Petition3

Software Patents stifle innovation

  1. Software patents are too slow. Examination takes 5 years, if not more. Patents limit competition for up to 20 years. The software life cycle and amortization of investments are much shorter.
  2. Software patents are too expensive. Litigation costs, damages and royalties are too high for a majority of the affected actors.
  3. Software patents ruin investment. A typical product can violate hundreds of patents. A complex product can violate thousands of patents. A single patent might ruin years of work invested in a software product. Software patents create minefields for developers.
  4. Software patents are overly broad as a right, as opposed to precise rights such as software copyright. Broad, intricate and clumsy rights are damaging for the market while at the same time more valuable for their holders. 'Inventors' may deliberately phrase their applications very broadly and negotiate with the patent office over breadth of the grant.
  5. Software patents do not protect products. As software products are complex, a product cannot be protected only with one or with few patents. To completely protect a product, one needs many patents which are not owned by him/her.
  6. Software patents create fear in the market. As a single infringed patent is needed to shoot emerging competitors in the cradle, developers refrain from the risk of entering a 'mined' market. Those legal ambushes deter innovation investment.
  7. Software patents deprive authors of the fruits of their work. Patent regimes dilute your ownership over copyrighted works because they overlap with the realm of copyright. Software patenting closes an alleged copyright 'protection gap', which was preserved by the legislator for reasons that advise against patenting, too.
  8. Software patents are not economically justified. Insufficient economic evidence supports an application of the patent system on software. On the contrary, most studies hint that software patent regimes restrain innovation.
  9. Software patents reward 'hot air'. Ideas are not scarce but cheap. Their disclosure barely justifies to grant rights to prevent them happening. Developers who read software patents consider them an offence: they disclose nothing useful.
  10. Software patents are difficult to research. Patent databases, software and patents are complex. Patent attorneys can do unreliable research for you. Only courts can decide if you infringe a patent. You take all risks and tolerate high costs. Patent Offices admits it is impossible to find prior art in source or binary code.
  11. Software patents are useless for defensive purposes. Software patents as defensive purposes do not work against 'patent trolls', since they do not have any product which you can countersue for infringement. When your business bankrupts after years of litigation, a troll will buy your 'defensive' patent and terrorise your competitors.
  12. Software patents discriminate against small players. Small companies are forced to bow into cease-and-desist letters about questionable patents or settle out of court as litigation is too expensive and takes a along time. As a small player you can either pull or cut your software or take a license if available. Small companies are excluded from large cross-licensing deals with large companies, since they do not have enough patents to exchange. Large companies are able to recreate a level-playing field with those cross-licensing deals and reduce the risk of infringement. Free software developers see their software becoming non-free with the collection of royalties.
  13. Software patents are a problem for large companies. Large companies view positively the exclusion of competitors from the market. But they have to face negative effects of software patents, such as litigation costs, damages, royalties, product removal, and shift of resources from the R&D (Research&Development) to the P&L (Patents&Litigation) department.
  14. Software patents creates a cold war.
  15. Software patents do not fit for service-oriented markets. The software market is mostly about providing services. Patenting suits service markets badly. Patents were designed for the industrial age.
  16. Software patents are not written by (and for) developers. 'Inventors' can write patent applications without skills in software development and sue software developers who independently 'recreated' their 'inventions'.
  17. Software patents endangers users. Users of software might be sued for patent infringement (ex: banks), since their activity might depend on an infringing piece of software. The product might be removed from the market (ex: Blackberry). Software underpins a lot of key activities in the economy (ex: creation of an airplane).

On the opposite, large players can countersue with their own patent portfolio but it won't help them against patent trolls. A litigation business without products will not cross-license. And you cannot enforce your own patent against a larger player either as it can force you into cross-licensing.

What shall our parliaments and governments do to preserve innovation?

  1. Apply sound economical justifications and impact assessments in a democratic legislative process.
  2. Pass legal clarifications to substantive patent law. Such clarifications comprise negative and positive tests for patent examiners to assess what is eligible to merit a patent. We made proposals and remain open to alternatives that contribute to the end of software patenting worldwide. Two major ones are:
    1. A claimed object that consists only of instructions for use of generic data processing hardware (universal computer), also called “program for computers” or “computer-implemented solution”, is not an invention in the sense of patent law, regardless of the form in which it is claimed.
    2. A claimed object can be an invention in the sense of patent law only if it contributes knowledge to the state of the art in a field of applied natural science.
  3. Overcome a patent reform discussion trapped into "non-obviousness" The 'American disease' of patent law requires a return to real steering instruments. The patent community has been using that filter to distract reforms and to get industry backing for dismantling of more meaningful examination filters.
  4. Apply democratic reforms of patent institutions. Patent offices have to stay neutral and abstain from lobbying. They must let patent examiners contribute their first hand experiences. Persuasion for patenting based on the assumption that small enterprises just lack awareness puts preconceptions over the rationale of market choice. In a liberal democracy public servants and policy makers do not educate but listen.
  5. Provide for non-infringement declarations which override enforcement of patents. Rather than you taking the risk to research patents, patent holders should declare upon you request if your product or standard infringes one of their patents. Such estoppal provides legal certainty and standard confidence.
  6. Get patent professionals out of policy making. Increase the influence exercised by economists on the governance of innovation policy. The quality of professional judges cannot be exchanged for 'technical judges' without legal training and eligibility to a judicial office or administrative 'case law'.
  7. Keep substantive patent law harmonisation away from Free Trade Agreements.
  8. Start an open debate about the patent crisis aimed at finding solutions. Economists can easily explain to you why free rider effects make patent opposition suboptimal. Other institutional unbalances pressure patent examiners to grant permissively. Let's review the institutional incentives and start reform.