G 3/08 Observations - Your Last Chance

If you want to have your say about whether patents for computer-implemented inventions are a good thing or not, it's time to get your skates on.  The final date for filing observations (or amicus curiae briefs, if you prefer) on the EPO President's referral to the Enlarged Board of Appeal is 30 April 2009.  See the IPKat's previous notes on the subject here, here, here and here for more details, and the EPO website here for the official lowdown.

As of the date and time of this post, the following observations have been filed:
  • Bakels, R. (PDF, 1.9 MB)
  • Balos, I. (PDF, 360 KB)
  • Bayer, C. (PDF, 270 KB)
  • BIKT et al (PDF, 4 MB)
  • Combeau, J. (PDF, 630 KB)
  • Cowsley, C. (PDF, 230 KB)
  • De Keyzer (PDF, 84 KB)
  • Duhme,Torsten, Dr. (PDF, 1 MB)   
  • epi (PDF, 280 KB)
  • Gustavson, Stefan (PDF, 220 KB) 
  • Hallén, Jacob (PDF, 225 KB)
  • IBM (PDF, 970 KB)
  • intellect (PDF, 585 KB)
  • Lawrence, William (PDF, 915 KB)
  • Philips (PDF, 390 KB)
  • Piratenpartei (PDF, 400 KB)
  • Piratpartiet (PDF, 160 KB)
  • ScriptumLibre (PDF, 410 KB)
  • Sterckx, Sigrid, Prof. Dr. (PDF, 2.7 MB)
  • Straus, Joseph, Prof. Dr. (PDF, 1.2 MB)
  • Sylvestris, T. (PDF, 525 KB)
  • Thum, Simon (PDF, 170 KB)
  • Wessel, A. (PDF, 81 KB)

  • Of these, some are very interesting, some are fairly pedestrian, some are horribly wordy and full of legal 'analysis', footnotes and references, some are fairly predictable ill-informed anti-patent rants, while others are merely irrelevant.  The IPKat (who still can't see what on Earth TRIPs has to do with the EPO) will let his readers decide which are which.

    One brief in particular, submitted by the IPKat's maverick sidekick Tufty Sylvestris (a.k.a. Tufty the Cat), argues that the referral is inadmissible because it does not comply with Article 112(1)(b) EPC.  This view, although in a small minority, appears to be shared by at least one other observer.  The IPKat prefers to remain on the fence, largely due to being completely bamboozled by all that legalspeak.  He hopes, however, that the Enlarged Board, whichever way they decide to go, will come up with some reasoning that we can all understand, instead of trying to impress us with their footnoting skills.  

    More legalspeak here. Doublespeak here.
    G 3/08 Observations - Your Last Chance G 3/08 Observations - Your Last Chance Reviewed by David Pearce on Wednesday, April 29, 2009 Rating: 5


    1. Unlike the post says, this is *not* an opportunity "to have your say about whether patents for computer-implemented inventions are a good thing or not." Actually, the questions referred to the EPO Enlarged Board of Appeal *assume* patentability of "CIIs" in certain cases, and only try to clarify under what conditions. Could the EBoA say "w8 a minute, we don't endorse the presumptions"?
      perhaps it is fortunate that "patent pope" prof. Joseph Straus argues that the EBoA is not admissible in this case!

    2. The EBA could so easily have agreed with a lone academic German line from Prof Dr Dr Straus, but now there's the problem of an English voice urging the same line (if my superficial skimming gave me the right impression). Politically, can the EBA adopt a line urged from England? Hardly. If Tufty had really wanted the Straus line to be followed, she/he should have remained silent, written from an address on the mainland or, from the plumb-centre English address, argued exactly the opposite of the Straus line.

    3. What is the objection against briefs "wordy" briefs "full of legal 'analysis', footnotes and references"?

      It is a fact that the Boards of Appeal of the EPO over the years developed very complicated (and inconsistent) rules, all based on this dreadful technology criterion which was identified as a "restatement of the problem in more imprecise terminology" by a UK court (CFPH case).

      In order to convince the present Enlarged Board that this is a crazy approach, strong (legal!) arguments are needed.

    4. IPKat can't see what TRIPs has to do with the EPO.

      The answer is simple. All EPO member states (except Monaco) are WTO members. If the EPO does not comply with TRIPS, the member states are obliged to grant the TRIPS-required patents that the EPO does not grant. Then we are back in the pre-EPO era, when separate patent applications had to be filed in all countries. Logically, the EPO member states will urge the EPO via their Administrative Council representation to grant all patents that the member states are obliged to grant.

    5. Amazingly well argued amicus curiae brief by Tufty Sylvestris.
      The mice diet of your cat appears to do wonders, David!

    6. If professor Joseph Straus, once nicknamed the patent pope from Munich, says that the referral is inadmissible, the Enlarged Board of Appeal is likely to follow him. Joseph Straus is a very influential man, and he would not waste his time (and reputation!) if he would not have inside information that this advice is likely to be accepted.

      But perhaps the Enlarged Board should indeed refrain from making decisions that are effectively political in nature. Courts are only allowed to make new law to some extent. They should not replace political decision making.

    7. Interestingly there is the feeling between non UK europeans that the UK has too much influence in the EPO, while the opposite feeling seems to exist between UK europeans. This is probably the best sign that the EPO is a reasonable compromise between the continental and common law approach.

    8. An incredible and dangerous allegation made by Mr Bakels questioning the independence of the EBoA! Those who have experience with the Boards of Appeal know better and hopefully ignore also the further fully superfluous remarks.

    9. For those who can read German, there is an interesting article in GRUR Int. 2008, vol 8-9, pp 713, on the admissibility of referrals to the EBoA considering the judicial nature of the EBoA.

    10. I intended for a long time to submit some observations. Ultimately, for the reasons suggested by Bakels, I decided not to. The questions to the Enlarged Board are based on two presumptions which I do not agree with and this, to my mind, makes the questions pointless.

      Firstly, there is the presumption that a claim is either excluded or is not excluded as a whole. There is no opportunity for some of the claimed features to be excluded. No such presumption exists for novelty - a claim can lack novelty as a whole, but the usual situation is that some of the claimed features lack novelty and some don't.

      This makes questions 1 and 2 impossible to answer in a sensible way, in my view.

      The second presumption is that, given that it is a requirement that a claim have technical character to be patentable, a claim which does have technical character is not excluded. Am I the only person to see the logical fallacy in that? Art 52(2) does not mention technical character such that a natural intepretation is that it carves out an area of excluded subject matter which may include fields that otherwise have technical character. Analogously, just because a claim has technical character does not mean that the novelty requirement can be overlooked.

      This makes it impossible to answer questions 3 and 4 in any sensible way, in my view.

      I think the whole exercise is a PR stunt to "prove" to their detractors that they only grant proper patents. If the EPO were really serious about analysing the scope of the computer program exclusion, they would look at some of the assumptions underlying the current case law. Thing is, there's no way to do that if all of the case law makes the same assumptions.

      On the positive side, if the EBA make a hash of it, maybe the House of Lords will take on the next decision that comes their way. The HoL seems to be one of the few venues left with the opportunity to start again from scratch.

    11. Joseph Straus is a very influential man, and he would not waste his time (and reputation!) if he would not have inside information that this advice is likely to be accepted.Come on, that's just a rather ugly accusation for which there is probably no concrete indication at all.

      The information he has is most likely exactly the sources he refers to in his brief. Why would he even bother if he knew the EBoA had made up their mind already? Just to "score a point"?

      That said, I certainly do not exclude that at least some of the questions will be declared inadmissible.

    12. If someone reads in my comments an allegation against the independence of the Enlarged Board of Appeal, I haven't expressed myself clearly enough, and I ask for apologies.

      I only suggested that prof. Straus may have talked to EBoA people who may have made comments like "we are still not sure whether we should be admissible".

      And I argued that some decisions may be considered inherently political. To some extent, that is a risk of any (high) court.

      Finally: if you blame me personally, please be fair and do not post an anonymous comment!

    13. @Gerontius: what do you mean by "no such presumption exists for novelty"? To my understanding, a claim either lacks novelty in the sense of Article 54 EPC, or it does not. You can talk about individual claim features that are known (just as you can talk about individual claim features that are excluded), but in a legal sense it is the claimed combination of features that is either novel as a whole or not.

      A single "excluded" feature does not render the claim excluded. A single "old" feature does not render the claim old. I don't really see a difference here; your first presumption certainly seems valid for novelty.

      Anyway, I think the president could not avoid making the presumption that EPO case law, as far as it is not considered to diverge, is correct. EPO case law seems to be consistent in equating excludedness to lack of technicality. The president has no power to ask the EBoA directly whether this is correct. However, if the EBoA is of the view that this basic presumption of EPO case law is wrong, so that basically none of the questions make much sense, then the EBoA is probably allowed to say so.

    14. If Joseph Straus is the patent pope, does this make him infallible? I have had a quick scan of his submission and feel that he is barking up the wrong tree with his "two boards" idea. I would have thought that foe each appeal, the boards are different even if they have the same composition. The board is formed to consider a particular appeal and therefore two differnt appeals have two different boards.

    15. The most common misconception: that all these submitters are so confident that it was an Amicus Curiae process while it is in fact about "written statements" which could be anything you want, for instance poems about the further technical effects.

      "Amicus Curiae" make no sense. The EBoA is no Court (yes, I know in the UK they don't understand the EPO and assume it acts almost like a court, stupid Brits)!! Without a Court there is no reason to write Amicus letters. And by the way, who wants to make "friends" with them?

      It is about "written statements". When you are a law student from Krakow and you are looking for clients why not inform the enlarged board that you are interested in the subject? The EPO may even invite you, nice dinners, you can join forces with Granstad, become the one and only EPO sponsored patent activist and call for harmonisation of the EPO practice with the US. More patents mean more business, more investment, more R&D. No R&D without a euro-cafc.

      Amigo letters are about job applications. Hi, I am your new friend (or: I am your old forgotten friend, let's better do business), please feed me.

      Hi, we found the last resort policy laundry argument in the basement: We need patents. TRIPS 27, inventions in all fields of technology patentable, therefore software qualifies as an invention. q.e.d.

      Or, we are a large client, let's have a UK agent write some stuff up who shares the German passion for "technical" aspects. Our little patent porn contribution. Open patent porn source.

      Or, I am software developer and I think I have some knowledge in the field cause you want to get the system right for me, no?

      Comes the academic, look what I know and teach my students. Cool stuff. This could be a question and that? Unfortunately you are unable to consider that. No problem, we need to fix the invention concept fundamentally somehow? Why don't you remove the complicated testing...

    16. In G 1/04, the EBoA spends some time to argue that the referral is admissible even though decision T 385/86 and T 964/99 both originate from Technical Board of Appeal 3.4.1. So the idea that two different appeals are by definition taken by two (different) boards does not seem to be shared by the EBoA.

      However, it seems to me that the "in addition" provides a (second) independent reason why the "two boards" criterion is complied with. So it should be sufficient if the decisions were taken by two Boards in "completely" different compositions. This does not give an answer for the case that the Boards had "somewhat" different compositions.

      Prof. Strauss seems to argue that the "in addition" merely strengthens the first reason given. I can't possibly see how this could be true. Either the "completely different composition" is irrelevant for compliance with the "two boards" criterion, or it is sufficient. The fact that the EBoA in G 1/04 included the "in addition" rules out that it is irrelevant.

      The argument that Art. 112(1)(b) requires a real divergence and not merely a consistent change (from "contribution approach" to the Hitachi-approach) seems a lot stronger to me. However, that would seem to imply that questions can only be referred after an extended period of legal uncertainty in which the president has to wait and see if the regular Boards can figure something out themselves.

    17. Klaus, at the very least you could try to exhibit some awareness that the EBoA cannot be equated to the EPO. The EBoA certainly is a court. The EPO certainly is not.

    18. The German professor Straus takes it far too serious. The process is political. No one cares for the case law inconsistency. The German master mind now compiles all sources to find out what is really inconsistent. Does not matter. No one actually cares. It is about the mysterious "further political effect", the outcome. Brimelow does not think German. She is a politician. More a herrenmensch than a mensch. She does not believe in her own propaganda and does not take the EPO case law serious. Let's face it: The referral g308 is about policy making with the EBoA as her puppets. How can you influence a political argument with legal research, Mr. Straus? Or was he asked to open the backdoor for his EPO buddies after contracting states where less ethusiastic?

    19. Wow, the amicus curiae brief by the "Institut zum Schutze geistigen Eigentums" (Institute for protection of intellectual property) is somewhat mind boggling. The idea of the brief seems to be that the questions should be answered in a way that strikes an optimal balance between extending patent protection as much as possible while creating a sufficient smoke screen for patent critics. As a goal of the institute that's fair enough, but do they really think that the EBoA is a purely political body that shares their agenda? (Admittedly, the briefs that simply state that software patents are no good aren't any better.)

      "Note: We want to mention, that there are practical problems with the "further technical effect" when the concept is misunderstood. As explained by the EPO in 2000 this concept should not be used at all in the examination but only as a "de-facto" deletion of the exclusions under Art.52(2) and (3). But we think that these problems can be solved by communicating the concept to examiners and local patent offices."

      So "further technical effect" was meant only as a clever trick to effectively delete Art. 52(2) and (3), and now the EPO should inform its examiners accordingly?

      And what on earth was meant by page 5, second paragraph...

    20. @Klaus: Remember that the suggestion by a British court to invoke an EBoA on this issue was initially dismissed by Pompidou, the previous EPO president, because art. 112 requirements were not met. And when Brimelow made the referral late 2008, she was criticized for her interpretation of this provision and/or the facts.

      Yes, Brimelow may be a politican, but the EBoA is a group of (technical) "judges". I guess they will be all too happy to have a reason to escape from Brimelows obvious attempt to abuse them for political purposes (= force UK courts to think German). Incidentally, Brimelow may not like that personally, but I assume her assistants who wrote the referral letters wanted that!

      Incidentally, Joseph Straus (with a single "s" at the end) is not German. He holds an Italian passport and was educated in former Yugoslavia.

    21. @Anonymous 9:39 PM
      The Enlarged Board of Appeal is part of the EPO, see art. 22(g) EPC. Amazing, but true. And the EPO is working to improve the independence of the members of their Boards of Appeal, see "legislative initiatives" on the EPO website. Apparently, there is something to improve!

      Indeed it is questionable whether (enlarged) boards of appeal may be seen as independent courts under art. 6 ECHR. Actually I know of one BoA case that was appealed at an external court. Unfortunately, that case was eventually settled out of court.

      And in the present (G 3/08) case, unfortunately there are no parties who can appeal at a "real" court.

    22. "Incidentally, Joseph Straus (with a single "s" at the end) is not German. He holds an Italian passport and was educated in former Yugoslavia."

      Oh, I didn't know that, but he is the top legal professor in Germany together with Krasser, no?

      Straus = "Max-Planck-Institut für Geistiges Eigentum, Wettbewerbs- und Steuerrecht" and he writes German.

      A representative of the "German holy grail" of patenting.

      In Germany some people argue how Bavaria is not really German not to mention the francons in Bavaria.

    23. "Klaus, at the very least you could try to exhibit some awareness that the EBoA cannot be equated to the EPO. The EBoA certainly is a court. The EPO certainly is not."

      You can claim that the sun turns around Earth but still you ought to provide sufficient evidence for your claims.

      Actually it is not a Court but an administrative institutoon of the EPO. The confusion again is very British because there in the UK we find a "Court of Appeal". Enlarged "Board of Appeal" sound similar but is not. It is all administrative and the power stays in the hands of the administrative council. While Mr. Messerli was replaced by a professional judge the members do not act in a judicial role when they serve in the enlarged 'Board of Appeal'. For instance they do not have to qualify for judicial office. The technical members clearly do not.

      Basically the EBoA was not set up for substantive questions, look in the EBoA case law history. It is abused to act like a Court and triggered with political matters.

      In a continental tradition Courts are not interviewed by their boss who asks them to provide a legal opinion. The EBoA is a Committee of administrative experts and they write up an "opinion", it is not a ruling or decision.

      If the BoA system was a Court (which needs to be proven), then why do we need EPLA? Why the other reforms to get it replaced by an actual Court of Appeal?

    24. Well argued as it is, I wonder if the EBA will take account of Tufty's observations. Maybe they will treat them as anonymous (clearly having been typed by an unidentified one of Tufty's human minions) which I think would not normally be accepted by the EBA (G 1/03 summary of facts and submissions section VI (3) - see below).

      (3) Disregarded statements
      One statement was filed anonymously and another one was received after the oral proceedings at the end of
      which the debate had been closed. Neither has been taken into account by the Enlarged Board.

    25. I would like to share my opinion of a person who is involved in witting software. When I observe all these discussions and progress in the matter of so called CII I do not see anything good for small enterprises and start-ups which want to produce some usable goods for the others. I will try to explain why.

      The current doctrine of patenting is simply not suitable for patenting of software because it bases on the assumption that any combination or mix of technical and non-technical features that interact bitwise may be patentable if it is novel (has an inventive step). To create a novel 'invention' it is now sufficiently to present any usable combination of functions to achieve patented 'invention' (because PO does not consider in practice a degree of inventiveness - it is not able to do it appropriately os it is not bothering of it). In the result of such assumptions any instance of software can be patented because any software is a combination of different functions. It become technical just by writing that it is stored on disk or run on computer (from this point of view any book describing code or game or discovery should be a patentable invention also because paper is also storage). From my perspective it looks like nightmare of mad man (but it is a reality). If is going about abstract things like algorithms a person who simply write some generator of software (creating even random but runnable code) will in fact build a generator of patent applications (just by converting code to text). I doubt that this was assumption for introduction of patent system promoting inventions. What is also insane in the situation, the current system is not ready (technically) for seeking for prior arts that are extensively present in IT since 50's (scientific publications, implementations, journals, software code and associated documentation available publicly in the internet) and it results in granting patents that simply repatent many ideas and solutions existing for years (sometimes also being trivial). In evaluation of each software patents should be considered tens of terabytes of existing data in state of art.

      Practically (at the end we all live in the real world) it leads to situation where:

      1 – Current systems if opened on CII will favour a deluge of low quality of patents. There is no technically possible to control quality of patents. The inventive step requirement in the context of software complexity is always fulfilled.

      2 - The only cutting mechanism is the patent fee that limits patenting. It hurts directly SME and start-ups ant not having funds for extensive patenting of CII (event if they could in the current situation because of not proper recognition of inventive step in software). It is obvious that if you are a small company that has profit of $50000 then you will patent much less that in the case of company with profit $5000000. It will be a snowball effect promoting big players and in fact dismantling small enterprises. Handling patents in the court cost so much that a small company will always bankrupt trying to clean up from dozens of trivial patents brought to the court by bigger competitor (none insurance will help here). The counter strike with a few patents of the small company with not influential lawyers will be mostly a hopeless action deemed on failure (justice is blind – at least since ancient Greece).

      3 – By developing software it is easily 'to create' inventions but in the current situation it will be very easily to infringe existing patent. The process of creating software bases on using libraries and producing some added value. It is iterative by nature. It makes production of new software cheaper because of sharing costs filling market niches. Merging of existing solution is a core part of development decreasing costs and causing software as cheap as possible (natural goal of each economy). The proposed system will cause a big problem here because each way of usage of any library and the library itself may by patented many times (flexibility of the abstract apace in non countable by nature). It causes always a multiple patent infringements in any simple written software (it is worth to note that used software contains usually hundred of functions – so probably would infringe many thousands of patents).
      It is very easy to infringe in writing software because it is not just like writing chemical compound. Existence of chemical compound is constrained by laws of physic, so the number of combinations is limited by nature and some of them cannot be made and cannot exist even if something has been written on paper. The space of combination in software is not limited by nature of physic laws because it is abstract discipline. The only constraint for that are computer capabilities. But they will not play role in evaluation of inventiveness when patent is granted.

      4 – Any insurance will not help in software industry. If the big player in software industry will decide to go after a smaller competitor with the innovative product it will be easily able to do it because of scale of possible patenting, complexity of software products, and fuzzy scope of patents on abstract 'things' or methods and snowball effect. It will be a time to forget about competitiveness on the market, the proposed system will promote rich players with support of the best lawyers. In fact the fines are disproportionately high to the patent cost and small enterprise profit from the software. The insurance against thousands of software patents that may be infringed will be probably not affordable for any SME – so who will pay for it (as always taxpayers :) )? The risk and scale of fines for infringements for SME (soon maybe also with criminal charges) will be too great in software industry to produce anything by themselves alone (that is the goal of the changes?). The most of IT companies have in this situation three options: become a 'reseller' ans servant of big player and leave R&D and production OR become a producer of 'paperware' – dozens trivial patents that could be used against big players by combining their software with some 'new' (even trivial) functions, so patent troll OR close enterprise.

      Above conclusions brought me to opinion that proponents of the software patenting do not thing in fact about SME and economy but only about their own prosperity at the cost of SME and economy. SME definitely are not planned to be winners under current proposals and actions. But the decision makers will decide if they want to maintain a rich world of currently available software with SME's filling all niches or a few big companies dominating market with some set of servants and many gaps in the IT usage that will be filled 20 years or more later because of patent risk.

      The economy has survived dot.com bubble, now the economy is trying to survive mortgage bubble. Now some try to construct IPR bubble. Who is finally paying for that – taxpayers and mainly people who want to create and sell something usable and earn in their own SME's. The law that artificially creates obstacles and then introduces payments for insurance against these obstacles will not help SMEs.

      People who run SME and their employees are main customers of products sold by big companies (middle class). If they do not have occasion to earn sufficiently they will do not buy. The economy is a matter of bottom-up growth (of all parts of society). Now this growth has been abused by overregulation at every level. Any wunderwaffe in the form of IPR and further regulations will not help in resolving current problems in economy. Just does the situation worse. A man that wants to earn and do something good for the economy and society has too many problems and risks with current regulations when running its business. Adding more will not help and proposed regulations will make painful times in software industry.

    26. @Anonymous of 12:43 AM:
      I know that the EBoA is part of the EPO. That does not mean the EBoA can be equated to the EPO, which is what I wrote.

      The European Commission of Human Rights had this to say in Lenzing v. the United Kingdom (Appl. No. 38817/97):
      "In addition, as the High Court in the United Kingdom found, the Boards of Appeal of the EPO provide a means of judicial review. Thus, Article 21 provides for an appeals procedure which includes the Board of Appeal and an Enlarged Board of Appeal. The members of these Boards are independent of the parties and of the decision of the division appealed from, have tenure and there must always be one legally qualified member of the Board. The Boards have powers to obtain sworn evidence and must give written decisions containing reasons. Further, the members of the Boards are not subject to any instructions from the President or anyone else in their work (Article 23). These procedures set up a form of "equivalent protection" within the meaning of the Convention case-law (see No. 13258/87 and No. 21090/92, both cited above)."

      I'm not British and rather familiar with the continental tradition. However, you seem to have a very different continental tradition in mind than the one I know. It is not uncommon for continental courts to have non-legally qualified judges. This is especially true for courts deciding in specialised matters, such as patent courts. According the www.bpatg.de, "Zu seinen 117 Berufsrichtern gehören 58 Richter mit einer naturwissenschaftlichen Ausbildung (so genannte technische Richter)." Is the Bundespatentgericht not a Gericht? Please read § 67 PatG. Should I now provide evidence that Germany is part of continental Europe?

      It is interesting that you seem to claim that the EBoA was not set up for giving opinions on referrals. Must Art. 112(1)(b) EPC be considered a typo?

      Indeed G 3/08 will be an opinion and not a decision, as it did not arise out of an actual case but from a referral by the president. It is rather similar to opinions delivered by the ECJ (art. 300 EC Treaty) and the ICJ (art. 65 ICJ Statute). Are those not courts either?

      Klaus: "If the BoA system was a Court (which needs to be proven), then why do we need EPLA?"

      Please study first the European patent system. The EPC only plays a role in the granting procedure. EPLA is intended to provide a uniform European patent litigation procedure.

      Klaus: "Why the other reforms to get it replaced by an actual Court of Appeal?"

      See the EPO website. Improvements are possible.

    27. @Anonymous of 4:36 PM:
      "The current doctrine of patenting is simply not suitable for patenting of software because it bases on the assumption that any combination or mix of technical and non-technical features that interact bitwise may be patentable if it is novel (has an inventive step)."

      I'm afraid that in this single sentence you mix up all the basic concepts of patent law. Now I understand that you are not an expert on patent law by profession, but it is really not very useful to try to draw conclusions before you have a good grap of what you are trying to draw conclusions from.

      "To create a novel 'invention' it is now sufficiently to present any usable combination of functions to achieve patented 'invention' (because PO does not consider in practice a degree of inventiveness - it is not able to do it appropriately os it is not bothering of it)."

      If what you mean is that the EPO equates "inventive" to "novel", then that is absolute nonsense. You are most likely blindly accepting whatever FFII tells you without doing your own research. It is *never* good to blindly follow what you are being told.

      You call the system "insane". Maybe you should first ask yourself how likely it is that the European patent system really is "insane" compared to how likely it is that an anti-patent lobby group would like you to blindly believe that the system is "insane".

      If you would just look up some actual decisions on software patents, you would see that the great majority of them hinge on the requirement of inventive step, and that more often than not this requirement is found not to be fulfilled. This is because the non-technical aspects are essentially ignored when assessing inventive step. Thus, a cdrom with a word processing application is technical and hence an "invention", but the application does likely not provide a technical effect (apart from very trivial effects such as performing I/O) so won't render a known cdrom inventive.

    28. Since both 30.04 and 01.05 are official EPO holidays, should 04.05 not be the last day for submissions?

    29. No one says, you can't submit later.

      The board just asked for written statements by the end of april for its own convenience

      I don't think you are right to argue that this is a Court.

      Also decisions of the Technical boards usually don't meet judicial standards.

      What you quote is no primary source but political.

      The EBoA belongs to the EPO.

      You may wonder what makes a Court a Court. So far no one has argued that the EBoA should be a Court and that members of the Committee should be called judges. Sure, that could be discussed next time in Venice.

    30. "You call the system "insane"."

      I said it in the context of feasibility of prior arts seeking. I do not see it possible with aprioriate quality at the current level of data mining techniques. They may find correlations between words but hardly semantic dependencies.
      I do not see software at the same level of complexity as chemistry. Later is simpler and possibilities are constrained by natural laws. Software are pure equations (infinite space of solutions) and common computer parts steering. At the level above any OS practicaly there are only equations because hardware is recognized in abstract way.

      "Thus, a cdrom with a word processing application is technical and hence an "invention", but the application does likely not provide a technical effect (apart from very trivial effects such as performing I/O) so won't render a known cdrom inventive."

      Ok. Currently, maybe patent on software on CR-ROM may be difficult to receive in EU. But there are plans to adjust to US so soon it might be not so obvious.

      "You are most likely blindly accepting whatever FFII tells you without doing your own research. It is *never* good to blindly follow what you are being told."

      I think that this is far going assumption. It is sufficient to read pdf included in this blog entry and know how software is written to come to some conclusions. It is very bad that about future of IT are deciding people who never wrote a line of code. Maybe they are great in other disciplines but I hardly belive in their omniscience.

    31. "Thus, a cdrom with a word processing application is technical and hence an "invention", but the application does likely not provide a technical effect (apart from very trivial effects such as performing I/O) so won't render a known cdrom inventive."

      I would like to add just one parallel. In the same manner CD-ROM containing music causes CD-PLAYER to emmit sound waves. Thus a music recorded on the CD-ROM is also patentable because it is in the same way technical as bussines software, games or music (abstracting from hardware - like a piece of code in Java or C#) executed, played on computer.

      I my opinion the common hardware-like patent concept just does not fit to the complexity and needs of software creators (especially for those programmers who creates bussines software abstracting from technology details e.g. computer drivers programming). Copyright, design rights are much better for complex abstract world.
      IT is now mature industry (almost sixty years of tradition sice the first computer). I am involved in programming almost 15 years. In fact writing software now is as difficult as writing books. You just sit, having a nice idea and write some instructions at your computer. Jump there, print something if some variable is greather than zero etc. Nothing more - simple as composing some scores of music. Everything in human like language like Java. You just using libraries prepared by vendors usually do not be worring about technical details how it works when some pixels are displayed in required order on the LCD.
      An as programmer who wants to create software that will be usable or funny for people I am simply worried about these ideas fully implementation. The may forbid my practice and simply leave out of job (after 15 years). A then what? - the only solution is to become a patent lawyer in CII :) :) :)

      Have a nice holidays

    32. @Klaus:
      It is interesting that you are demanding evidence, when you yourself have no better than "I don't think you are right" and "Also decisions of the Technical boards usually don't meet judicial standards". What is your basis for that?

      What do you mean by "what you quote is no primary source but political"? The EPC, the EC Treaty, the ICJ Statute and the Patentgesetz are primary sources, no? And I would not call a judgment of the European Commission of Human Rights a political document.

      You write "So far no one has argued that the EBoA should be a Court". What are we having this discussion for, then? The facts are that the Boards of Appeal including the EBoA look like courts, talk like courts and act like courts. They are not called "courts", but neither is the judicial part of the House of Lords in the UK, or that of the French Conseil d'État, which both most definitively are courts for all practical purposes. The same holds for the EBoA, unless you can point out some characteristic common to all courts, but absent in the EBoA.

      Maybe we can leave it at this, and conclude that your "most common misconception" was a misconception on your side? Or if you react, please come with some real arguments and something to back them up.

    33. @Anonymous of 8:01 AM:
      I'm not aware of any plans to adjust the EPO practice to US practice. Even the submissions from "big industry" are not asking for that. And certainly G 3/08 will not adapt US practice, since the EBoA can't change the EPC.

      Insofar as software consists of pure equations, a patent is very unlikely to be (validly) granted. However, in a lot of cases software is more than pure equations. For example, if you structure your data or program in a way that can be shown to improve cache efficiency (and not by merely throwing away most of your data), this might give a "further technical effect". The question is then if the prior art already suggests to structure the data or program in that way in order to improve cache efficiency. If the prior art does, I think this will usually be disclosed in some paper or on some internet site and not only be deeply hidden in (open) source code. However, in principle well-commented source code is perfectly searchable using existing search techniques. What has to be searched for are not the "pure equations" but the application of the cache efficiency improvement technique.

      I think of more concern is what kind of effects should be considered "technical". Case law is not consistent. For example, many decisions state that effects on the human mind are not technical, but a few decisions bluntly state that improving amusement is a technical effect. Another difficulty is the distinction between "this program is faster because it does less", "this program is faster because its algorithm is O(n) instead of O(n^2)" and "this program is faster because it specifically improves how the hardware parts interoperate".

    34. @Anonymous of 3:22PM:
      Yes, a CD-ROM containing music is just as much an "invention" as a CD-ROM containing a word processing application.

      A patent can only be granted on inventions that are new, inventive, and industrially applicable. When assessing the inventive step of a CD-ROM containing music, the features relating to the music are not taken into account. Since CD-ROMs are well-known, a CD-ROM with music will never satisfy the requirement of inventive step. So no patent can be granted. However, it doesn't seem fair to say that an optical disk does not qualify as an "invention" (even before you start to look at novelty and inventive step).

      Now suppose a medical researcher discovers that particular sound waves cure cancer in a medically verifiable way. Should a patent be granted on a CD-ROM with those particular sound waves? Or should a patent only be granted on a CD-player with the CD-ROM loaded, or on a CD-player that is playing the CD-ROM? Or should a patent not be granted at all, because the invention resides in sound data?

    35. One begins to see something rather useful emerging from the President's initiative. Not what many might have hoped for, namely, clarification. Rather, the open house on observations from members of the public being taken up with enthusiasm, and showing that the EPO has already got it pretty much dead right.

    36. "Insofar as software consists of pure equations, a patent is very unlikely to be (validly) granted..."

      That is a nice acknowledgement. I hope that this practice will survive current and further situations.

      "For example, if you structure your data or program in a way that can be shown to improve cache efficiency..."

      My opinion here.
      Well cache may be a physical object - a piece of memory wired to disk as device storing data before you write them on disk. If cache algorithm is used with conjunction with this device and with clear purpose of speeding access to data from device then I think that patent is possible but only on connected pair disk memory device+algorithm (whole). You simply have new better device. It also should be proved that disk memory device+algorithm working effect in physical activity resulting in better accessing data and it presents an advance in technology of writing data on disk (so in physical activity).

      But the same cache algorithm (set of equations) also can handle a data in some bigger algorithm (software) used in processing abstract data on PC. For example 'Last Recently Used' algorithm may be implemented in array of data in Java to efficient (from memory point of view) fetch some data from memory of a computer in some loop. In this case we have still equation, maybe connected with other biger equations but an equation. It should be not consider as patentable invention but copyrightable as a content prepared for computer (like new music using some known motifs prepared for CD-PLAYER). It does nothing more than data processing here and such software output is a pure information. So with conjunction with general computer it also should not be patentable (maybe steering of LCD or sound card is technical but it is done at the level of computer OS drivers and device drivers could be patentable with devices as coupled novel physical entities installed on PC). PC simply is used here as content (here software) a playing device (so as CD-PLAYER or guitar playing music scores). So the PC with drivers (low level software accessing specific devices and presenting abstract API) should be patentable but a content played on it (seq of equations using computer drivers API) should be copyrightable (it is mere computer usage).

      So the LRU itself never should be patented alone as data processing algorithm (so as complex equation but still the equation). But with conjunction with some device may create not existing earlier new technical whole - then it should be patentable in my opinion - simply as a new novel thing using algorithm to do work in a better way. If you use PC (old known thing) with software you should be granted patent only when a you prove that in the result you have something physical what is new - for example signal steering new kind of medical equipment. Then you should be able to get patent on software+PC+medical equipment because software on PC works as driver for particular medical device.
      But displaying pictures on LCD at different position with different text may not be patentable - it is mere use of PC. So program running on computer and simply processing data like words processing tool should not be patentable (output is only pure information presented and such software uses abstract API supplied by OS and do not solve any technical problems related to computer working). So mere usage of already patented computer in form of prepared and played content (here software) should not be a subject of patent but copyright.
      So to patent software with a device you should prove that software is a form of driver steering this particular device and the whole fulfils rest of conditions for a patent (novelty, inventiveness).

      And finally abstract computing machine + abstract algorithm combination should never be patentable. It creates single broad virtual entity because 'abstract computing machine' is simply also algorithm (see abstract Turing machine).

      "Now suppose a medical researcher discovers that particular sound waves cure cancer in a medically verifiable way. Should a patent be granted on a CD-ROM with those particular sound waves? Or should a patent only be granted on a CD-player with the CD-ROM loaded, or on a CD-player that is playing the CD-ROM? Or should a patent not be granted at all, because the invention resides in sound data?"

      Well, I personally think that it should be patentable not as "a music on CD-ROM with CD-PLAYER" but as a "device for curing cancer". And I would put following conditions here:
      - that music (its instance(s)) on CD-ROM have to be strictly described in detail. I could not be just claimed as "general CD-PLAYER with any kind of music curing cancer" because it would embrace music even not tested by 'inventor' so hypothetical invention not developed one. So to the description of invention cannot be put simply 'a music'. It must be the music in the best way described as sequence of scores or properties enclosing strictly a family of healing sounds. Somebody may find different music not fitting to the description above and on that new whole he should also be granted with a new patent (because he handles body cells with different acoustic sequence - maybe better), so it is different novel device for curing cancer using sound waves. And the difference in those devices are different sequences of waves handling cells in different way maybe with the same good result.
      - It has to be proven that the effect of device is different from mere pleasure (or not) caused by sound waves and in fact it must be proved that there are some additional previously unknown physical effects (cancer cell dying) caused by this specific music (and this is also important in case of software). These technical effects have to be described in detail because such a device cannot be a hoax.
      - Usage of this music as a music for other purposes (when you play it on guitar instead of CD-PLAYER) should not be forbidden (therefore we call it device for curing cancer - not a CD-PLAYER+CD-ROM any more). (And in similar way usage of LRU algorithm only as a part of the content for PC also should not be forbidden even if LRU was patented as part of disk device).
      Then go forward with patent. You have found something new :)

      However I cannot find something technical in business software like worksheet or word processor. Its purpose is to process data. They do not handle hardware on other way than using API of drivers to do data presentation or get input data (so it is mere usage of e.g. graphics card and not a novel, inventive graphics card or new kind of faster disk). It is content for PC. However OS kernel algorithms scheduling particular type processor (or family) may be considered as a driver so should be patentable.

    37. Contributions, apparently from software writers like that last one (May 03, 7.17pm), just confirm that the EPO has it right, and that the software folks would agree, once they have mastered the EPO caselaw. Incidentally, when it comes to legal certainty, isn't this an area of patent law in which the civil law jurisprudence of the 24 Technical Boards of Appeal, cumulative now for 24 years, over thousands of Decisions, with no one Board deferring to any of the other 23, beats common law binding precedent up,down, inside out and backwards?

    38. Brimelow (like LJ Jacob in Aerotel/Macrossan) saw a conflict between decision of different boards, Straus (like Pompidou) sees non. The EBoA has to decide.

      The mentioned decisions of the BoAs don't stick to the literal wording of art 52 EPC, but Straus urges the EBoA to stick to the literal wording of art 112 (1) b) EPC.

      In their CII amicus curiae brief Schricker/Straus voted for the patentability of computer programs. Estermann summarized their vote: "By means of coherent legal arguments and appeals to legal constraints, the extension of patentability of computer-implemented inventions has to be extended and any political discussion prevented". Now Straus tries to prevent even a legal discussion to reach his goals.


    All comments must be moderated by a member of the IPKat team before they appear on the blog. Comments will not be allowed if the contravene the IPKat policy that readers' comments should not be obscene or defamatory; they should not consist of ad hominem attacks on members of the blog team or other comment-posters and they should make a constructive contribution to the discussion of the post on which they purport to comment.

    It is also the IPKat policy that comments should not be made completely anonymously, and users should use a consistent name or pseudonym (which should not itself be defamatory or obscene, or that of another real person), either in the "identity" field, or at the beginning of the comment. Current practice is to, however, allow a limited number of comments that contravene this policy, provided that the comment has a high degree of relevance and the comment chain does not become too difficult to follow.

    Learn more here: http://ipkitten.blogspot.com/p/want-to-complain.html

    Powered by Blogger.