Comments for Fixing the software patent problem

The problem of the patentability of software leads to lengthy debates concerning the nature of software, entailing a legislative disagreement between the various state legal devices. A phenomenon of legal insecurity appears around the question. Indeed, the article presents perfectly the differences between the American and European systems, but by observing closer the jurisprudence of States within the European Union on this question , we realize that Europe has difficulty to convince everybody, and collides even sometimes violently in the reluctance of some States, such as France, which seems to refuse systematically patent software. (Tribunal de Grande Instance 18th of June, 2015, one a visa application as defined in Article L.611-10 of the Intellectual Property Code).

French Courts express clearly an ill-at-ease concerning patent software. But what are the reasons? Do French judges want to destroy all the innovation incentives in this field? From my point of view, I think that it is rather the opposite, to refuse patent on software maybe a right way to promote the development and the distribution of knowledge in the ground of Information technology.

First of all, algorithms which are endowed the software are fundamental because they are basis of the IT development. The fact of patenting a software prevents the competitors from using algorithms contained in the software to develop new software. As the Professor Donald Knuth showed in his letter of 1995 sent to the American office of patents, programmers are using significant number of software to realize their work every day. If patents on the software had usually been admitted in the 1980s, an important amount of programs we use today would certainly not have been born, and the IT progress would have been terribly difficult. Worse, if the most fundamental, basic algorithms, had been patented, it would have been impossible to do anything.

Secondly, the absence of patent on software allows a more important sharing of the knowledge. Indeed, a company which conceives IT programs can easily be copied by his competitor knowing that the price of reproduction is almost nil, but it allows the company, secondly, to take advantage of the work achieved by these competitors for, in his turn, copy him and obtain a margin beforehand. And so on. Without patents, companies will more focus on the software quality improvement.

Lastly, the company which does not want to see its software copied can completely sign a contract with his customers by forbidding them to propagate outside of their company the software supplies. Customers will have only interest to respect this agreement if they want to be able to take advantage of regulary updates proposed by the supplier of the software.

To conclude, the question of the patent software is maybe a rather specific domain of the intellectual property. limit as far as possible the possibility of patenting a software and by shrinking as much as possible the duration of this legal protection seems to me the best way to favor the innovation.

Sources :
https://www.legifrance.gouv.fr/affichCodeArticle.do?cidTexte=LEGITEXT000006069414&idArticle=LEGIARTI000006279405

https://www.village-justice.com/articles/Brevet-logiciel-France-justice,20334.html

https://www.contrepoints.org/2013/02/14/114815-le-logiciel-libre-est-lavenir-du-logiciel

https://framablog.org/2014/08/11/donald-knuth-brevets-logiciels-1995/

Hello,
I would like to contribute to the topic and share my point of view here.

First of all, I totally agree with Eric Goldman. We can look at the origin of patents to see that their goal is to compensate the problem of appropriability. As Belleflamme & Peitz (1) put it: government ought to intervene to promote dynamic efficiency, so to provide the right incentives to create and innovate. However it does not seem appropriate that you can patent a computer program capable of achieving a result (e.g.). I think it is too broad and because there are many ways to achieve a similar result, allowing such a thing would result in strong monopolies. For instance: if you have a monopoly in a computer program capable of solving calculus, you have a monopoly on a broader range of programs’ possibilities than if you just had a monopoly on one specific program (such as microsoft excel let’s say). Thus the damage done to the consumer would be larger because you would have hinders entries on several programs and your monopoly covers what could have been a bunch of small monopolies but competing with each other in some cases. In this case I argue, the deadweight loss (a measure of market inefficiency) is larger in the one case than in the latter. And, in a consumer perspective you would have less choice with broader patents than with very specific patents. My answer for this question is: patent should not be too broad. This point of view has been shared by the Obama administration (2).

Secondly, I would like to talk about asymmetries between software’s. Indeed you have software that can be used for a very long time (such as Microsoft excel) meanwhile others would be used for a very short time. I think for example of trading software, those are usually in harsh competition and there are always new improved software’s by the end of the year. So the length of a patent depends of the context. As shown in this article (3), in very competitive markets, patenting can be very handy.
The point I am making here is that we should allow a case-by-case study for software. We cannot set a fixed duration for every ever patent. There are areas where 5 years would be plenty enough because after a year the software is out-dated but in some case the software could be use for 50 years (I doubt that could be the case but for the sake of the intervention I would say so) with the patent, and thus that would generates issues regarding competition. So I argue here that authorities should “study” the market before granting a patent for a certain period of time. Here we face a trade-off between three things:
How long the patent could be used before turning into a useless patent
How long is enough to make sure that people have incentives to invest
How long is not too long to let others enter the market before it’s too late
To answer those questions we could look at the history and the past law suits so that we have a background to see in which sectors patent should be shorten or extended. Where it is likely to generate monopolies and so on.

Lastly, there are also good consequences of patents. Once you apply you have to disclose you invention and 18 months after you apply for a patent you have to the invention becoming “public” and it has therefore a positive externality on society because people can look at the invention and all the details and the science can progress faster. This should not be neglected neither.

(1) (1) BELLEFLAMME P., PEITZ M. (2010), Industrial organization: Markets and Strategies, Cambridge university press

(2) http://www.nytimes.com/2013/06/05/business/president-moves-to-curb-patent-suits.html?_r=0

(3) http://bits.blogs.nytimes.com/2012/11/20/smartphone-patent-wars-show-the-system-works-patent-chief-says/

Before entering the problem of the length and breadth of a software patent itself, I would like to consider first the question of the harmonization, either for the hardware than the software patents. Although the criteria regarding patentability of many institutions worldwide are quite similar, there are still a lot of differences and sometimes confusions because the inventor does not know to which institution he has to refer or in which category his invention belongs to. This is due to the globalization effect implying that a new invention brought to the market often reaches quickly other markets, where different conditions regarding intellectual property apply. According to the International Trade Administration (ITA), an inventor can protect his invention only in the country or region where he has been granted a patent from the local institution. Taking into account the criteria of other regions or countries becomes therefore a difficult problem that leads to additional costs and losing precious time.

Moreover, the absence of harmonized patents results in difficulties and confusions, especially regarding the question of the software patents. It is indeed illustrated in the article where the EU and US governments have different approaches concerning the granting of software protection. In these kind of globalized markets where information is at the very centre, I do believe that harmonization would be a progress as it would facilitate the purchase of intellectual property across countries.

If I had to implement the harmonization, I would rather choose to implement the EU software patenting system worldwide rather than the one of the US. The EU definition of a patentable software invention is according to me more relevant compared to that of the US which is more abstract and too wide. It highlights the invention itself, a concrete solution to a problem, while the US system grants a maybe too wide protection and focuses too much on the function of the invention. That is why I do believe that granting software patents to concrete and implementable inventions is more appropriate. It goes a step further and reduces the scope of the invention, encouraging somehow people to innovate.

Concerning the length of that type of patents, I do not think it is necessary to grant patents that last more than 5 years. The software industry is a fast changing environment, softwares are created and improved everyday all over the world and inventions becomes quickly overwhelmed by new inventions. The idea is that these new inventions will make no more sense for the previous inventions to be longer protected as they will not bear the risk of being copied anymore. On the contrary and especially in the software industry, these inventions will serve as an important tool for other inventors who will invent something else a bit later. That is why I do believe that software patents should not last too long: previous inventions have positive effects on new inventions by being a necessary platform or tool on which to rely. Finally, my summarized recommendation is that legislators should be aware that granting too long or too wide protections could prevent other ideas to be developed.

Sources:

https://www.stopfakes.gov/article?id=Is-My-US-Patent-Good-in-Other-Countries

The patent of software has been highly debated since many years. First thing we can notice, is that the patent is essential in order to encourage technological innovation. In the case of the software patent, it is the ‘invention’ which is implemented in the software which is being patented. According to this, anyone who try to use that invention, no matter how it is implemented or in which language or even if they have developed the invention independently, will be infringing. An exemple of software patent is Apple who has patented the “bounce-back” feature of iOS. This allows him to prevent Android competitors from copying this aspect of appel user interface.

One of the main reason why companies choose patent instead of copyright, it is the strength of protection provided by the patent system. Indeed, copyright only attaches to the software code itself. So it is easier to reimplement the ‘idea’ behind a copyrighted software work.

The patent system protects diverse industries from everyday kitchen utensil to a nanotechnology chip and they are all governed by the same rules even though they innovate very differently. In my opinion, some rules must be change to be adapted to a specific innovation. Indeed, 20 years of protection might make sense when talking about drug companies which need long patent windows because of clinical testing, or complicated physical processes but for software I agree with the amazo founder ,jeff Bezos, who suggests in his open letter that an appropriate term for software patents would be 3 to 5 years because softwares aren’t where you might have to tool up and build factories. A good software innovation can catch a lot of wind in 3 or 5 years.

I am pleased to give my opinion taking inspiration from some relevant elements discussed in this article by P.Belleflamme.
At first I will try to raise some issues regarding the EU vs US criteria on software patentability, then I will proceed in giving an answer to the optimal lenght-breadth combination of this kind of patents.

I noticed that there are not so clear-cut boundaries between US and EU criteria on software patentability, in pratice. The case of Amazon “1click” software (1) is emblematic in this regard: the specific software arguably represented a perfect example of implementing a business method. The patent, not surprisingly, was effectively granted in the US in 1999 and not in Europe. However, that was not due to it being considered a business method or ‘software as such’: even the EPO originally considered it patentable and it effectively reached the third stage of examination.
At the end it was the inventiveness step that was found lacking, when findings about prior art emerged. Of course inventiveness is a principle shared by the americans.
There are other cases in which the patent was not granted in Europe contrary to the US: the case of Microsoft’s “Time based hardware button for application launch” (1), for example. The reason why it reached patentability in the US was that the claim of it effectively improving the functionality of a button, was accepted.That was not deemed sufficient in Europe.

My opinion is that, in these cases, court discretion plays a much bigger role than application of written law: Some clearer guidelines and articles are strongly needed.
What’s more, I share Lemley’s view (2) in saying that it is not ‘bad’ to patent a really inventive software, it may be good to reward such a step. The real issue in software patenting being granting rights only to the specific methods, such as in the pharmaceutical industry, rejecting all ‘functional claimings’. I agree this could offset the relevant problems (patent thicket and patent troll). Patenting only specific methods also lowers the extent of monopoly power by large software developing firms.

Finally, relating to the breadth-lenght issue, it is clear that in a fast paced industry, as already said by my colleagues in previous comments, the lenght has to be decisively shorter comparing with other industries (say, the pharmaceutical one). The point is that it has not to be paired with large breadth (as generally with patents (3)) because in general the costs of R&D to be recovered are lower. The relatively low costs associated with high possible gains are a reason why it is always convenient for firms and private inventors to be on the frontier of software innovation.

References:

(1) http://www.clarkemodet.com/blog/2014/12/software-patents-US-vs.-Europe.html#.VjplWoQ79E4
(2) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2117302
(3) Belleflamme and Peitz, 2010, pp. 517-521

In his present article, Paul Belleflamme subsumes the differing approaches to software patents between the US and EU. Furthermore, he illuminates this topic with respect to its three main dimensions, which are the questions of patent matter, breadth and length. In the following, I concentrate on the questions what the optimal breadth and length of software patents should be. In doing so, I start by defining the goal of the protection of intellectual property, and the optimality it needs to achieve between different societal objectives. Afterwards, I put these differing perspectives into the current context, which are the variations in software patent practices between the EU’s EPO and the US’s USPTO. Hereby, I particularly relate to the underlying political forces and the influence of certain interest groups. These latter two arguments also explain why I am personally in favor of short lengths of software patents, and at the same time supportive of a narrow patent breadth, while totally excluding the patentability of those innovations who solely target functionality.

To begin with, I want to make clear that the protection of intellectual property via patents does not constitute an end in itself. It rather serves as an intermediate means to an overarching objective: to keep the growth rate of intellectual property on a level that is optimal for both current and future generations of society. This “optimal” rate is of course unknown. Hence, one might look at the trade-off that defines this rate: on the one hand, incentives for companies to invest into R&D need to be maintained, which is only possible in competitive market economies via institutions that guarantee adequate capital return on R&D, e.g. through profits from transitory monopolist positions. On the other hand, society also profits from the possibility that new innovators, such as e.g. small start-ups in the technology scene, are able to quickly improve upon existing methods and hence increase general welfare. For instance, software products form the base of almost everybody’s daily routine in developed countries, ranging from the common online bank account over the daily utilization of smartphones to the usage of word and excel in most jobs of the modern knowledge economy. All of these innovations were initially developed by small start-ups, most of which have by now become large enterprises. Many facilitations of daily life are especially acknowledgeable in areas that are hard to monetize, e.g. the increased possibilities to contact with friends and family via the communication program skype. Hence, one might make a case here for possibly increasing social and societal returns to scale from having particular software products available for continuous improvement through competing companies, and hence a case against excessive granting of software patents.

Of course, so far I only mentioned the interests of a widely dispersed group of profiteers, which are naturally in a worse position than established software companies to intermediate their interests in a targeted way to policy makers and the relevant institutions. Given the massive presence of large-scale software producers in the US (IBM, Microsoft etc.) and their likely influence on legislative and executive procedures, it is not overly surprising that patentability of software products is indeed far more common in the US than in the EU, and seems to be much less balanced with respect to the above mentioned broader societal objectives. Furthermore, since European officials and politicians are well aware that they are lagging behind the US in terms of technological creativity, they also face more incentives to keep a level playing field, such that potential startups are not facing additional hurdles in their attempts to catch up to US companies.

Given the enormous technological progress in these fields of entrepreneurial activity within short periods of time, I hence consider it as essential to only grant short-term patents in these domains (say a maximum of five years) in order for Europe to not lose further grounds vis-à-vis its strongest competitor, the US. Furthermore, it also think that it will be equally important to strictly exclude those software products from patentability which possess so-called “functional claiming”, since these would guarantee the incumbent US companies to inhibit potential European competitors and hence solidify their dominant position eternally.

Both arguments above are admittedly biased towards a Pro-European perspective. However, and in order to come back to the social objectives mentioned in the beginning, I believe that it is better for European economies to actively partake in the race for technological progress than rather taking only the passive part. This is of unique importance since these new technologies are increasingly affecting also sensible parts of our digital selves, for example the protection of privacy rights of personal data, which is considered a much less important issue in the US. Last but not least, I also think that the maintenance of technological competition between the EU and the US increases the benefits of all those of us who profit from daily software utilization, be it for commercial or non-profit purposes.

Software patent is a very controversial matter, there’s no doubt about that.
In my opinion, there’s 3 issues in the current system that the legislator needs to tackle in order to create a safer and healthier competition: when should a software patent be granted, which protection should this patent receive and should the legislation be the same all over the world?

1. When should a patent be granted for softwares? The problem with software patent is that it usually protects a concept more than an actual creation. Therefore the legislator must be very careful when granting a patent for softwares, because he could give a too broad protection and it could allow the innovator to sue a lot of people because of that protection.
Because of my limited knowledge on the subject, I can’t really give any advice on that matter, but I think the legislator should spend time on each case to make sure the protection granted is not too broad.

2. Then comes a different problem if a country decides to grant patent software: how long and how broad should that patent be? This is definitevely a tricky question: should the legislator choose to protect the software very hard and for a long time or should he offer a short and soft protection? In my view, the good choice lies in between. Because the software industry is such a fast changing environment, granting a long-term patent would surely be counter-productive, and could slow down an important industry in our more and more tertiary economy. On the other hand, I think that software patents should be well protected, and broad enough to guaranty the protection of true innovation.
But then again, my knowledge is quite limited on that subject and I’m not fool enough to think that the solution is that easy.

3. The question of allowing software to be patented is surely controversial, but I think no one would argue that the legislation on that matter should be harmonized across the world. Having a much stronger legislation in Europe undeniably creates a competitive disadvantage with the USA, which certainly leads to a brain drain from our brightest minds. Knowing that a patent in the US can protect their software surely encourages them to develop their software over there rather than here where their competitors could steal it. An agreement between the US and Europe on that matter would probably be profitable for us, and that could happen soon because of the negotiation for the much-debated TTIP.

Allowing softwares to be patented is definitevely a difficult question, and there’s probably not a finite solution for it. But I think that the system would already be better if it was harmonized across the globe, and before doing anything, legislators around the world should first decide to do it all together.

The aim of a patent is a recognition of an innovation and a guarantee for its owner against an unwilling replication. So in this concept, patent is a greatfull tool and provides a certain security. It is usefull to protect a business and It is convertible into cash. As far as this description, i agree with the concept of patent.

However, this lets place to some downward spiral as the software. It is problematic when the protection is about the functional claiming and not about a way to achieve this function. Indeed patent allowing a « function claiming » is a barrier to innovatio. There is no sense to limit the progress.

Moreover this creates some monopoly and breaks the perfect competition which means a loss for the consument. That reveals in higher prices and less choices. This way of thinking is contrary to the concumer’s protection UE policy. It seems essential to adjust the breadth of patents.

Concerning the length, It’s more difficult to establish a common rule. Because it depends on the domain of the software. Some software don’t need an excessive time of protection and some well. So It will be more efficient to grant a the patent’s length in function of the kind of software.

No doubt, software patents’ debate is a ripe topic, but before we dwell into what is right and acceptable to ALL, we need to understand what drove the different views across the world. Both the US and EU have expressed contrasting opinions on how to treat software patents and the reason for differing dynamics is the difference in the scale of the software industry in these parts of the world. If we look at the top 10 major software giants, we observe 8 of those 10 to be from the US. These companies drive the economy as no one else does and being liberal in granting patents was in some ways feeding their egos and encouraging them to invest, grow and expand further. On the other hand, the EU has been more of an unbiased moderator and is driven by opinions of delegation of multiple countries. This is why a company like SAP in Germany cannot dictate the rules (indirectly) to the entire EU.

Now coming to what is more logical, I believe the EU approach is more appropriate in handling the software patents. The reasons are obvious – 1) Promotes more opportunities to innovate in terms of process/algo, 2) Promotes entrepreneurs, 3) Reduces the legal burden and unnecessary examination costs (and hence benefits consumers), and most importantly, 4) continues to protect the implementation which is technically progressive.
This is something that the US has also slowly started to accept. The latest Supreme Court ruling on software patents in July 2014 (Alice Corp v CLS Bank International) denied rights to patent-eligibility on grounds that applying a software process to an abstract idea does not simply cut it.

Now looking at the interesting concept of length and breadth:- ‘Patent breadth’ is something that defines a patent’s scope and will bring more transparency in the industry of patent trolls. ‘Patent length’ on the other hand is an important lever for growth and innovation – It should be adjusted based on the applicability and the level/grade of innovation (“depth”) by studying the patents on a case to case basis.

Now for a moment, let’s look beyond patents and take a step back – Won’t software innovations be better protected through copyrights? “Copyrights allow the abstract ideas behind a software problem to be created by anyone, but protect an implementation of those ideas in concrete form, so developers who implement their own ideas do not have to worry that someone will put them out of business”(2) – something that is desired. Also, copyrights are straightforward to enforce and allow easy identification of what is being protected – Can’t we move beyond the software patent debate from looking at this option?

References:-
1. http://en.swpat.org/wiki/Alice_v._CLS_Bank_(2012,_USA)
2. http://oneminute.rationalmind.net/software-patents/
3.http://archive.fortune.com/magazines/fortune/fortune_archive/2007/05/28/100033867/index.htm?section=money_latest

Patents are a means to protect the innovation carried out by an individual or a company. It helps protecting the original idea or thought or the process meant for developing something which is not obvious and is genuine. It also gives the innovator a sense of protection that his innovation won’t be copied and he has enough incentive for innovation. The patented innovation is restricted for commercial uses ad the innovator has the sole rights.

Patents can be classified into product and process innovation patents. Generally software patents would be a subset of process innovation. In case of software, the investment may sometimes run into millions of dollars and few years of development time. To protect the time and effort put, softwares can be patented. The question that arises here is what should be the guideline for providing such and the time horizon for the protection.

To remove vague descriptions, the guidelines should clearly describe the rules for granting a patent. The patent should be on a software which helps solve a problem via a descriptive process. I completely agree with Richard Posner’s view that software patents should have a smaller time horizon. Software, unlike drugs, have an impact on a large chunk of population. With faster churn rate in thetechnology sector where things change frequently, patent regime for a long term would hinder or restrict spread of knowledge. Here we can bring our knowledge of marginal benefit and cost. The marginal benefit for the innovator should be just equal to the marginal cost so that there are agents in the economy who can innovate. Obviously, a cost benefit analysis should also be considered keeping the society in mind. Having a longer patent regime would cost the society a large deal as it would be deprived of the new technology.

So there has to be a balance between the interests of the innovator and society. We have seen the Apple vs. Samsung war on patents. Whenever there is ambiguity, there would be conflicts. Additionally, the software world saw the rise of Open Source Software where some of the finest ones were developed by people all throughout the world. Linux as the Open Source operating system, Mozilla as free web browser, Scilab as a substitute to Matlab are few examples which illustrate that the world can do great things collectively.

The question that remains is how to motivate like minded people to bring together to eliminate patents. Till then the patent regime would continue. For the better good of society, the regime length should be less than 5 years

In the first place, it is important to define exactly what a software patent is. “A software patent is a patent that is provided to enhance computer performance by means of a computer application.” (1) It is a current matter as the technology is constantly growing and has become a key sector.

Regarding the fact that patent are ruled by country, I do not agree with it especially in our present society. Companies are often based in several continents and it would be difficult for them to use a patent in their American headquarters but not in the ones in Europe. As consequence, that would probably pose them a problem if they are not allowed to use the same processes in Europe and in America.

Furthermore, the lengths of patents are also questionable. Seen that we live in a world that progresses really fast, patents’ duration should not be too long. I agree with the statement of the Electronic Frontier Foundation that says that a patent should last a term of five years.
Although we can put a limit to the length of a patent, it is harder to set limits to a patent’s breadth. The United States have a different vision than Europe concerning this subject. They have a way broader way to define patents. They do not necessarily have to refer to process but they can refer to concepts also.

On the contrary, it is more restricted in Europe and I am a partisan of that vision. In this case, software patents represent innovations in a technical process only. It avoids the multiplication of patents about innovation business processes and the risk to go over the same object. Nevertheless, the business processes are very useful; it should be under more rules so that it is not so easy to get a patent on a simple idea. Software patents should only cover a particular way to achieve a goal but not the goal itself.

In conclusion, I think that we should go towards a globalization of the software patents’ world so that it becomes easier for multinationals present on different continent. Because softwares are in continuous evolution, the patent’s length should not be too long and its breadth too large.

(1) Techopedia, Online http://www.techopedia.com/definition/22199/software-patent, consulted on 7/10/2014

I’m certainly not an expert in the area, but reading the article and getting informed on the issue opened my eyes on a world that I barely knew was suffering from such a problem.

As a starting point, I would like to quickly summarize what are the main problems that occur in the software patenting field. As exhaustively pointed out by numerous authors, we can notice that the current patenting model -even the EU one, which is way stricter than USPTO- doesn’t fit the software field; here are some of the main reason. In a first group we find the structural problems, the ones that relates with the quite big amount of differences that can be noticed between a classical patent subject -i.e. a particular pharmaceutical formula- and a software: software is, basically, often too hard to describe precisely, and therefore gets patented at an high level of abstraction, which leads to a nearly-impossible work for the patent office when it comes to find out if the idea that has to be patented satisfies the requirment. The overclaiming, for example, should produce patents rejections on various stages, but it often doesn’t, due to the complexity of the work that should be done. Not to menction the fact that software’s life is usually shorter than the time required to patent it.

A second ideal group can be formed by the issues that makes software patenting uneconomical. We all know that -and this is particularly true in certain fields, like smartphone industry- firms are now “investing” huge amounts of money on patent portfolios just in order to avoid them to fall in the competitors hand. Furthermore, there are some facts that show how software patenting could be avoided: opensource software is the most impressing example, but it’s not the only one. As Eric Goldman (http://www.forbes.com/sites/ericgoldman/2012/11/28/the-problems-with-software-patents/) points out, the mix of copyright -multiple aspects of software can qualify for copyright protection- and trade secrets could provide an acceptable defense, and thus a sufficent incentive to produce. Also, we must never forget that we’re talking of an untraditional kind of industry, one of the peculiarity being the incredibly high “first mover advantage”. If we consider the average software-related product lifecycle, it’s easy to find out how dominant is the position of the first seller.

That said, I think it’s hard not to agree with the ones that claim new forms of regulations for software patenting. I really don’t think that a situation in which, for example, “defendant’s costs to find the patentee’s version of the invention beforehand are greater than the defendant’s costs to invent it on his own” (https://www.usv.com/posts/we-need-an-independent-invention-defense-to-minimize-the-damage-of-aggressive-patent-trolls) can be considered sustainable. Still, as experience taught us, merely delegate the solution of this kind of problems to the politicians alone could represent a risk: in the worst cases complications increase, in the more likely one solutions are late and incomplete. My opinion is that, in most of the cases, the market itself represent an adequate incentive for firms to invest in R&D, and that in such cases patents are more of a limit than something else -a limit that also have the downside of favouring the biggest over the others.

According to me, EU approach to software patent seems to be better due to its specificity. Given the rate at which technology is changing, it is highly likely that we will see many different approaches being invented to a common problem. US approach will restrict going for new and innovative approach since old ones would already be patented while EU approach will incentivize companies for investing in new approaches since every approach can be patented individually. One example in front of us is the smart phone operating systems. The technology for that has seen rapid changes and now has reached its maturity stage. Companies will only invest in this area when they are assured that their innovation can be patented and that where I feel, defining the breadth of a patent will help.
For breadth of a patent, it is very necessary to patent an approach not the problem which is solved by software. It may become subjective at times, and hence require neutral authorities to define the limits. Also, having a universal body across countries to take care of patent issues will ensure that same practices are followed in every country which will, in turn, help MNCs.
Regarding the length of a patent, a common number can’t be applied to every area. Every software development requires different level of effort and investment and hence, need different patent length. For low effort areas, patents can be of 2-4 years of length while high effort patent may require longer patent lengths like 10 years. Again, it can be subjective and may vary over time.

After the reading of the article and of my researches, the European Patent Office’s approach seems, in my opinion, to be more logic and to work better.

More logic, why?

For the European Patent Office, to be patentable a technology needs to be novel, inventive, susceptible of industrial application and obviously innovative.
When we apply this four requirements to a software, it respects the three firsts. Software can be new, it can be inventive if it require some creativity to do and if it’s not a simple evolution of an older one and it can have an utility for industries.
But is it really innovative? is a software not just a piece of paper writes in a mathematical language like a literary work done by an author?
I agree with the European Patent Office’ point of view which is to not consider the final “product” as innovative because it uses conventional technology.
But, I think that software needs protection to not be claim as the invention if someone who is not the true inventor. In my point of view, a copyright is a good alternative because it provides some protection but can be use by someone else too. Thank to this possibility to use it, some others and smaller companies can do research with the software and perhaps find a true invention really relevant.

Working better?

There is two main reasons why I believe that the American and Japan’ system is less effective.
Firstly, as Mark Lemley claims, I agree that there is a patent thickets. For instance, the number of patent filed in the US and in China is respectively around 48 600 and 38 900 (footnote n°1) while Germany, the “higher” European country, faces only 18 600 requests for patents.
Secondly, the number of defendants in patent lawsuits is majority from software patent. For example, this number was just above 6 000 (footnote n°2) in 2011 which causes a higher cost of patent litigation . Moreover, the number of software patent’s lawsuits never stopped to increase since 2007 and so it will certainly continue to go up in the following years.

In conclusion, I believe that with the European approach which limits patent to an innovator solution to a problem and not to a “strict and finished” software is better than the US’s regulation. But, the perfect solution does not exist because the IT environment is complex. Moreover, this environment is also changing so rapidly that we need all-time to reconsider Software patents.

References:
[1]http://www.vox.com/2014/7/7/5862284/9-charts-that-show-patents-are-bad-for-the-software-industry
[2]http://www.lemonde.fr/europe/article/2005/07/06/le-parlement-europeen-a-enterre-la-directive-sur-le-brevet-des-logiciels_669825_3214.html
[3]http://www.sciencepresse.qc.ca/blogue/2010/06/28/brevet-logiciel-fait-vagues
[4]http://www.iusmentis.com/patents/uspto-epodiff/
[5]http://www.wipo.int/ipstats/en/statistics/patents/

Footnotes:
1) WIPO (Worldwide Intellectual Property Organization)’ statistics: http://www.wipo.int/portal/en/
2) http://www.vox.com/2014/7/7/5862284/9-charts-that-show-patents-are-bad-for-the-software-industry

I think patents have to exist, there is no doubt about this. Inventors need to have some rewards for their research and they deserve some advantages such as having exclusive commercial rights to use the invention or utilizing the invention for his/her own purpose. That seems normal but the regulations about approving patents are too broad.

First of all, I think that there should be a global regulation for all countries. It would simplify the aggregation of patents and patent holders would know exactly what they can do with their patent. It’s not logical that in the US you don’t need a technical inventive creation to receive a patent whereas in European Union it is mandatory.

Secondly, the patent office has to tighten its rules. Inventors deserve patents when they come up with a bright idea that is inventive and new. But they can’t have rights on other unrelated products built by others because their patent was to broadly described. The problem is that it doesn’t only affect the companies but also the consumers.

A third problem with the approval of patents is the leniency of the patent office. They approve patents without seeking further explanation of how the new product works, based on what calculations and how these calculations were found. Certainly these inventions deserve recognition but don’t we have to look more closely before granting a patent?

In conclusion, for me, the length of the patent is not the biggest issue unlike the breadth. A stricter regulation should be implemented worldwide with very specific rules concerning the length and breadth of the patent.

Generically, patents appeared as a reward from an investment made by a inventor/company in the creation of a new idea/technology. By restricting the use of this new knowledge to the patent owner, for a certain time, we’re giving a way to justify the investment (and the risk incurred) either by exploring or selling the respective technology.

If the concept itself appears to be very logical and a global accepted solution to the appropriability problem of knowledge, it’s appliance to the real world is not even close to be harmonic. The two main patent offices USPTO (US) and EPO (EU) have different – opposite wouldn’t sound wrong too- ways to operate.
USPTO accepts very broad patent descriptions, sometimes only regarding a specific function or design, and leaving further complains to the courts. A known example is the patent USD504889, owned by Apple Computer, Inc, which only claims “We claim the ornamental design for an electronic device, substantially shown and described”. The drawings and descriptions define a rectangular and thin electronic device, which fits not only in IPad, but also in almost all tablets.
In the opposite, EPO only grants patents describing technical advancements to the actual level of knowledge, leaving beside other types of contributions, as explained previously in the article. A patent was not granted by EPO to a software for internet auction because “… it made no inventive technical contribution to the level of existing technology”. Although it was a improvement of business processes.

Personally, I am more inclined to this second stricter view of patenting. In a world where information flows faster and faster (reducing development costs) and the velocity of software improvements increase exponentially, I think that in this specific case, market offers enough attractiveness for constant investments in development. It’s true this will lead to a greater instability; an example was the recent electronic game, Flappy Bird, that in a year became a global success, generating 50,000$ a day, and disappeared. The possibility of reaching worldwide makes the software market attractive but to sustain long-term market share a constant investment developing and marketing is required. The same idea applies to academic and professional software. My opinion is that a software brand should differentiate itself by its quality over time instead of looking for a patent to claim rights over wide usages or goals. If granted, this patent could also prevent the introduction of a better/more efficient software that solves similar problems.

In conclusion, regarding software patenting, my vision is closer to the European Patent Office (EPO) standards, which lead to a more open and competitive market (at least outside the courts). Without risking any predictions, I think that the acceleration of the IT will lead to a more specific and specially shorter patenting, in terms of breadth and length respectively.

References:
https://www.google.com/patents/USD504889

The software patents shouldn’t be that generic, patentees try to gain more money patenting the final goal of the program more than the method, that is terribly wrong because if there is an engineer who can solve a problem with an easier method cheaper and faster, he might not publish his method because he won’t want to pay de patent, and that will stop the software advances, it is unfair to try to own a solution and not a method, it produces more costs to the society just for the selfish person (company) who wants to received more benefits from their code.

In the other side, one thing that is conducting to this patenting situation is the piracy, people just crack programs and they can use them for free forever, this situation is really common and for example in México if you are not very good with computers you can just buy whatever program you like for one euro instead of buying the legal one or purchasing the license of the program. This situation is not very fair for the programmers and maybe that is what makes them be so ambitious with the patents.

My point of view is that it is okay to patent software, one method but not the solution itself, and the piracy maybe won’t be such a big trouble if the patents last only 2 years instead of 5 years. Maybe I can illustrate my point with an example: If there is now available the new “©Microsoft Office 2015” with new features and faster than the 2010 version and you want it, it’s fair for you to pay for it, but if you’re poor and you just need a program for make your homework you should be able to download the 2010 version for free, although you won’t have the new features and is more slow. My point is that the new technology should has a price, however the useful but not sharper and newer one shouldn’t.

Both normative and economically, software patents should last a conditional, relatively short time-span and should be exclusively awarded to “genuine” inventions. Patents are thought to encourage innovation, not to guarantee monopoly power. I believe the US setting fails in accomplishing the previous by, first, having a broad definition of what an invention is and, secondly, the patent time-span. On the other hand, the EU approach seems to be too hardware-oriented, and may fail to provide innovation incentives to software developers.

An invention should be defined as a new technical solution or a genuinely new functionality (in both cases, either hardware or software oriented). This means that patents should not be provided to “obvious” extensions of prior work. For instance, it is hard to consider Amazon’s “one-click shopping” (known as the option of enabling users to make single click online payments using payment information provided in previous checkouts) as a genuine invention. To me, it is rather an extension of payment checkout software. Having a broad definition of ‘invention’ such as one that allows Amazon’s one-click to be under a patent is not only unfair but also discourages innovation.

First, as seen above, a setting such as the American allows companies to attempt to patent almost anything. In some cases, actors may not even develop these patents: they may just be willing to seek rents from law suits. In order to avoid this, patent holders should be required to develop a product under a patent in a 5 years time-span -otherwise it should be obliged to license the patent. Seemingly, firms that exclusively buy patents to blackmail other players or to gain a dominant position, without actually exploiting these patents in productive processes should be banned.

Secondly, even if holders actually embed their patents into products, a broad definition of invention allows every single “obvious extension” to block other players from innovation. For instance, Amazon’s one-click can’t be implemented by other non-licensed software developers. The latter undermines the efficiency of new checkout software, even if it comes up with a new “genuine invention”, provided it can’t integrate one-click as a normalized feature. Finally, small players may be discouraged from developing new checkout software: these don’t have the resources to face law suits neither the money to afford the one-click feature.

In short, programming is about writing code. Novels and books are protected by copyright; the sentences and words that compose them are not. As I have argued, a wide definition of software invention would lead into impediments of new players to come up with genuine inventions, given that programming would need to avoid certain sentences and words. Regarding the length of patents, having browsed the US software patents directory, it is hard to find a single patent truly deserving a 20-year monopoly. As said above, it is crucial to fight against players not willing to develop products (or licensing) or actors that use patenting as a strategy to displace competitors. To do so, I would suggest a 5 to 7 time-span, conditional on product development and/or licensing to other producers.

Certainly, Amazon’s one-click has patent in the US while its application was rejected by the EU. In this respect, I believe that the EU framework provides more security to prevent monopolies on simple extensions from prior inventions. Therefore, I believe that the EU framework is closer to what should be done -at least in terms of the definition of an invention.

Nowadays, the software world moves faster than many others, and the debate about the patentability is increasingly rage. Laws are broken and big companies are fighting with each other about the patentee… But why do they brake laws? Why can’t they wait the end of the protection?

Firstly, let’s have a look at the patent duration. In the Software sector, I think that the duration mustn’t be too long. It would be preferable that innovation are patented for 3 years. It would let the competition unchanged. Indeed, on one hand companies shouldn’t have to wait during, for example, 20 years until the possibility to develop the innovation which is a nonsense because the software world moves faster than many others. On the other hand, a short period will encourage others companies to still stay competitive in the market. And as a result, the company which has the patent, we’ll be encourage to develop this innovation to find an advantage. So, I think that 3 years is perfect, it is not too much, not too short (which could reduce the motivation to innovate).

Secondly, concerning the breadth, which, according to ‘’ Industrial Organization Markets and Strategies, measures the degree of patent protection”, the book says that because patent length and breadth can used as substitutes; “if the marginal rate of substitution of patent length for breadth is larger on the incentive to innovate than on social welfare, the optimal patent is broad and short; otherwise, it is narrow and long”

Another important question is to know when the patent should be given. In the software world, I think that an innovator should obtain his patent when the software is finished and operational. Indeed, having an idea is not enough to bring innovation. Patent encourage innovation, but if companies could patent only their idea, which come before the “real innovation”, it could discourage the others companies who work on the same project.

In conclusion, I think it’s important to reduce the length of each patent to 3 years but it’s also significant to note that the length also depend on the marginal rate of substitution of patent length for breadth. Moreover, before finding an appropriate patent length and breadth, the software must be finished and operational.

Patents play an essential role in encouraging creativity and innovation in technology and promoting the development of better and more efficient ways of doing things. The existing patents may have been adequate till date, but the world has become increasingly digital at a very fast pace and thus, it is now necessary to reconsider how they relate to digital technology and make changes if necessary.

I think that the USPTO patent regulations are far too broad to be appropriate in today’s context. By giving out patents for a software or the end-product, the United States is indirectly restricting innovations in better ways of achieving the same goal. This would result in inefficiencies that could otherwise be eliminated. Also, such broad definitions of what is applicable under a particular patent would cause confusion and legal hassles that further result in a waste of time, money, and effort. By the same arguments, the European Union’s approach seems to be more efficient and relevant.

There should be both regulations in terms of ‘length’ and ‘breadth’ for patent laws. Ideally, ‘length’ regulations should be customized to the type of digital technology, based on an estimate of its period of relevance. However, this may be time-taking and involve too many resources, making the patent process lengthy. So, a broad ‘length’ such as 5 years as mentioned can be set, and changed when necessary. But ‘breadth’ regulations should definitely be implemented. It should encompass the process and method and not the overall result.

Defining the patentability rules objectively would help save bureaucratic time and effort, but a certain amount of flexibility is necessary to make sure the process is efficient and serves its purpose. Basically, it should just be ensured that the rules do not limit and instead, achieve the ultimate objective of rewarding and encouraging further innovation in the same field.

Intellectual property rights are, I guess, a very complex domain. The role of a patent is multidimensional; I think the most important one is the protection of someone’s creation but it is also a tool to encourage innovation, precision, competition… It makes it difficult to set boundaries on the “patentability”. Furthermore if it is for software.

Firstly, about the difference between US and EU, my point of view is that EU approach is more consistent. Having an idea is not enough to bring innovation and if we consider patent as a tool to protect innovation, then it is not applicable when the support isn’t novel as well.
But I feel like it is not totally right. When I think about smartphone applications, I don’t think there is thousands of languages to create an “app” (but I don’t master any of them so I might be wrong) and although, some particular applications are revolutionary! It wouldn’t be fair to ignore these innovative concepts.

Secondly, it seems that adjustments are necessary but which one? Posner’s idea does’t convince me. If we decide to patent some softwares, one of the benefit I would count is having time to develop the concept, enough time to go further with the idea. Nowadays, the software world moves faster than many others and shortening the period wouldn’t help.
On the other hand, Lemley’s suggestion seduces me more. As I said earlier, I think the support should be novel too to be patented but we can see a support not only as material or technical but as a combination. Most of innovations are built on others that were built on others, etc… Scott Berkun says in “The myths of innovation”: “The combination might be novel, or used in an original way, but the materials and ideas all existed in some form somewhere before…” (talking about the invention of the keyboard). So the function could be, according to me, a criterion of patentability.
Another possible way to design patents that would fit better the challenge of softwares could be based on derivating different categories according to the device or the domain of application (whether it is for medicines, or games). Then, the requirements would be adapted for its function.

Nevertheless, these ideas make the patent procedure even more complicated and the question become: does it really stimulate innovation? Isn’t it a brake more than an incentive? It might be but then, creators should have the needed help to accomplish their project while patent would be more fitted to the creation.

First I think that patents on software are indispensable for one main reason: Patents motive the innovators. If a company spent out a lot money in research and development for a product it’s really normal that she wait benefits, and that rival companies don’t use this technologies for their own products. It would be unfair.

To obtain a patent I find that the software must fill two conditions: To be finish and operational. Just have an idea of project should not be patented. Because there is not efficient product and also because it could be discourage the efforts of others companies that works on similar projects. So associations who deliver the patents should only patent the structure of the software and not his goals. Therefore the company can use his innovation and is protected from the others companies that can’t use the application but only try to create a similar product which use a different way to arrive to these same goals. An example with a drug: If you invent the first vaccination against the AIDS, the process of fabrication and the components must be patented, it’s the reward for the efforts of the company. But If another enterprise find a different mixture with other process,… but has the same effects in my opinion it could be also patented.

About the duration of the patent. I think that it must not be too long. For the example 3 years is sufficient to spread out and your software. And it’s also a motivation for the company to ameliorate the initial product. During the three first years they have the time to take some significant advantage on the concurrence. And 3 years is also not too long to discourage the others. If you give a patent of 20 years on a product I think that you could have two problems: The first will maybe stop the development of the initial product and it will discourage the others … So we could be in a situation of monopole that is not the best for the market. I have also an idea of “extension of the patent”: If after 3 years the company has developed and ameliorate the initial product, the innovators should the opportunity to keep their product under protection for a few extra years.

Finally I agree with a patent that encourages innovation. So not too long and only focused on manufacturing methods, components … but not on the goals of the software.

Patents allow a company to protect the fruits of their investments and thus encourage technological innovation. On the other hand, these securities granting an absolute monopoly on the invention, they can be dangerous. The allocation of these monopolies costs to society. The patent is a restriction of freedoms, particularly because, as a monopoly, the patent reduces competition, because it helps prevent others from using certain processes or the production of certain goods. Thus, the patent may hinder further innovation by preventing the free use of knowledge.

The patentability of software therefore remains a touchy subject. On one hand give the right to the patentability of software can allow small businesses to fight a giant preventing copying its products without compensation, on the other hand, the same giant has the financial strength to drop hundreds patents that would give him control over computer science.

I think that with patenting software we support the artificial creation, which curbs innovation and the spreading of knowledge. Indeed, in computer science, especially in the field of software, innovations are cumulative and incremental; every innovation is based on those that preceded it.

In addition, I think the length of time of software can difficulty be associated with the length time of a patent: the software will be obsolete in a few years, but the patent will continue to decline innovations based on patented technology for decades. Software patents therefore give no incentive for innovation: they are used in a roundabout way to ensure by law a competitive advantage.

If we wanted to make software patentable, I think we should imagine a legal exception. The duration of the standard patent is twenty years and this term does not correspond with the life of software. To reduce the monopoly of the original designer, and to preserve the original purpose of the patent (protecting investments, encourage innovation without endanger competition), it would reduce the time of the validity of the patent for a software.

As any material in the field of patents, we face two objectives. On the one hand, the promotion of R&D and on the other hand the protection of innovation by some legal instruments to provide the motivation to innovate. However, it seems to be difficult to conciliate this two objectives.

In fact, it is particularly difficult in the field of softwares because of the presence of «cumulative innovations» (1), which means that to develop a new software it is essential to use subsequent innovations to create it. A new software is necessarily built upon existing technologies and innovations, thus based on embedded knowledge. In order to reach the ultimate aim that consists in increasing social welfare, I think that software patents should not be granted.

In the current state of things, on the contrary of the US Patent and Trademark Officer (USPTO) that grant software patent without any technical innovations, the European Patent Office (EPO) does not grant software patent. In EU, to be granted, an invention must bring a contribution in a technical field (2).

I think that if software patents are granted in Europe, one will fall into the same problem that already exists in the USA. In the USA, we observe a software patent thicket indeed. The patent thicket is defined by the economist Carl Shapiro as «a dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialize new technology ». The overlapping software patents lead to several negative consequences (3), one of these consequences is the blocking innovations because of the fear of lawsuits. The example of smart phones wars in the USA that routinely erupt is relevant. «The smart phone patent wars have produced a large number of US lawsuits involving major industry players like Samsung, Google’s Motorola Mobility divi- sion, and Apple, with many counterpart suits filed overseas» (4). Nowadays, in the USA, it is difficult to innovate without having the fear of being suited by software patent owners, which would lead the innovators to a long and expensive contentious road.

To conclude, instead of putting energy on trial, it would be wiser to devote energy to innovate. Meanwhile firms sue each other, they forget that one of the main purposes of creating softwares is to develop powerful tools for innovation. Unfortunately, holders of software patents have often an incentive to use their software patent as a sword instead of using it as a shield.

References :

(1) Belleflamme, P. and Peitz, M. (2010) Industrial Organization: Markets and Strategies.
Cambridge: Cambridge University Press.

(2) http://www.epo.org/news-issues/issues/computers/software.html

(3) The Software Patent Thicket : A Matter Of Disclosure, Rosa Maria Ballardini, SCRIPT-ed – A Journal of Law, Technology & Society, 2009, Vol. 6-2, pp. 207-233.

(4) Stuart Graham and Saurabh Vishnubhakat, Of Smart Phone Wars and Software Patents, Journal of Economic Perspectives, Volume 27, Number 1, Winter 2013, pp. 67–86.

In my opinion, software patent mustn’t be granted too easily, thus I agree better with the EPO approach which restricts the number of functional patents. Indeed, I agree with the idea that this kind of patent could restrict the future research and hence slow the innovation’s growth.
As we seen in this article, the US grant a lot of patent that the owners use them to prevent the others to create different software which use the same function but the creation is really different.

I think patent would exist because they allow the owners to use their invention without others benefit from the efforts of this one. And hence patent give greater incentive to innovate.

In terms of duration, I think that patent mustn’t be too long because in the field of technology, the emergence of new innovations is quicker. A patent granted for a too long period becomes useless after 2 (for example) years because of new innovations more interesting.
An other argument is that this 2 years gives an advantage to the owner to develop (in advance) other innovation based on this patent.

To conclude, I think patent for software is required to give greater incentives to innovate. But, we must be careful and not deliver a lot of them because we could have a disincentive to innovate.

Patent law allows the owner have exclusive rights about manufacture, use, transfer, licensing and marketing of their patent, it have very strong protection of software. However, patent protection requires that the software has novelty, creativity and practicality, but 90% of the software is difficult to meet this standard, thus, in order to most of the software cannot obtain patent protection. And software patents need to go through a strict examination and approval, this approval at least one year to complete. But, the economic life of software is often only one or two years, after the patent is approved, the software probably already eliminated by the market. Even if the software to get a patent, but the patent maintenance costs are also quite high. So we can see the software patent system doesn’t work very well.

About patent duration, i agree that apatent covering software should survive for a term of five years. Because the software updates quickly, and the economic life of software is often only one or two years, and the patent maintenance costs are quite high.

For patent breadth, i think software patents cover a very broad, but they should be more specific and narrow it down. This happens because those large organizations tend to use these patents for legal warfare. This patent is actually preventing rather than encouraging innovation. And even worse is that small companies or individuals can not afford time and money of patent lawsuit. Finally, software patents defend a monopoly.

Having a patent system have a lot of disadvantages.
If we haven’t this system, I don’t think it would slow down the innovation process. The patent itself isn’t a lasting profit, it’s the realization of the new product which brings the true profit.

The patents only slow down the production process and are harmful for the customers. They are some artificial way to protect the companies which are not able to quickly launch the new product or which are not able to make this idea profitable.
Furthermore they create a new market on the true one, which add virtual added value where the true one is the result of the idea: the finish product.
However, they are some sectors where the patents are useful as other constraints slow down the release of the product as in the pharmaceutical sector.

But for sectors like software, the scope of the patent is too difficult to define and instead of still increase the complexity and slow down the innovation maybe we should simplify and remove the patent system. Moreover this sector move so fast that the time to write and confirm the patent a new solution is often available and free as open source.

I think there isn’t a right length or a right breadth of all patents. Indeed innovations are very different according to the sector of activity. In some sectors, to develop a new product is much more tough than in other sectors. Time, costs, human resources, etc. are completely different if we innovate in pharmaceutical or in new energy than in software, app, etc.

A new way of energy conservation (or a new vaccine) requires several years of experiments and effort, a lot of staff, resources and money. Moreover when we start this kind of project, we have neither any certainty of success, nor possibility of clearly plan it. On the other hand, if we develop new software most of these parameters are measurable, or at least more predictable. Basically for an app we only need some programmers, a designer and some good computers with adequate software. We are far from the lab, the doctors, the searchers, the materials, etc. needed in energy or pharmaceutical!

Thus we need to have more incentive to innovate in the firsts sectors than in software development. That’s logical; any manager will evaluate risks of his project before launching it. We can clearly identify here which one is more risky; pharmaceutical/ new energy ones.
That’s why different lengths of patent seem to be appropriate. This incentive is mainly an economic incentive. Indeed how is it possible to convince investor our project is profitable if we know that a few months after the commercialization of our product, our competitor will already have the same? In this case, companies will decide to wait others firms develop new ideas and they’ll “just” keep it and develop it a little bit more…

Thanks to patent, companies have interest to develop products. In easily “copyable” environment (like pharmaceutics where we can easily split up components of a medicine and define which molecules are used) a long length is required! Seeing that we need to patent medicine several years before its commercialization (as soon as tests), numerous years of patent protection are useful.
Considering software for the apps, which are developed in a few months, a patent protection of some years is more appropriate (on the software itself, but for the app a “normal” length is fine). Indeed in these few years the software and resources involved into will be profitable, and the app will continue to yield to its creator.

Here I focus me only on two extreme sectors. I’m convinced it’s possible to generalize it to many innovations, creations, etc. We can find a system of “return of investment” (not only financial one!) which granted the length (and the breadth) of the patent to the need of development resources. It’s obvious that actual system is easier to manage, but the suggested one will incite people to innovate if they have a (more) clear vision of how they will be protect by patent!

Patents have a double purpose: they protect R&D investments and they facilitate the diffusion of knowledge and the use of new innovations. Since the disclose of patents gives detailed insight about the innovation, other researchers can indeed benefit from the publication of the patent and even future research can be accelerated by building on existing work. Patents also allow companies to gain a competitive edge over their competitors by giving them exclusive rights over their innovations. Patents are given for inventions that are of practical use, show an element of novelty and include an inventive step that cannot be deduced by a person working in the same field. While reviewing these criteria might have been easy for something mechanical (which is what patent law originally strove to protect), it is much more difficult in check for software ideas.

The European Patent Office’s therefore restricts patent protection for software providing business process since these types of software are simply composed of algorithms, of math formulas, rearranged in a new way. In these cases, software patenting may actually hinder innovation by not allowing everyone access to them. Furthermore, in the software industry, the development of new software is sequential and mostly depends on preceding work. By setting a high patent fee on software, the patent holder can actually impede future progress. Another point is that patents also can give rise to monopolies. Since August, New Zealand has adopted the same approach as the EPO for software patents. (http://www.torontostandard.com/technology/new-zealand-says-buh-bye-to-software-patents).

For software patents including technical processes, the WTO’s Agreement on Trade-Related aspects of Intellectual Property Rights (TRIPS) has fixed the patent protection length at 20 years. In my opinion, 20 years are far too long for software patents. Even 3 years are a very long time in this highly growth-intensive and dynamic industry. They should be as short as possible and fixed at maximum 2 years in order to promote innovation.

I’m not an expert regarding granting of software patents, but after reading the article I get the impression that the situation in the US is not optimal and the amount of granting by the USPTO should be decreased. Even though the EPO follows a more restrictive policy and only grants software patents to processes solving technical problems rather than providing a „business service“ using existing computer technologies, I think the level of restriction is still not sufficient.

As for ways how to decrease the amount of granted patent protection, I sympathize with the criteria mentioned in the article.

Regarding the aspect of patent length, I very much agree on the proposition by the Electronic Frontier Foundation. From my point of view, 5 years is a duration that fits the subject matter of nowerdays’ digital technologies, whereas 20 years is not applicable at all – to put this in relation, the internet was opened up for public/scientific use about 20 years ago which resulted in a big increase to the speed of change with respect to the founding/launch and termination of companies and services. Almost everything we would summarize as related to digital technology has been created since, supporting the thesis that companies or services related to digital technology have a much shorter lifespan. To provide another hypothetical example, if Apple had managed to cover every single technical aspect of the iPhone in 2007 and in doing so prevent competitors from creating similiar devices by using wider-breadth patents, there wouldn’t be any kind of competitive smartphones for the next 20 years. More importantly, it would have resulted in a rather big deadweight loss for the consumers, as the smartphone market wouldn’t have grown to its current extent.

Breadth somehow is a more complex issue to face. In his paper, Prof. Lemley proposes to only provide patent protection for processes to achieve a certain goal, rather than the goal itself. In my opinion, this approach would effectively decrease the amount of overlapping patents, but in the same time require the companies to innovate as there is a limited amount of ways to achieve a certain result. However, this would also require the patent offices to execute a more precise analysis of every process’ functionality and sub-steps to achieve their respective goal.