CaliuCampaign against software patents | Català

Work in progress at The European Parliament (Committee on Legal Affairs and Internal Market - JURI)

Rapporteur Arlene McCarthy's document on issues

The JURI committee working document [1] on need and issues at stake for the software patentability directive COM(02) 92 by Arlene McCarthy presents some doubts that we hereby try to address.The working document is part of the work of the committee on Legal Affairs and Internal Market of the European Parliament, that is the one in charge of the software patentability directive COM(02) 92.We have sent this comments on the study to the web forum of the committe [12], that is open to everyone to discuss the directive.

The FFII has more elaborated comments on McCarthy's documents, in the same line as these.

In the section "Why the Commission think a directive is necessary" it rightly explains there is serious legal uncertainity, but attibutes it to differing national courts decisions, instead of the divergence of EPO practice from EPO statutary law. See [2] for a detailed explanation for the Swedish case. See also the EPO workers union dennouncing their management for not allowing them to serve the public interest[5].

In section "some arguments", it asks for effects of software patentability in creativity and vitaly, and free software. Here's [3] a little bibliography. Software with a "technical effect" as the EPO judges it, means any software, it is only a requirement on the wording of the claims. This is also clear in the EP commissioned study by Bakels. This is why you only find references to perils of software patents, not of "software with technical effects" patents. The only criteria we know that works to separate patentable from unpatentable subject matter is the traditional technicity criteria, that requires a teaching on uses of forces of nature to achieve overseeable effects to grant a patent [4].

There are differences between software and material inventions that affect their economic balance and their patentability. The basic contradiction arises because of the fact that software is only information:

Patents are monopolies in the use of some new machine in exchange for publication of information on that machine. When the machine becomes software, and therefore machine=information, patents collapse in "monopolies in the use of some information in exchange for publication of information". So we are allowing people to monopolise information in exchange for publishing information that cannot be used because is monopolised. The fact that publication equals manufacturing in software (and other logical creations) makes patenting software (and all other logical achievements) absurd.

This epistemological contradiction leads to economic contradictions:

So when the document says:

      The arguments against patenting software which satisfies the
      tests of the EPC could well be levied at patenting in general.
      All patents grant a limited monopoly as a reward for invention.

it is not exact. There are specific arguments that affect software and not inventions. The same rules to judge the value of the patent system can give different results when applied to different subject matter. If this is not yet clear, you can find a somewhat more detailed explanation at [9].

But even if there weren't specific arguments for logical creations, the document recognizes that there are arguments which cast doubt on the usefulness of the whole patent system. So it would still be unwise to expand it to any new area.

The piece:

      Moreover, open-source software has flourished alongside the
      existing patenting regime in recent years. Again, under the
      present patenting regime many firms have invested billions of
      euros in R&D ­ investments founded on, among other things, the
      availability of patents.  It is argued that it would be very
      unwise to jeopardise this investment by reducing patentability
      of software.

is also inexact. Software has not been patentable anywhere during most of the time it has flourished, and patents have been little used when and where they have existed, except by some large players. Both free software and non-free software has flourished without patent protection. There are documented cases [6] of threats to free software by software patents. And in Europe, the effects of patenting software have been less than they would be if the directive passes, because software patents are still illegal under EPC and member state laws, and the issued patents are not as enforceable as the directive would make them. Expect even more trouble if the directive passes. If fact the Bakels study explains how patents may well appear because of software success and not the opposite.

The rest of the document is quite sensible and welcome. The legal certainity must be restored and the burden of proof is for those who want to extend the patent system (not many arguments are presented in favor of that move). Note that the directive does only introduce more legal uncertainity. By requiring a "technical contribution" without defining "technical", but abandoning the traditional and useful definition of technical character by stating that software is part of the technique, it in effect leaves decisions to the EPO arbitrarily, and prevents any possible balance by national courts or anyone else.

In fact the frequently asked questions documents [7] published by the DG Internal Market itself, proposes some examples and is unable to decide on the patentability of their own example based on the proposed directive (it says it would be highly likely, but not certain, that the Amazon one-click patent be rejected, and they don't want to interfere on the application processing at the EPO, thereby denying any true intention of driving EPO under democratic legislative control). It is amazing that they claim any legal certainity at all in their directive.

Then there are a few open questions that we'd like to try to address:

1.Is such a directive necessary or can we just leave matters to the Boards of Appeal and the national patent courts?

The directive is unnecessary and harmful. But it cannot be left to the EPO Boards of Appeal since it has demostrated erratic and arbitrary case law with a deceptive interpretation of EPC.

Leaving it up to national courts is more or less the current situation, and it lacks desirable legal certainity, so the best option might be to delete point 3 in article 53 of EPC (the "as such" clause). To make it clearer that software is not patentable. The point is merely clarifying what is implied by the rest of the Convention.

If fact the problem lies in lack of democratic control and selfish drive at the EPO, and the solution would not be so much in harmonizing national laws, that are already harmonized, but to execute measures to bring EPO under control, or replace it by some new EU institution. A list of measures to restore sanity into the patent system and avoid further patent inflation is in [8].


2.Will the directive achieve its ends, in particular without unwanted side effects

It wouldn't acheve the stated aims. The only goals that it would achieve would be greater control of the market by a few large international corporations, less innovation, and above all, more power and less accountability for the EPO. It would have lots of side (or frontal) effects on software innovation, vitality, quality, competition and freedom.


3. Does Art. 5 run counter to TRIPs in imposing artificial restrictions on the type of claims permissible in this field? Is it intended to be retroactive? If so, with what effects?

I don't think so. As the Bakels study says, TRIPs does not mandate subject matter. This allows us to return to the sane technicity doctrine of the 70s.

4. Is the proposal legally watertight (legal certainty)?

Not at all, they can't even decide certainly on their own example. It would make the system even more uncertain so that patent offices and patent professional could capture more revenue and power in being indispensable to navigate throught the formalistic rethorical but meaningless requirements.


5. Is there any merit in following in the footsteps of the USA (patentability of business methods) or should patentability of business methods be clearly and specifically precluded?

This is clearly rejected by everybody, and the USA could be expected to correct this abuse some day. Anyway, patentability of business methods is not a separate problem. Once you allow logical creations to be patented (like software), then any logical creation, in cluding business methods, can be claimed by dressing it in software claims. Business methods patents are already granted by the EPO without the legislator ever telling them to. This should be stopped.


6. Should the system of (compulsory) licences be reviewed to prevent abuse of the patent system?

Possibly yes, but this is an independent problem. You cannot build an inconsistent patent system and expect to fix it case by case with last reasource measures as compulsory licenses. Compulsory licenses should probably be used more than they are now, but cannot become the norm, since they would introduce even more legal uncertainity.


7. Is the issue of trivial patents a problem? If so, how should it be addressed?

Yes it is. It is very hard to adress, but the requirement of techinicity of the novelty would reduce patent inflation a little. The best way to address trivial patents would be to change the balance of incentives for the EPO, that now gets money when it grants a patent and doesn't when it rejects it. Addressing them by giving ever more resources to an incompetent EPO should not be tried. A system should be devised with proper balances in place, so that the players regulate themselves. Social games have been proposed where the patentee pays anyone who can show the triviality of their patents so that patentees are encouraged not to patent trivial ideas, and the public is encouraged to debunk them.


8. What risk, if any, is posed to open-source software? If so, how is it that open-source and proprietary software seem to co-exist at present?

Risk is already present [6] and would increase if the 30000 EPO patents on software would be enforceable. Expect more trouble if most of the world accept software patents. Nowadays it is not clear what national courts will say, so the risks are somewhat less because patentees refrain from going to court. Cease and desist letters and other private threats are already damaging, but not so easy to detect. But it is not only free software that is at risk. Any software developer would be seriously restricted from exercicing his or her profession.


9. Is it possible to argue that patents may stifle innovation, if so how?

It is argued in many studies [3], and a little bit above. Patents would make basic ingredients of software innovation scarce, and would criminalise creativity. Software innovators would be more busy trying to find out whether they infringe some patent or not than innovating. See also [9] and [10].. Barriers to entry would be higher, so fewer creative individuals and startups would be able to find revenue from their creativity.


10. Is it possible to quantify in economic terms/employment the benefits/disbenefits of software patentability?

Not exactly, so far as I know. But the arguments are compelling enough to show that disbenefits outweight the (nonexistent) benefits. For example, it is clear that SMEs would be damaged, and 2/3 of workers in information and communication technologies in Catalonia working in firms with more than 10 employees are in firms with less than 200 employees [11]. More statistics would be needed that I don't have, and for the whole of Europe, but the impact would be huge.

Other businesses (not software) would also be affected by more expensive, less innovative software, and of lower quality, due to reinforcement of monopolies. But I cannot quantify it.


11. What impact, if any, will action in this area, one way or another, have on SMEs?

Software patentability would be damaing to SMEs by increasing their costs, reducing their capacity to innovate and compete and increasing barrieres to entry and bureaucracy. Reinstating software as not patentable would allow them to work in a more level playing field.

12. What is the impact of TRIPs?

I don't know. I'd say none on subject matter.


13. The EPC is not limited to EU countries. What are the implications of this in the event that the directive is adopted?

Legal uncertainity would increase in the EU and not necessarily in other EPC countries. The EPO would be reinforced and maybe would have more ease in setting policy for all of them. By the way, many candidate countries to the EU are not in the EPC. This is one more reason to replace the EPO with an EU dependant patent office under parliamentary control.


14. Do recital 18 and Art. 6 of the proposal for a directive need to be reworded in order to allow decompilation of programs for the purposes of interoperability as is permitted under f Directive 91/250?

I think not, but I don't know. ANyway, patents should not affect software, so it isn't important, I think.


[1] JURI Committee Working Document by Arlene McCarthy

[2] CEC & BSA trying to impose unlimited patentability on Sweden

[3] Bibliography on software patents

Also on software patents effects:

[4] Technicity criteria
Dispositionsprogramm decision by the German Federal Court

[5] EPO workers union dennounces EPO out of democratic control and not serving public interest

[6] Effects of software patents.

[7] DG Internal MArket FAQ on the directive and press release


[8] Eurolinux demands
Many languages:

[9] Caliu position on software patents. Context and problems summary. Dated March 1st 2002.
Proinnova version of the document, slightly more up to date.

[10] One reply to the Comission Consultation

[11] Statistics for Knowledge Economy Information Society in Catalonia.

[12] WEB forum of the Legal Affairs and Internal Market committee of the European Parliament.
Topic number 6 starts the discussion on software patentability and the proposed directive.

$Id: mcCarthy-opinion.html,v 1.13 2004/05/28 09:28:29 xdrudis Exp $