Den 29. aug. 2012 kl. 19:11 skrev steve harley: > on 2012-08-28 23:25 DagT wrote >> >>>> No, your theory works for small software inventions, but those are not the >>>> complete picture. You need a better definition. >>> >>> i think your point is not disputing my definition, but rather returning to >>> the question of whether some or all software should be patentable >> >> No, I am simply looking for a definition. In fact, I do agree regarding >> small purely software inventions, but their complex relatives really need >> patents. That is why it is difficult. > > i think it boils down to your desire for, as you stated, a "fair" definition; > i think the problem on your side is one of defining what you mean by fair — > if you can do that, you might have your answer (and the fact that it has to > meet your definition of fair is why you cannot accept others' definitions); > from my perspective it is much simpler because i think it's "fair enough" to > simply disallow any software patent
Well, then you think software as program lines, but that is not as simple as it sounds. The reason was illustrated previously (I cut out some text here :-) but anyway I am also putting myself in the position of the peoples who make the rules, as I sometimes have the job to bend them. Unless you get a clear definition it is impossible to make an effective rule. Bad rules turn out very expensive for the people with the least money. >>>> Another thing is, of course, that computer programs may be implemented as >>>> hardware... >>> >>> if you mean as firmware, or as a configuration of an integrated circuit, >>> it's still software if it is a description (loosely speaking) of inputs, >>> outputs and a logical sequence; and if you are talking about specific >>> computer programs, they are covered by copyright (copyright laws, at least >>> in the US, are also flawed, but i don't object in principle to software >>> copyrights) >>> >> No, it is very possible to make a machine that perform logical sequences. > > yes it is, but i was trying to parse your proposition into something that > made sense; i'd be interested if you can describe a software patent that can > be meaningfully implemented in a machine; and you can copyright the sequence, > and patent the machine, but you can't patent the sequence We did play a bit with that in my job. Heres a link to a toy: http://www.trademarkfactory.no Heres a link to the patent: http://worldwide.espacenet.com/publicationDetails/originalDocument?CC=WO&NR=2007133087A2&KC=A2&FT=D&ND=3&date=20071122&DB=EPODOC&locale=en_EP As you can see the patent is both a one armed bandit with exchangeable signs and a software invention :-) But that is a crued example, the more elegant is that you can make microchips with burned in, non-changeable, circuitry performing logical tasks. Actually, many gear systems may be described as electronics, hydraulics or or software. The technological implementation is not related to the invention. > > >> Copyright is not effective as you loose any protection by simply rephrasing >> the code, not reinventing it. > > it is effective enough from my point of view (too effective in some ways, > e.g. shrink-wrap licenses with arcane restrictions) Copyright is related to the written code text, no matter what is describes or if it works. >> Anyway, there are some very strange things in the US copyright law. > > i totally agree > > and i appreciate your perspective — it sounds as if you work in areas where > software patents are effective (if still imperfect) incentives, whereas i, > though i am a software developer, am mainly considering impacts on the > evolution of software as a cultural phenomenon; my position is that on > balance the patent system works against the ideal of "incentives for > innovation" in the area of software; i think the software market may > naturally have enough incentives, and if it doesn't, some system other than > patents may better solve the problem OK, I think my last word here is that my work is helping small firms protect their inventions. In many cases patents are the only was to get investors, and also the only way to compete with large firms who otherwise simply copy their ideas. Either by themselves or by making alliances with others. As I think Steve Jobs said small firms are better at finding great ideas as they have to find new fields not dominated by the large ones. That is why Apple, and many others, simply buy these small firms instead of trying to do it themselves (and many small firms want this). So we use it as a tool to help the small ones. The reason why I usually don´t say to traditional programmers that they should file patent applications is simple: Software patents have uncertain validity, as it is difficult to find out what is known already. So their value is limited. That is, however, in Europe. A representative for Microsoft ones said that the reason why they filed so many patent is as a defensive precaution agains US patent trolls. It´s a lot cheaper than getting into some strange case against some guy with a patent he never use for anything else than suing others and a lawyer who gets payed part of the "triple damages". If you have a lot of patents yourself there is a chance that you may say: "Yeah, maybe we infringe your patent, but you certainly infringe a number of ours". If you look very carefully neither Apple nor Microsoft take many small software firms to court. It is bad for publicity. The war between Samsung and Apple is something very different. DagT -- PDML Pentax-Discuss Mail List PDML@pdml.net http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
