Talk:Software patents under the European Patent Convention

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Amazon's One-Click US patent has no equivalent in Europe[edit]

I removed the "since 1-click is patentened by the EPO as well". This is a fallacy since US Patent 5,960,411 entitled "Method and system for placing a purchase order via a communications network" has not equivalent in Europe. --Edcolins 22:03, 19 May 2004 (UTC)

Criticism of EPO decisions[edit]

I removed the following sentence:

Some EPO decisions have been harshly criticised in member states, especially Sohei and Pension Benefit System.

Please cite your sources. Thanks. --Edcolins 19:59, Nov 24, 2004 (UTC)


What "should" happen is not at stake here.

There is one interpreatation by the EPO that is not followed in many EU member states. In some countries there is no "as such" clause in national law. The "as such" clause was introduced by the German delegation to clarify the 5.2, historically the EPO's current interpretation is legal escape.

Matter removed[edit]

I removed the following sentence:

However, despite initial intent of the drafters of this language, the EPO has granted a number of patents whose only technical effects were indeed effects taking place within the computer itself (more efficient compression, image storage formats, various algorithmic improvements), or effects which are social rather than technical (many business method patents).

Please cite your sources. Where have you read that the drafters of the European Patent Convention wanted to exclude inventions with technical effects lying in a computer itself? You also claim that the EPO has granted patent with "effects which are social rather than technical". I understand that you mean only social and not technical at all. Is it correct? Where have you read that the EPO case law accepted such things? Thanks for the information. --Edcolins 21:45, Dec 8, 2004 (UTC)

EPO practice[edit]

The fact that the EPO grants patents does not mean that its interpretation of the EPC has any value.

There are a lot of legal experts such as Lenz who do not follow the EPO interpretation of EPC 52.

And there are a lot of courts that do not either.

In fact it was proposed to delete the 52 provision for the EPC renegotiation round in 2000, but member states did not agree.

EPO practise anticipated a deletation of the provisions and did not correct its pratice after the 2000 decision. Many scholars regard this as a example for legal escape.

Please cite your sources on "a lot of legal experts". Mr. Lenz who teaches at a japanese university seems to be more or less the only law professor fundamentally criticizing EPO CII case law as inconsistent with the EPC, and thus implicitly accusing EPO BoA judges of bending the law.

Here's the UKPO's Peter Lawrence, speaking last December [1]:
"There is one point I think which the officials certainly are dying to make about these EPO examples. I mean we would not say that everything the EPO has granted should have been granted, and indeed one of the reasons why a number of member of states are quite keen on the directive is because of a concern about maybe there was drift at the EPO as well as elsewhere".
and the UKPO's Peter Hayward, at the same meeting:
"...Can I just make a comment on the EPO's position. We said we are maintaining the status quo. Back in around 2000 we became aware that there were certain groups of examiners in the European Patent Office who were perhaps adopting a slightly more generous approach because they thought we should actually move towards US - since then the EPO has clearly laid clearer standards - the EPO Boards of Appeal have laid down clearer standards - but those patents that were going through the system then - patent applications that were going through the system then - some of them have now surfaced as granted patents. I suspect that there are quite a number of those that would not now be granted by the EPO - it was a brief blip in their history, if you like".
YMMV as to whether you're as sanguine as Mr. Hayward about how far you think the EPO may or may not have rowed back since then; and where it may go in future. Jheald 23:10, 2 November 2005 (UTC)
The critical legal experts include late president Kober. While he may have been compelled by his office to express himself in a more diplomatic manner in - public - statements, his endorsement of the statement that the IBM decisions anticipated a change in law among lawyers is, nevertheless, an unambiguous remark as to the illegality of these decisions at the time they were issued. The fact that the changes didn´t take place but were explicitely rejected at the diplomatic conference allows the conclusion on the basis of the statements of an acting president of the EPO that the board of appeals during his presidency acted against the law.
A UK High Court deputy was quite critical of the EPO Technical Boards of Appeal in CFPH LLC's Application [2005] EWHC 1589 (Pat) — Mr Prescott QC. See paragraph 54 "...Unfortunately the jurisprudence of the EPO is not constant. Like the declination of the compass, it can shift over time. Therefore, cases decided by our courts a few years ago may not be saying quite the same thing as the EPO is saying now. That is particularly true of the topic of patents and business methods. Our courts have followed decisions of the EPO which the EPO itself no longer applies...." —Preceding unsigned comment added by (talkcontribs)

Small edit[edit]

I updated the title of the link to my page ( because I was erroneously called a patent lawyer. I'm a European patent attorney, not a lawyer. Also I changed the "under the EPO ideas" title to something more neutral. Arnoud Engelfriet

Thanks for this correction and apologies for this. --Edcolins 19:37, July 12, 2005 (UTC)

Legal precedence and source[edit]

I corrected the misconceptions on the value of legal precendence before the EPO. The EPO widely follows the continatal civil law. It is more than common to read sentences like : The division chooses to diverge from the reasoning of decision T...., also in first instance decisions. This would have to be considered a procederal and legal violation under common law, it is not under the Eurpean convention. —Preceding unsigned comment added by (talkcontribs)

Thank you. Indeed. I recently amended the article about the appeal procedure before the EPO in the same direction.
By the way, could you provide a source for:
"More recently, a more restrictive stance can be observed at least with the first instance organs of the EPO. In 2005 the vast majority of terminated procedures in the field of computing did not give rise to a European patent. Out of 8.000, only 1.200 were granted and 350 refused. Most applications, about 6.000, were withdrawn by the applicants in reponse to the search report or substantial objections in examination. After an absolute moratorium of several years, an increasing number of patent applications is refused due to exclusion from patentability under Art. 52(2)(c) EPC in line with recent decisions taken by higher national courts of the member states. It is to be seen whether these EPO first instance decisions will be contested and if, whether the Boards of Appeal will uphold them."
This is an interesting paragraph, but a source would be needed for this I think. Cheers.--Edcolins 19:23, 28 January 2006 (UTC)
Given the collection and topics of the articles you have provided to wikipedia I am confident you are in a position to verify their vericity on Monday. Up to that moment you have to take my word for it, as we take yours for your contributions.—Preceding unsigned comment added by (talkcontribs)
Do not take the accuracy of my contributions for granted. Wikipedia's readers don't have to take your word and contributions for granted I am afraid. An assertion should not be found in an article without a verifiable basis, in particular when statistical figures are given (the above paragraph). Please cite your sources! ... Since the EPO Annual report 2005 has not been issued yet, I don't know where you have picked these data. For now, I have removed the paragraph until a source is provided. Thank you. --Edcolins 14:04, 29 January 2006 (UTC)
The nature of the information provided makes it clear from which source the information is verifiable.
The EPO on their website suggest the following:
Alternatively, if you have a general enquiry relating to the patent granting procedure (EP/PCT) or the EPO' s products and services and would prefer to contact us by telephone, please call EPO Customer Services on:
Tel: + 49 89 2399 4636
If only things which are published in the EPO report are to be mentioned in Wikipedia, there is no point of having articles about the EPO, a link would suffice.
Actually not only the text provided by Edcolins, but more so the harsh censorship of any contributions from other sources makes it appear that the EPO is misusing wikipedia as their public relations platform. I suggest they run what they consider information under and not It is highly inappropriate that the EPO excercises censorship and writes the articles about themselves particular on controversial issues such as software patenting.
I don't think that the the Republic of China should publish the articles on Taiwan and Tibet in Wikipedia, or SCO about Linux. I suggest that some editors consider whether their professional affiliations are not in irreconciliable conflict with the role of a neutral editor.
Should the harsh unjustified censorship under the guise of requesting sources on even the most mild issues which could be considered cristism persist, I suggest to put on vote whether most of the articles relating to the EPO should be not taken off. —Preceding unsigned comment added by (talkcontribs)
I am afraid it is not sufficient that the nature of the information looks like verifiable. It must be verifiable: "The burden of evidence lies with the editor who has made the edit. Editors should therefore provide references. Any edit lacking a source may be removed." We are not asking the earth, just a source...
Of course it does not need to be in an EPO report, it can be anywhere (from any reliable source). There is no censorship of "any contributions from other sources", but information from unverifiable sources or unknown sources may be removed. Thank you again for your kindness and help. --Edcolins 21:25, 30 January 2006 (UTC)
I think you misunderstand your role here. Who gave you the permission to define new wikipedia rules? —Preceding unsigned comment added by (talkcontribs)
It is elucidating - some might consider it betraying - that Edcolins contests the provision of a self-evident source rather than the vericity of the facts. He evidently acts on behalf of the EPO or at least believes so. The facts submitted may not have been published but they are verifiable and don´t just look so. His own contributions are full with subjective assessements such as "what some have described.."
What is more disturbing is the totally one sided preception of reality. He does not cite a single national decision which has invalidated a patent for exclusion from patentabilty, but only those supporting his and possible the views of the EPO, while at the same time the vast majority of UK and German decisions have annulled computer implemented inventions. It only remains to be speculated upon why the EPO evidently has an institutional opionion on the matter which is kind enough to let us know through their wiki editor, rather than acting as a neutral observer which it is obliged to by their institutional duties. May it be that software patent adversaries don´t pay fees for their opinion to the EPO while patent holders and seekers do? —Preceding unsigned comment added by (talkcontribs)

Landmark decisions vs relevant decisions[edit]

Replace the term "Landmark" which is a positive and subjective judgment on importance by "relevant". The choice of decisions cited up to now is arbitrary but not random. The article failed and still fails to cite decisions which do not support the author´s opinion. I will look up a few others beside T 158/88. —Preceding unsigned comment added by (talkcontribs)

Original research[edit]

I removed the "statistical" part of this:

While according to the the European Patent Register 85% of the Patent applications which qualify as software patent applications according to some definitions of the expression "software patent" presently do not give rise to a granted patent ever but are refused on various grounds or withdrawn by the applicant in response to the search report or substantive objections of the European Patent Office (EPO), some of them have been granted by the EPO since the '80s. At present. applications in this field account for 7% of the total number of applications to the EPO, but only for about 2% of all granted patents.

No source nor explanation of how to arrive at these percentages is provided. To me it looks like original research on the basis of the European Patent Register. --Edcolins 09:41, 12 February 2006 (UTC)

The source is given, counting is not original research: "However, research that consists of collecting and organizing information from existing primary and/or secondary sources is strongly encouraged. In fact, all articles on Wikipedia should be based on information collected from primary and secondary sources. This is not "original research", it is "source-based research", and it is fundamental to writing an encyclopedia." —Preceding unsigned comment added by (talkcontribs)
The manner in which you made your count involves selecting particular parameters (which are not dislosed). It does not merely amount in collecting and organizing information from primary or secondary sources. The choice of these particular parameters, e.g. selecting a particular classification IPC, involves deciding which classifications encompass software or so-called computer-implemented inventions and which classifications do not. This is a difficult process which involves many hypotheses and does qualify as original research IMHO. --Edcolins 08:14, 14 February 2006 (UTC)

Rather than having a ping-pong match of edits and reversions, can I suggest that it would be useful to have the raw figures here on this talk page, ie the actual number of patent applications, grants and rejections that you are deriving these percentages on, and the criteria used to select them? That way we'd know exactly what we were talking about; we could also be sure eg that applications still in the system weren't being miscounted; and that we're not just talking about a blip year. If (but only if) we can verify the numbers and what they mean, I think the percentages could be a useful addition to the article. (Also, I wonder, how many of them were "won't search" business method rejections?). But for any statistics like this, I do think it is important to know the raw numbers, and the counting criteria.

I hope this would be an acceptable way to go forward. -- Jheald 09:34, 14 February 2006 (UTC).

The method and source are now specified in the Article. I note that nobody, neither Jheald nor Edcolins, had any objections to the previous statement "software patents by some definition". They have not objected to hardly verifiable statemens as : "So far there does not appear to have been any case", "Though many argue ..". The present version is better than what it was, and it is verifiable much more than the personal opinions of other authors still in the text.
I hope this ends the discussion.
I have tried to obtain the statistical data indicated in the second paragraph of the article by introducing "G06F" in the "Classification (IPC)" field and "2005" in the "Application Number / Date" field [2]. I get 3944 documents! How can you easily obtain statistical data about the granted patents and so on? Do you take into account European patent applications filed, published, or granted (or withdrawn or deemed withdrawn) in 2005? Please explain. --Edcolins 10:34, 20 February 2006 (UTC)
Criteria other than "filed" can be obtained quickly by counting already for 2005. This is because of the 18 month secret period, so for the filed ones you´ll have to wait some more months and in the mean time be happy with the periods before 2005. The verifialbility criterion does not mean that you get a link and read the same number somewhere else: Nobody contests the following statement: "The 2005 (U.S) defense budget amounted to $401.7 billion, an increase of 4% over 2004 and 35% since 2001, with over 50% being spent in research & development." in the USA article, they give no source nor method, as it self-evident. There is no link to a page showing the same numbers. Same thing here: if loading the documents is too much of a pain call the EPO (less of a pain than trying to call Mr Rumsfeld). Without personal offense, it is clear from the subject and detail of your contributions relating to the EPO which clearly have not been obtained from scanning press releases that in particular you would not have any difficulty to verify or if case should be falsify them. In this light, the fact that you attack counting as research under a very extravagant personal interpreation of the procedures underlying the making of wikipedia, rather than verifying or falsifying the data is peculiar. The only plausible explanation I have is that you or whoever on whose presumed or actual interest you censor are not happy with those true numbers and don´t want them to be known. I may be wrong of course, but only you can shed light on your motives.
Instead of making personal attacks, which is inappropriate, try reaching a consensus. You should explain from which reputable source you obtained the information or explain how any reader can verify it. You probably know that this is an official policy on wikipedia:
  1. Articles should contain only material that has been published by reputable sources.
  2. Editors adding new material to an article should cite a reputable source, or it may be removed by any editor.
  3. The obligation to provide a reputable source lies with the editors wishing to include the material, not on those seeking to remove it.
Is criterion 1. met? No, I don't think so. --Edcolins 17:53, 20 February 2006 (UTC)
I gave you the answer, nothing I can do about you not wanting to hear it. I consider the European Patent register a reputable source just as a telephone book and it´s public. My style in the Talk may differ from yours, you may not like mine, I´m indifferent to yours. I´m not indifferent to somebody purposefully providing exclusively data - at times wrong data - supporting his cause, supressing under pretense of policy data which for whatever reason he does not want to see on wiki. As long as wiki accepts basically anonymous edits it is not the Britannica. And if it were, neither you nor me would be allowed to contribute at least not on issues on which both of us have a pronounced even if opposite personal opinion. A different process leads to a different product. Wiki is directed to mature readers who don´t simply take for granted what they read on it, but also have to judge the intentions of the editors. I think I am pretty transperant if not I have no problem making it transperant here. I think software should not be patented as I don´t want that what I produce as an engineer is harvested by lawyers with no added value whatsoever. Lawyers whether by education or vocation may think differently.
You sought advice from a user Durin who voted you into sysadmin. Read all his User page and what he says about censorship. The numbers I gave are correct, they can be verified much easier than the GDP of the US and what may or may not be in your not so humble opinion the development of patent law in Europe. Don´t like them don´t believe them but don´t you dare to censor them. I learned my part of the lesson and I no longer censor chapters I find uttermost redundant and biased see below
How to get the percentage of "European patent applications in the field of electric digital data processing (G06F IPC) that did not give rise to a granted patent but were refused on various grounds or withdrawn by the applicant" is still a mystery to me. Of course the European Patent Register is a reliable source, but which fields are you filling with which criterions to get the 85 % figure? Jheald's suggestion was wise IMHO. As far as calling the EPO is concerned, the EPO phone number is public indeed but that does not make a future hypothetical phone conversation with the EPO public or published ("has been published by reputable sources").
Again, which fields are you filling with which criterions to get the 85 % figure? If the answer is by counting, this is original research. If not, please provide the criterions. --Edcolins 12:18, 21 February 2006 (UTC)
Counting is not original research. If somebody reputable such as the New York Times or Jyllands Posten says Mr X and Mr. Y each own ten bucks, the statement in Wiki that Mr Y and X together own 20 bucks is a condensation not research. If anything is original research it is for instance to pick in the thicket of national decisions only those which have maintained the patentability of software, and evidently purposefully hiding the majority of recent national decision which rejected it. I have not and will not censor this section while evidently biased and rendering an objectively wrong picture. At a certain point in time when its raining and I have nothing else to do, I´ll simply put the references to the latter on Wiki.
Your arguments as to verifiabilty are legal sophism without even any inner logic :
"Verifiable" means that something can be verified, therefore, any -ability refers to a possible and hence hypothetical future act. This future act is not the publication but its verification. The fact that you don´t want to verify it does not mean it can´t be verfied. Hardly any use I call as evidently you don´t believe me, but maybe other readers would, I mean call not necessarily believe me.
On the contrary, your logic coercively and imperatively requires publication of these numbers. We all want to assume that Wiki, overall, is a reputable source. I am sure you claim a merit in this cause. If consequently you assume that the undisputed article on "Softwarepatente" in the German section is reputable, you´ll note that you´ll find very compatible data. However, if you think that Wikipedia is not a reliable source why do you bother - in the first place - to censor it. Before you end up in an endless loop without finding the reset button or even worse reply to this, let me press the eject button for you. I simply could not resist the temptation to answer your appreciated but probably involuntary humor and sophism with my unappreciated but at least voluntary one.
The discussion is now getting repetitive from your and my side, no sense continuing it. If need should be, it can only be usefully discussed by means of the arbitration schemes on articles and sysadmins forseen by this ingenious social project called wiki. Need would be if a clear and possibly professional representative of an interest group uses wikipedia for lobbying and/or industrial public relations and continues to censor reliable information not fitting his objectives. While not a proven fact, the suspicion of possibly professional lobbying, is not a personal attack but is supported by the circumstantial evidence of the sheer number of your contributions, their temporal distribution, their close to absolute monothematic nature and single view point under the eyes of everybody. No need from my side to initiate arbitration would be, if others - freelance or professional lobbyists or not - continue to stay verbose, unverifiable and pointless in sections such as the dead EU directive and generally single eyed, as long as they excercise no censorship. Simpler: you´ll have to live with me and vice versa. Your demonstrated dedication to this Borgesian past time makes it unlikely you´ll disappear -while less dedicated- nor won´t me. I may be as one eyed as you are; together we make the readers see a wider horizon, even if neither of us may ever see this horizon beyond our respective limited fields of view ourselves.

EU directive[edit]

Should not have been here in the first place, because the EU is not regulating the EPO. >The directive has its on article which is more than it deserves. What is a likely influence of one on the other is speculation. In any case something that definitely did not happen definitely has no relationship or influence with "Software patents under the European Patent Convention". —Preceding unsigned comment added by (talkcontribs)

The EU directive should be mentioned here since it is closely linked to the situation of software inventions under the EPC. National courts of EU member states (except Malta) decide on the validity of European patents. The EU directive is about harmonizing the patentability of software inventions at a EU level, and this will at least, by definition, have a direct influence on the validity of such European patents at a national level. --Edcolins 08:14, 14 February 2006 (UTC)
´quo usque tandem ubutere patienta nostra´

The EU directive isn´t about anything, some - evidently you - believed it WAS. Did not happen. Pointless to write about things which didn´t happen. It is more than enough to have one pointless article about a non-existing directive. Does not deserve redundant repetition not even a link there to. BTW Edcolins unreasonable behaviour will require arbitration not only of his articles but of his role as a sysadmin. Evidently making one´s bucks on intellectual property is incompatible with providing free access to information. Get a life beside your hobby as the big inquisitor and censor. Maybe the Chinese government has to offer you a job in their internet big brother office, wiki doesn´t. If the Chinese don´t, you can always breed orchids, so long man. —Preceding unsigned comment added by (talkcontribs)

I have updated and condensed the material on the failed Swpat Directive. IMO it is relevant because it has been arguably the major cockpit for discussion of the law in this area for the last three years. Those discussions are part of the history of the law in this area, and how it is perceived. The scars of the process - on history, on individuals, on the legislative process, on legal perception - remain. They are a relevant part of the current landscape.
The heat of the debate over the Directive also underscores the trickiness of the question of just what is the legal position on this subject in Europe at the moment; and the questionability of how far EPO case law can be taken as a proxy for the court-room law of the member states. -- Jheald 16:10, 15 February 2006 (UTC).
If people feel, that the EU directive on software patents which already has its far too long article needs another small article inside another article, I and others may be able to live with this kind of verbosity without censoring it. Quite honestly, the statements are purely personal opinions by no means verifiable and the last paragraph is totally speculative and more importantly beside the point. It´s like adding a medical opinion on moles to a biography of Cindy Crawford or vice versa depending on your field of interest. The community patent or the EPLA have a much wider scope and they would be an Earth quake in which the software patents are but a hush, or if you choose to believe the numbers in the introduction a 2% effect. It is actually anything but wise to connect the two issues of an elephant with an unbeloved, most unpopular mouse having passed Tom´s guts already twice (failed EPC reform and failed EU directive), at least from the point of view of the elephant. However, those who neither want mice nor elephants in their gardens will rejoice over their opponents lack of wisdom. They may even feel a sadistic joy in seeing the mouse lick its "scars" in the mourning discourse in this article. So let´s leave it for everybody´s comfort.
The German government which is the driving force behind patent policy these days issued a leaflet which said that the purpose of EPLA was legal certainty for software patents! Podmok 23:18, 31 May 2007 (UTC)

Original research (cont'd)[edit]

See WikiProject Fact and Reference Check#Help on whether or not the 2nd paragraph of the article is original research. Your opinion is welcome. --Edcolins 21:12, 15 March 2006 (UTC)

Disputed paragraph[edit]

While according to the European Patent Register in 2005, 85% of European patent applications in the field of electric digital data processing (G06F in the International Patent Classification (IPC)) did not give rise to a granted patent but were refused on various grounds or withdrawn by the applicant in response to the search report or substantive objections of the EPO, some of them have been granted by the EPO since the '80s. At present, applications in this field account for 7% of the total number of applications to the EPO, but only for about 2% of all granted patents.

This paragraph looks like original research to me. But even if it weren't, how is it relevant to the introductory paragraph of this article? Fagstein 03:00, 21 March 2006 (UTC)


According to a powerpoint presentation given at the WWW2006 conference in Edinburgh in May 2006 by Dr Clara Neppel, an EPO patent examiner in Munich,

  • over 50% of patent applications in the category cluster "computers" are ultimately refused or withdrawn.
  • over 70% of patent applications in the category field "electronic commerce" are ultimately refused or withdrawn

(slide 38 out of 40).

-- Jheald 19:55, 6 June 2006 (UTC).

Should this page be renamed[edit]

I'm aware that "software patents" are the hot topic at the moment, but given that I cannot think of a single case in recent EP history that JUST involved the computer program exclusion, might it be a good idea to rename this article "excluded subject matter under the European Patent Convention" or something. Making it more general to include the business method exclusion in particular (since you can then start talking about the 1-click application properly) would open it up. Different articles on the different exclusions would just contain so much overlapping subject matter as to be useless.

What do people think? I have no desire to make such a change myself because I don't know enough about the inner workings of Wikipedia to know how it might affect all the other articles that link to this one under its current title. Yes, I'm sorry I'm a relative newbie to Wiki. :)

GDallimore 01:20, 9 September 2006 (UTC)

Moving page[edit]

I decided to move the page to "Patents for computer-implemented inventions under the European Patent Convention" since it is an article about EPO practice and the EPO do not use the term "software patent". Then I realised the new title was a bit lengthy, repeats the word "patent" and is, in any event, largely about the examination of patent applications, so I'm going to move it again to simply: "Computer-implemented inventions under the European Patent Convention".

Sorry for confusion. Any comments? GDallimore 12:42, 15 February 2007 (UTC)

I think it's a bad idea. Jheald 19:54, 15 February 2007 (UTC)
GDallimore has added his comments interspersed with Jheald and has signed each point separately, also adding JHealds sig to his comments. GDallimore 10:33, 16 February 2007 (UTC)
  • All the other wikipedia articles on this topic are called "software patents in <jurisdiction x>" Jheald
  • That doesn't mean it is right to call this one by that title, or that any of the others are correct for that matter. I'll come to this later when talking about UK language. GDallimore
  • Almost universally, people talk about software patents rather than computer-implemented invention patents. For example, during the debate about what was (officially) supposed to be known as the CII directive, the name Commission internal market group called it in internal discussions was "the SoftPat directive"; the UK patent office's focus group was called the "Software patent focus group"; and the term "software patent directive" was universally used by politicians and the media. Jheald
  • The EPO do not use the term software patents. This is primarily an article about EPO practice. Therefore software patent is inappropriate in this connection. This is not an article about "computer-implemented invention patents" either. It is primarily an article about the practice for examining computer-implemented invention patent applications. I know you know the difference. In my time as a patent attorney, I've heard people say they want to "patent their trademark", "copyright their invention" etc etc. Newspapers and other supposedly reliable information sources make similar basic errors. The fact that, on the internet, many people use the term "software patent" is not a reason to use it incorrectly in connection with an article about EPO practice when examining applications. GDallimore
  • Interestingly, even lobbyists in Brussels supporting wider and more certain patentability now avoid the term "computer implemented invention". The idea which the term encouraged, that there was a significant difference between a software patent and a computer-implemented invention patent, is now seen as a notable public relations failure; something that particularly encouraged MEPs towards amendments to reduce software patentability. Jheald
  • This is not an article about lobbyists who will always bend words to suit their purposes. This is intended to be an intelligent, factual article about practice under the EPC. Go look at the CII Directive article or the software patent debate article if you want to write something detailed about lobbyists.GDallimore
  • UK judgements don't use the term "computer-implemented inventions" either. Typically the term used has been "software-related patents". It should be remembered that this article is about the law under the EPC across the whole of Europe, not specially software patents at the EPO. Jheald
  • It is about the EPC and the EPO are the main users of the EPC. The EPO language is therefore to be preferred. Also, note that the EPC does not mention "software", only "programs for computers". The UK article needs renaming to the appropriate language used in UK law/practice/decisions/judgments. I'm not yet sure what the best title is, but wanted to start here before making mass article movements.GDallimore
That is called positivism. Positivism is not NPOV. No, we don't have to use dogmatic definitions when we discuss the teachings of the roman catholic church. Officially the Algerian War was French police action, not war, because Algeria was regarded as a part of France. We don't have to adhere to the institutional language, in fact it should not be done. Podmok 23:22, 31 May 2007 (UTC)
  • An important topic the article discusses is potential software patents which are not granted. An important ground for not granting them is if they are "not considered to be inventions". Renaming the article computer-implemented inventions forecloses that discussion. Jheald
  • Agreed in part. The article needs a (brief) discussion of how lobbyists are upset about the EPO practice of referring to anything done with a computer as "an invention". I have, in fact, put such statements in many other articles, but hadn't got around to doing it in this one yet. This upset over EPO practice needs to then be countered (for NPOV) by highlighting that while "invention" has been downgraded, "inventive step" has been upgraded and the repercussions of this. However, that is not a reason to have an incorrect title.GDallimore
  • Finally, for many people "computer implemented invention" is seen as a POV slanted term, a piece of purpose-made Newspeak, precisely because it seems to shut down the Article 52 discussion. (Some would go further, and identify it as an attempted piece of spin, which failed; and which, like much failed spin, is now seen as an embarassment by both sides). Jheald
  • POV was my main reason for changing the title: Software patent is equally POV because the term has a pejorative connotation - look at your own website: the term software patent is used to mean a patent that should not have been granted. Since, therefore, it is impossible to avoid a title which is considered to be POV the one to choose must be the one that is used in actual practicing of the EPC. I would have no problem in principle with there being ANOTHER article called "software patents under the EPC", with a discussion of the problems that lobbyists have with EPO practice, but such an article is always going to be unecessary and stubbish in view of software patent and software patent debate. Also an article that is primarily about EPC practice cannot and should not have "software patents" in the title as that is promulgating the lobbyist lie that "computer-implemented invention" and "software patent" are synonyms when they quite blatantly aren't: patent means granted patent, invention is something for which a patent might be granted. Different things. No question.GDallimore
  • The defining aspect of the potential patents, in regard to the legal questions they raise, is that they relate to software. Jheald
  • You've said it youself: potential patents. Just because a lot of people don't know the difference between a patent and a patent application (cf the furore over the "emoticon patent" that was just a published patent application) is no reason to dumb down Wikipedia or to promote the confusion over the difference between the two.
For all these reasons I'm changing the article back, to the older simpler title. Jheald 19:54, 15 February 2007 (UTC)
For what it's worth, I second Jheald's arguments. Rl 21:59, 15 February 2007 (UTC)
I hope I've made my point. I won't change the title back yet, though.GDallimore 10:33, 16 February 2007 (UTC)
I didn't revert your initial move because I've seen your work and know you don't make such changes on a whim. And as expected, you make some good points. — That said, I'd be more inclined to agree with you if the EPO used the term "computer-implemented inventions" to differentiate inventions that make use of a computer from pure software patents. But they insist on using the same term for pure software patents, which to me indicates that they were less interested in precise terminology than in making the apparent contradiction between their practice and the European Patent Convention less obvious. It appears to be a manipulative use of language and only aiming to confuse. Rl 17:36, 16 February 2007 (UTC)

Moving page, new proposal...s[edit]

From the above discussion, I understand that neither "software patents" (pretty loaded, sounds like only granted patents are concerned, and does not meet WP:PRECISION) nor "computer-implemented inventions" (may sound doublespeak, and may be viewed as endorsing the EPO view of the EPC, while the expression is not in the EPC) are neutral. I would like to suggest a third option. I wouldn't use:

  • "Patentability of software-related inventions under (...)": patentability implies inventions, so this would be redundant - in addition the term "invention" is somewhat construed in a particular way in the EPC, i.e. a purely abstract invention is not considered to be an invention under the EPC (or at least under the current interpretation of the EPC by the Boards of Appeal),
  • "Patentability of computer-related inventions under (...)" : Art. 52(2) does not relate to computer-related hardware, but only computer-related software (incl. middelware, which is a type of software) - so too general,
  • "Patentability of computer programs under the European Patent Convention": nice, close to the legal text, but computer programs are all made to form software, and software are all made of computer programs, right? So, software is shorter and simpler for readers.

So, what about "Patentability of software under the European Patent Convention"? Seems neutral, to me, and precise and simple enough (please read WP:NC, if you have 5 minutes). --Edcolins 21:27, 16 February 2007 (UTC)

One thing I would say is if you're going to move this page then move the whole set: Patentability of software under U.S. Law, under UK law, under TRIPs etc etc.
I agree your title is more precise; but I think it's also less snappy and involving. So I think the series is better under the current title -- even if "Patentability of software (under legal regime xxx)" is what the reader should understand by it.
Is the term "software patents" loaded? I think it is actually helpful to the discussion to recognise the patents in Diamond vs Diehr, ABS braking system, or Koch & Sterzel as software patents -- patents on systems where the only new physical integer was new software.
Not all software patents relate to software running on PCs. (And not all of them would have been ruled out by the European Parliament). Jheald 00:27, 17 February 2007 (UTC)
How about simply Computer programs under the European Patent Convention, which is still snappy and includes nothing POV at all - no association of "inventions" with "software/computer programs" for example? The word "patent" is in the title once, so I think "patentability" can be lost at the beginning but Software under the European Patent Convention lacks rhythmic balance :) (apologies for my semi-poetic soul).
WP:NC suggests that an article should be named according to common usage rather than something which is necessarily factually correct, which does imply that "software patent" would be the most appropriate title. However, when looking at common usage, it's important to note what understanding comes with that usage. With this in mind, I'm keen to lose the "software patent" from the title of this article (and others) due to the way it is used by the general public. I've done my best on Wikipedia articles to make usage of the term NPOV and merely descriptive of a category of patent but the more articles, blogs, websites and rants I read, the more I come to the conclusion that the term is used pejoratively by the general public to mean "a patent that should not have been granted". I therefore do not think we can have the article with its current name because the general public will look at it and see "Patents that should not have been granted under the European Patent Convention". They may still see that with the newly proposed title, but that's due to their own belief of what should be patentable rather than the baggage that experience has shown me automatically comes with the term "software patent". GDallimore 18:10, 19 February 2007 (UTC)
I do feel quite strongly that we have a series of parallel articles on Wikipedia, essentially contrasting how the same family of patent applications might fare differently under different jurisdictions, and it's helpful to signpost that parallelism by giving the articles similar names. So it's useful to have "patent" or something similar in the shared bit of the title - otherwise something like Computer programs under U.S. law doesn't signal that it's about patents and patenting. Jheald 19:12, 19 February 2007 (UTC)
wrt your concern about "software patent" becoming so linked with "patent that should not be granted" - perhaps that is as good a reason as any that an article which includes talk about patents which arguably should be granted (on any grounds - moral, political, economic) probably should be titled "software patents..." :-) Jheald 19:20, 19 February 2007 (UTC)
PS. I don't think "computer" is necessarily the same as "PC", if that was your concern, James. Koch and Sterzel makes that clear, IIRC. GDallimore 18:11, 19 February 2007 (UTC)
That's a very important point -- that when we talk about "computer programs" in European patent law, we do not necessarily primarily mean programs for PCs; but in effect any software. (I believe the UK delegation specifically on the record waved away concerns that were put them about that drafting in the run up to 1973).
But that's maybe another reason that "computer programs..." isn't the right title for this article -- because I think, in the public mind, "software" is the more general term, and "computer programs" does suggest an association specifically with general-purpose computers. WP titles should try not to mislead... Jheald 19:12, 19 February 2007 (UTC)
Another possibility would be Software and the European Patent Convention, Computer programs and the European Patent Convention or Computer programs, software and the European Patent Convention (for precision and rhythmic balance, it seems the best ;) !!!)? Using "and" would make it much more neutral and maybe less subject to dispute (this is used elsewhere: see for instance Template:WMD, "By country" section). By the way, it is not absolutely necessary to have the same title for the series, the title should best fit the content and practice. The template I have just added would provide the naviguation tool in the series. --Edcolins 15:07, 20 February 2007 (UTC)


"should be construed using a Purposive Approach." -- not NPOV, it is an opinion Podmok 23:24, 31 May 2007 (UTC)

It is an opinion expressed by courts throughout Europe and is therefore a reliable and sourced opinion. Therefore it does not violate NPOV. If you can find a reliable and sourced opposing opinion, that can be added to offer the opposing view to help balance the article, but I doubt very much you'll find one. GDallimore (Talk) 13:08, 1 June 2007 (UTC)
But it is a normative statement. You misinterpret NPOV as positivism. What "should" be done always is normative.Podmok 12:33, 8 June 2007 (UTC)
I think it's you who is misinterpreting NPOV. It is not a requirement that an article should avoid taking one view or another, that is impossible, it is a requirement that articles should represent all significant views fairly, proportionately and without bias. The view that the EPC should be construed using a purposive apporach is a significant view with no significant opponents (although there are arguments that Art 52 can't be interpreted purposively since the purpose is not clear!). What this statement lacks is proper referencing. Please do not confuse WP:VERIFY with WP:NPOV the two as you also did here. Also, you might want to explain why you added a synthesis tag to the PCT software patent article since that makes no sense to me.
Lack of referencing is a big problem in most of the patent articles and sourcing them all is a mammoth task that is only being completed in a piecemeal manner. However, since this particular statement happens to be correctly representing the majority viewpoint, no-one has previously got around to questioning it. Tell you what, I'll add a source now... GDallimore (Talk) 13:57, 8 June 2007 (UTC)

misleading quote[edit]

"The European Patent Convention (EPC), Article 52, paragraph 2, excludes from patentability, in particular discoveries, scientific theories and mathematical methods; aesthetic creations; schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers; presentations of information. (emphasis added) Paragraph 3 then says: (3) The provisions of paragraph 2 shall exclude patentability of the subject-matter or activities referred to in that provision only to the extent to which a European patent application or European patent relates to such subject-matter or activities as such. (emphasis added)"

---> why not quote the EPC 52.2/3 in toto? Paragraph 2 does not exclude from patentability but explains what is in particular no invention. "(2) The following in particular shall not be regarded as inventions within the meaning of paragraph 1:", exclusion from patentability is only an effect.Podmok 12:30, 8 June 2007 (UTC)


more refs for "SA SAGEM c./ M. le directeur de l'INPI":

  • Une " méthode " ne peut être l'objet d'un dépôt de brevet: Aux termes des dispositions des articles L. 612-12-5° et L. 611-10- 2° du Code de la propriété intellectuelle, une demande de brevet dont l'objet ne peut manifestement pas être considéré comme une invention doit être rejetée, dès lors qu'il s'agit d'une méthode applicable dans le domaine des activités économiques ou bien lorsque l'invention n'est manifestement pas susceptible d'application industrielle. CA Paris, 4ème Ch. Sect. B, 10 janvier 2003, SA Sagem c/ M.le directeur de l'INPI pp.202-204 pp.790-791

—Preceding unsigned comment added by Swen (talkcontribs) 22:27, 29 April 2009 (UTC)

Found in par. 128: "Only one French case was brought to our attention, Informa v Catalina Marketing 21st June 2002 (3rd Chamber of the Cour Civile de Paris) at first instance and 28th April 2004 (4th Chamber of the Cour d’Appel de Paris). The patent claim was to an apparatus consisting of a conventional supermarket bar code reader and a printer arranged so that if a particular product was subject to a special offer by way of a discount coupon for later purchases, a coupon would be printed out. It was held that because there was hardware, the claim was not to a business method as such. Art.52(2) does not appear to have been considered by the Cour de Cassation." --Swen 21:10, 5 May 2009 (UTC) —Preceding unsigned comment added by Swen (talkcontribs)

Is software patentable in Europe or Not?[edit]

I realize that this is not a trivial question, but the article fails to address it adequately.

I tidied up the Article 52 section a bit. But the statement "a way of making a computer run faster or more efficiently" needs reference. More case law would be great. I note that none of the recent IPhone etc patent cases seem to invoke this clause. Tuntable (talk) 08:17, 16 April 2012 (UTC)

Have to conclude there is bias in this. There are lots of calls for "cite your sources" when the point is on one side, yet whoppers get by without any opposition, one of which is below:

"Like the other parts of the paragraph 2, computer programs are open to patenting to the extent that they provide a technical contribution to the prior art."

Essentially, that quote represents a conclusion from some form of original research on the part of the author. The specific stipulation that computer programs are "open to patenting to the extent that they provide a technical contribution to the prior art" is a conclusion, not a statement by the EPC, unless that complete and exact language can be shown to be used en-toto by the EPC.

The word "open" in the phrase "open to being patented..." is being used to open the gate to a conclusion disguised as a statement of uncontested fact. By adding the word "open" to the sentence, it seems to imply something normative about the ability for software to be patented, when in fact such patents are rare. It also implies that the condition of "being a technical contribution tot he prior art" is necessary AND sufficient for software to be patented in the EU . But that is exactly false. Absent further explication about what qualifies software for patents, if anything, it appears to be citing necessary and sufficient conditions. In reality it is only a necessary condition. Further, it is universally necessary for ALL kinds of patents, not specifically those surrounding computing. Worse, such a condition ("a technical contribution to prior art") is presumed to always be the case, just as it's presumed to always be the case that the patent is not a law of nature, and it is presumed to always be the case that the invention is not impossible to build i.e. a perpetual motion machine.

Absent further discussion I am removing the entire line; it adds nothing and projects a false state of software patents under the EPC.Jd111111 (talk) 18:39, 22 October 2012 (UTC)

It is not a false statement. It may not be your reading of the EPC, but it is the reading taken by the EPO and at least the courts in Britain. I think your problem is that you disagree with the way the law stands, but that does not mean there is a problem with the article. GDallimore (Talk) 20:31, 22 October 2012 (UTC)
I fully agree with GDallimore's reply above. In a nutshell, the short answer to the question “Is software patentable in Europe or not?” is: Software is not patentable if the focus is on the software itself (e.g., a software-automated business method itself). Software is patentable if the focus is on a novel improvement in a technical topic (e.g., a software algorithm, data-structure, and usage context that brings about faster access time). For example taking this to the Nth degree, I suspect (although this is largely unexplored in case law) that an otherwise unpatentable business method implemented in the logic gates of an ASIC would in fact be patentable in Europe because of the technical improvement of drastically faster speed of integrated-circuit hardware when compared with sequentially-executed instructions in software, assuming that the drastic speed improvement itself was valuable. Likewise for the software-like case of emulating that hardware in a kind of ‘software’ as loadable gate-array in an FPGA. And likewise, the otherwise-unpatentable business method might be patentable in Europe if cleverly implemented on a massive array of parallel sequential-instruction processors, if a drastic speed improvement without the need for custom hardware is effected. — (talk) 15:02, 13 January 2015 (UTC)

US Patent law[edit]

Total wikipedia amateur here, but I work in the patent industry:

"The position in Europe can be contrasted with that of other countries such as the USA and Australia. In these countries, the mere use of a computer is sufficient to make a business method patentable even if the computer is not being used in a novel or inventive way and only the underlying business method provides the patentable features."

This seems to be outdated given recent developments in US patent law. See

An abstract idea, such as a business method (financial hedging in the case of Alice) is not patentable simply because it is implemented on generic computer hardware. Rather, the claims must include limitations which amount to significantly more than the claimed abstract idea, and the generic computer components typically recited in those types of claims are no longer sufficient.

Again, I don't really know anything about how the wiki-editing community goes, but maybe someone should update the article. (talk) 20:23, 26 January 2016 (UTC)

Many thanks! I have just removed the outdated paragraph. --Edcolins (talk) 20:39, 26 January 2016 (UTC)