[SWLUG] Software patents - what you're up against

[Dafydd Walters]

To those who may be going to the upcoming DTI meeting, if you haven't 
already seen this at LWN, this article highlights some of the arguments 
you're going to up against:

http://lwn.net/Articles/113518/

Also, if you weren't already aware, the UK patent office has published a 
4-page "fact" sheet (quite a bit of which is actually the UKPO's opinion) 
which most of the attendees will likely have read: 
http://www.patent.gov.uk/about/ippd/issues/softpat.pdf  To the casual 
reader, the points made in this document are likely to seem quite reasonable 
(and in fact some of them are), but the UKPO advocates keeping the status 
quo. The "status quo" probably means allowing the tens of thousands of 
frivolous and broad patents on software ideas that have ALREADY been granted 
by the European Patent Office to stand unchallenged - see the European 
Software Patent Horror Gallery here: 
http://swpat.ffii.org/patents/index.en.html .

The UKPO's document does not address the fact that the Irish presidency 
directive proposal is so badly worded (it expressly allows so-called 
"program claims"), that it would, in all likelihood, actually open the 
floodgates for broader software patentability.  What the directive is 
lacking is verbiage that clearly defines what software is (and therefore 
what cannot be patented).  Processing, handling and presentation of 
information should be explicitly excluded from patentability, as should 
innovations in the field of data processing. Furthermore, to make sure that 
"pure software" can never be patented, the directive needs to spell out the 
fact that an invention requires that natural forces are used to control 
physical effects beyond the digital sphere (e.g. you can patent the way an 
innovative garage door opening system works even if it contains software, 
but you could never patent software itself).