The team is joined by Guest Kats Rosie Burbidge, Stephen Jones, Mathilde Parvis, and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Hayleigh Bosher, Tian Lu and Cecilia Sbrolli.

Wednesday, 7 May 2008

Software patents: all talk

The IPKat has just listened to an interesting talk from BP/G radio on the subject of software patents in the US and Europe, involving a conversation between US attorneys from Merchant & Gould and the well-known (and softly spoken) UK and European patent attorney Paul Cole. No conclusions are arrived at (of course), but it might give the listener some food for thought in preparation for tonight's Nodus meeting, which includes a presentation on the same subject by Jeremy Reed of Hogarth Chambers.


Anonymous said...

Software patents are good because they are a cheap weapon to get rid of small players.

Anonymous said...

It was an interesting talk. I was again struck by the parallels between patent lawyer discussions of (software) patent issues and quack/woo discussions of homeopathy etc. The economics of the patent system and any consideration of whether it is actually beneficial or not are always absent in the former - just as (genuine) science and the limits to the utility and applicability of CAM are always absent in the latter.


Anonymous said...

NB: it's Paul Cole, not Coles.

I'm always struck by the parallels between the MMR scare and software patents: a few inflated anecdotes but no scientific/economic evidence cause mass panic and confusion which takes years to sort out, leading to widespread loss of faith in the medical/patent community as a whole and accusations of conflicts of interest and self-serving behaviour (on both sides).

Another parallel:

The flames of the MMR scare were fanned by separate concerns over the use of a mercury-based preservative in some US vaccines, even though the UK's MMR vaccine did not use that preservative.

The flames of the software patent scare are being fanned by separate concerns over US businss method patents, even though the UK(I)PO and EPO would not allow such patents.

Anyone can draw parallels and spout anecdotes. Both mean nothing.


Anonymous said...

"Anyone can draw parallels and spout anecdotes. Both mean nothing."

That depends on whether or not they have a useful point to make. My point about the parallel between the science free fantasy world the woos live in and the economics free fantasy world many patent proponents live in is neither meaningless nor original:

I wouldn't say that your parallel between MMR scares and software patent scares is meaningless either. However, the idea that the anti-software patent argument amounts to nothing more than an anecdote fuelled scaremongering exercise is badly flawed:

The parallel between MMR and software patents also falls down badly on consideration of where the burden of proof should lie in each case.

Lastly, it is simply not true to say that the software patent 'scare' is dominated by concerns about business method patents (or 'bad' patents, as some would contend). Nor is it at all credible that business method patents would not be or are not allowed by the EPO. They have granted them in the past:


Anonymous said...

Thanks for that.

Giving a list of links that fail to provide any economic evidence that software patents harm innovation or that the UK and Europe are granting business method patents unless you're willing to call a franking machine a "business method" proves my point entirely.

If your economic evidence is the deeply flawed (as pointed out by several other economists) Bessen/Hunt Working Paper mentioned in one of the links (a report that effectively defines "software patent" to mean any patent that happens to use the words "computer program" somewhere in the description and therefore includes patents covering the PS2 and forced B/H to create some interesting theories about why hardware companies are filing so many "software patents") then I would point out that they said themselves that there have been insufficient economic studies.

If you actually look at the total range of published information available, rather than just cherry-picking a couple of reports that support your view, then the situation is more complex than you seem willing to admit. I've not read all the different reports myself, but I have read several "reports of reports" by various academics, and the conclusions are usually along the lines of:

1. Nobody agrees on anything, particularly what is actually meant by a "software patent"
2. People get very emotional about the topic.
3. Insufficient hard research has been done and the research that there is is full of personal opinions, not facts.
4. Patents on computer software are different from patents on other types of invention, but not all that different.
5. Knowledge of the prior art and experience of examiners is improving to ensure increasing patent quality.


Anonymous said...

"Giving a list of links..."


a) I wasn't even attempting to provide (links to) economic evidence that software patents harm innovation. I provided those links in order to refute your eccentric assertion regarding the EPO and B/M patents, and to illustrate my points about the nature of the debate.

b) I'm not willing to call a franking machine a "business method", no. Nor, I expect, is the author of that paper on B/M patents in Europe which contains a chapter on business method patents - not franking machine patents - in the franking machine industry.

Anyway... It is gratifying that you do seem to take an interest in the economics. It would be even more gratifying if, instead of constructing elaborate strawman arguments based on things I never said and links I didn't post, you would take your own good advice. Rather than cherry-picking and exaggerating, try reading a little more of the literature (including the up to date stuff). Then, bearing in mind where the burden of proof rests, convince us all that, overall, the benefits of software patents do indeed outweigh their costs.


gerd said...

Software and business method patents are the scum of the patent system.

The most stupid defense against kicking them out: We don't know what software patents are, define software patents.

We don't need definitions, we point at your shit and we want to get rid off it, feel free to make your definitions of what constitutes "real shit", whether it needs to stink or where the 'shit stinks' tolerance level for peasants is,

The real solution is of course to bail out the software patents and the parasites who want them. If the lawyers want they would propose a clean cut rather than to pretend to be stupid. Else others would do it for them, it would be 'bloodshed'. Make your choice...

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':