[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On Wednesday 15 March 2006 1:31 am, Benjamin A'Lee wrote: > On Tue, 2006-03-14 at 16:58 +0000, Neil Williams wrote: > > There is now no barrier at all. The objections have been simply ignored > > and the UKPO has a workaround catch-all that both enables them to say > > they don't allow software patents and at the same time allow every > > software patent application that includes a claim that the idea preceded > > the creation of the software implementation. > > As in, everything? It's fairly hard to have an implementation *before* > the idea. Precisely, that's why this is so crazy. Some of the responses on the uk-parl list at FFII are quite helpful but the archives aren't public (so I can't quote here). If you subscribe here: http://lists.ffii.org/mailman/listinfo/uk-parl you can view the archives. The judgement is seriously flawed, put it that way. Prior art - forgotten, application - examiner doesn't appear to have understood the claim, simple logical appraisal of the claim - beggars belief. http://www.patent.gov.uk/patent/legal/decisions/2006/o05706.pdf Take this snippet: (brackets denote a comment from the examiner who found in favour of Sun): "A Java virtual machine is commonly implemented in software by means of an interpreter for the Java virtual machine instruction set but, according to the [patent] application, it may be software, hardware, or a combination of both. (This is still something of a mystery to me. I thought the fundamental attraction of a Java virtual machine was that it enables a program [sic] written in the Java programming language to be executed on a wide range of hardware platforms without requiring modifications or additions to the hardware. I find the concept of a 'virtual' machine implemented in hardware to be inherently confusing.)" Duh! How does he think the Java code is executed on so many different platforms without some kind of virtual interpreter that may or may not be in hardware!?!?!! The VM IS the layer between the 'universal' code and the underlying hardware. It matters not whether that VM is expressed in hardware or software. e.g. winmodems - with the correct software driver, the hardware works fine. It's a split device yet it performs precisely the same functions (with a few extensions) as an external/serial 100% hardware modem. The whole point of Java on a mobile phone is that the VM can be, at least partially, in hardware. The fundamental advantage of the Java LANGUAGE is that programmes written in Java can be executed on any platform that HAS a Java VM. It's the VM that does the hard bit. Yet without the VM, the Java classes themselves are quite, quite useless. Simon? You're the Java proponent here - have I got this completely wrong? > > "There is no doubt that the invention as claimed would involve a computer > > program for its implementation; the applicant also says that this is the > > case. But as CFPH indicates, that does not establish, in and of itself, > > that the invention is not patentable," the UKPO representative explained. > > If the only possible implementation is computer-based, then surely it's > a software patent (a patent that covers inventions that can never be > anything but software...). Ah, but anything we can do in software can also be done in hardware which means it's not software - or is it? See the problem? Being computer-based is no protection - the patent law only forbids software patents. This was meant to protect algorithms but we all know that an algorithm can be implemented in a chip as easily as a binary. It's only a question of will and funding. Patents were explicitly designed to protect inventions like chips - physical entities a lawyer can hold in their grubby little hands. Patents were also explicitly designed to AVOID software - this is the crux of the issue. At what point is a patent an idea, a piece of hardware or a piece of software? You only have to *claim* that you *can* make a piece of hardware instead of a piece of software and you're home and dry with the UKPO. That's why mobile phones are the current battle-ground and why Nokia was/is lobbying hard in FAVOUR of software patents (which they would almost exclusively use in machines that actually employ a hardware/software combined solution). What Nokia want to prevent is someone implementing their hardware work in a purely software implementation - because hardware costs money to develop and deploy but generally runs faster than software. http://www.theregister.co.uk/2004/04/13/eu_patent_protest/ It's also why Sun have pushed so hard for this bytecode patent. > > It's worse than the USA system. The actual patent is for Sun's Java > > bytecode, the patent on which has now been upheld in UK law by the UKPO. > > Hmm. Wouldn't that make Microsoft's .NET illegal, since it's > bytecode-based? Can't imagine they'd be too pleased about that... The problem is the ruling that the UKPO cites as case law - the interpretation is dubious and the UKPO itself doesn't appear completely comfortable with the judgement (but seem happy enough to use it when it suits them). (Patent was filed in 2002 - was C# around then?) > > Democracy? What's that? > > Out of interest, are there any countries that *do* have sensible patent > laws? Interesting question. There was a determined fight to protect what little sanity is left in the UK/EU patent system: http://wiki.ffii.org/MutualRecognition060315En "Members of the European Parliament successfully opposed a move by a MEP to jeopardise the annual Lisbon report. He had proposed to ask the Commission for mutual recognition of patent law, which would have resulted in a flood of patent suits all over Europe, lower quality standards, and worsen the software patents problem. Such mutual recognition would make the patents that the national patent office of any EU country grants enforceable against companies in all other member states. [This would be, ] among other things because patent applicants would "shop around" to find patent offices that are most willing to grant patents which would then be valid in the entire EU." -- Neil Williams ============= http://www.data-freedom.org/ http://www.nosoftwarepatents.com/ http://www.linux.codehelp.co.uk/
Attachment:
pgppHGM0acsb4.pgp
Description: PGP signature