In this mostly-quiet period, I've been thinking about what harmed/harms our software patent lobbying, and I'm wondering if it's our complete reliance on the term "software patents".
We've invested a lot of energy in that term, so it should be kept, but it's practically impossible to have a meaningful conversation with the European Commission or EPO about "software patents". They just use their own ludicrous definition and tell everyone that they agree with us and that current and proposed legislation doesn't support "software patents" etc. etc.
Then they say that they do support Computer Implemented Inventions, and they define that in a broad way that encompassing all inventions that make use of a computer - and we're left unable to say that CIIs are bad.
...but it doesn't have to be this way. Terms could be avoided altogether by just mentioning "problematic patents in the software field" and then giving a one-sentence explanation like "such as patents on how to display, store, or communicate data".
So what I'd like to ask is, do people think this would help, and can anyone suggest better wordings? Does that one-sentence explanation pretty much cover what we care about (without covering something we don't have a position on)?
(Keeping in mind that the more precise it gets, the more excuse the EC/EPO person has of using a "I don't know the exact technical details of this" response.)
Meanwhile, things are happening in India. Their patent office has published a draft for a new manual, which can be viewed along with comments from the likes of Microsoft and Redhat here: http://www.ipindia.nic.in/ipr/patent/manual.htm
I think we need a new stratagy to tackle the problematic aspects of software patents, and not software patents themselves.
The main reason that patents are so problematic in software, is that the speed of advancement in software technology is so much greater than in the traditional fields that patents were designed for.
Even if we do stop software patents, people will come up ( and have come up with ) with work-arounds that patent the raw methods, and don't mention the fact that they are software-implemented.
I think what we need is a clause, that allows the free implementation of patents in software after a shorter time period ( maybe 2-3 years ), which would basicly mean that software patents ( whether labeled as such or not ) would effectively expire in 2-3 years.
2008/7/1 Ciaran O'Riordan ciaran@fsfe.org:
In this mostly-quiet period, I've been thinking about what harmed/harms our software patent lobbying, and I'm wondering if it's our complete reliance on the term "software patents".
We've invested a lot of energy in that term, so it should be kept, but it's practically impossible to have a meaningful conversation with the European Commission or EPO about "software patents". They just use their own ludicrous definition and tell everyone that they agree with us and that current and proposed legislation doesn't support "software patents" etc. etc.
Then they say that they do support Computer Implemented Inventions, and they define that in a broad way that encompassing all inventions that make use of a computer - and we're left unable to say that CIIs are bad.
...but it doesn't have to be this way. Terms could be avoided altogether by just mentioning "problematic patents in the software field" and then giving a one-sentence explanation like "such as patents on how to display, store, or communicate data".
So what I'd like to ask is, do people think this would help, and can anyone suggest better wordings? Does that one-sentence explanation pretty much cover what we care about (without covering something we don't have a position on)?
(Keeping in mind that the more precise it gets, the more excuse the EC/EPO person has of using a "I don't know the exact technical details of this" response.)
Meanwhile, things are happening in India. Their patent office has published a draft for a new manual, which can be viewed along with comments from the likes of Microsoft and Redhat here: http://www.ipindia.nic.in/ipr/patent/manual.htm
-- Ciarán O'Riordan (+32 477 36 44 19) \ Support Free Software and GNU/Linux http://ciaran.compsoc.com/ _________ \ Join FSFE's Fellowship: http://fsfe.org/fellows/ciaran/weblog \ http://www.fsfe.org _______________________________________________ fsfe-ie@fsfeurope.org mailing list List information: http://mail.fsfeurope.org/pipermail/fsfe-ie Public archive: https://mail.fsfeurope.org/mailman/listinfo/fsfe-ie
Ciaran,
Ar 01.07.08 02:13, Scríobh Ciaran O'Riordan:
...but it doesn't have to be this way. Terms could be avoided altogether by just mentioning "problematic patents in the software field" and then giving a one-sentence explanation like "such as patents on how to display, store, or communicate data".
So what I'd like to ask is, do people think this would help, and can anyone suggest better wordings? Does that one-sentence explanation pretty much cover what we care about (without covering something we don't have a position on)?
I think there's a danger that this explanation could be twisted to claim that we're trying to block patents on display, storage and communication devices, which will almost certainly have a software or firmware element -- imagine a wireless digital photo frame with some novel aspects ( now patent it! ;). In such a case the software shouldn't be patentable but the device may be. If the device is patented as a whole the software elements shouldn't be enforceable.
The explanation should also cover "manipulation" or "processing" of data in general: we still care about algorithms.
How about "such as patents on software to display, process, store or communicate data"?
Hope this helps,
David
"Rory Browne" rbmlist@gmail.com writes:
I think what we need is a clause, that allows the free implementation of patents in software after a shorter time period ( maybe 2-3 years ), which would basicly mean that software patents ( whether labeled as such or not ) would effectively expire in 2-3 years.
This is difficult to ask for since the TRIPS agreement requires that patents last 20 years.
Another problem is that the start of a patents' term of validity is back-dated to when the application was made, IIRC. Because the EPO is so overloaded, application processing time is currently measured in years. So a 2-3 year duration might be completely unacceptable to the EPO since it would lead to patents that are expired-on-grant (we would have a hard time selling this idea to politicians). This should be fixed, but we might be, ah, spraying water against the wind if we try to fix the general the problems of the European patent system while trying to fix our specific problem.
And if the pharma lobby get worried that this idea will spread to other domains, we could gain new powerful enemies that we don't need.
This is difficult to ask for since the TRIPS agreement requires that patents last 20 years.
Perhaps - in which case you play them at their own game. The 2-year software limitation clause would only affect software patents, which are disguised as non-software-patents. The patent itself would be valid for the full 20 years for non-software implementations.
Alternatively just exempt software-implementations from the get-go.
Another problem is that the start of a patents' term of validity is back-dated to when the application was made, IIRC. Because the EPO is so overloaded, application processing time is currently measured in years. So a 2-3 year duration might be completely unacceptable to the EPO since it would lead to patents that are expired-on-grant (we would have a hard time selling this idea to politicians). This should be fixed, but we might be, ah, spraying water against the wind if we try to fix the general the problems of the European patent system while trying to fix our specific problem.
On the other hand, they may welcome such a limitation on patents, since the patents would be less valuable, leading to a reduction of applications, thereby easing their workload. Then again, maybe the EPO has a different sense of logic to me.
And if the pharma lobby get worried that this idea will spread to other domains, we could gain new powerful enemies that we don't need.
-- Ciarán O'Riordan, +32 477 36 44 19, http://ciaran.compsoc.com/
Recent blog entries:
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_software_notes/openstreetmap...
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_software_notes/launching_you...
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_software_notes/updating_debi... _______________________________________________ fsfe-ie@fsfeurope.org mailing list List information: http://mail.fsfeurope.org/pipermail/fsfe-ie Public archive: https://mail.fsfeurope.org/mailman/listinfo/fsfe-ie