flies out of my toolchest. Stop the plundering of my toolchest!
We could emphasize that patents properly assigned should not limit others in any substantial degree. They should protect ideas that others would otherwise not have, except after seeing the patent disclosure. Those are the ideas that the society wants to buy from the inventors at an offered price of granting them a time-limited monopoly.
We should explain that many of the so-called novel ideas that today receive patents are novel only because they were not needed before. These ideas have become usefull because of developments external to the idea, like the advent of the computer or of the internet. The reason nobody has implemented the idea before is that it was not practical before, not that it required any intelligence or effort to find it. Such ideas are similar to making a closet that fits in a corner where the walls meet in a 105 degree angle. You don't need to make it until you are called by a customer who lives in a house with such a corner. That should not make the 105 degree closet patentable, and the same applies to all other obvious patents.
Regards.