Weekly Edition
Return to the Press page
| |||||||||||
Bruce Perens expounds 'Open Source State of the Union' (NewsForge)
NewsForge
covers Bruce Perens' Open Source State of the Union speech
at LinuxWorld.
"Perens feels the biggest challenge to open source going forward is software patents. In the U.S., 50% to 95% of software patents should not be granted, he said, because they are not inventions and are written extremely broadly. He expects that after SCO suit is over, we'll see a number of patent lawsuits brought against Linux."
(Log in to post comments)
Patents Posted Jan 22, 2004 23:55 UTC (Thu) by ccyoung (subscriber, #16340) [Link] does this mean that the rest of the world will sail on with Linux while the US continues to shoot itelf in the foot and pay the MS tax?
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 22, 2004 23:58 UTC (Thu) by brugolsky (subscriber, #28) [Link] Bruce undermines his argument with nonsense statistics like "50-95%". Stated another way, either 1:2 or 1:20 patents are valid. That's a pretty big range - which is it? The answer is relevant to any discussion of the economic value of patents.Even more absurd than the number of software and business method patents granted is the 20 year monopoly on an idea.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 11:17 UTC (Fri) by kcowtan (guest, #18913) [Link] "That's a pretty big range - which is it?"That is an unfair criticism. His meaning is very specific: He belives that 50% are definately invalid, and 5% are definately valid, and he doesn't know about the other 45%. Pulling an arithmetic stunt to conceal his meaning doesn't help anyone.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 12:16 UTC (Fri) by Differance (guest, #18916) [Link] Even more absurd than the number of software and business method patents granted is the 20 year monopoly on an idea.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 11:23 UTC (Fri) by bug1 (subscriber, #7097) [Link] 'Bruce undermines his argument with nonsense statistics like "50-95%"'It would undermine his argument if he gave a single figure and tried to convince us that he had the ability to acurately judge all granted software patents and determine their legality. I imagine his numbers are based on a few studies by others, he may have got his range this way. If more than half of software patents are invalid then the exact number is irrelivant, its serious evidence that the USPTO is negligent and incompetent. It would be more just if the USPTO secretly used some sort of random number generator to detemrine if a patent should be granted, that way the long term average of invalid patents would hover around 50%, probably an improvement on the current situation. I dont suppose the USPTO can be held responsible for their actions ?
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 12:15 UTC (Fri) by Differance (guest, #18916) [Link] Even more absurd than the number of software and business method patents granted is the 20 year monopoly on an idea.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 24, 2004 2:41 UTC (Sat) by flewellyn (subscriber, #5047) [Link] Even more absurd than that is the idea of comment spamming us with the same message all the time.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 15:52 UTC (Fri) by donwaugaman (subscriber, #4214) [Link] > It would be more just if the USPTO secretly used some sort of random number generator to detemrine if a patent should be granted, that way the long term average of invalid patents would hover around 50%, probably an improvement on the current situation.No, that would be worse. If there were a simple random chance of getting a patent approved, and no penalty for nonapproval, the percentage of invalid patents would approach 100%, as there would be nothing to stop people from patenting really trivial things (or even repatenting stuff that had already been patented - refile LZW, anyone?) and anything could be patented in an average of two filings. As broken as the software patent system is, such hyperbole is unwarranted.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 14:58 UTC (Fri) by aleXXX (subscriber, #2742) [Link] Yes, good choice to exclude a huge range of apps just because they link to a GPL lib (Qt): k3b, quanta, kdevelop, konqy, ... many moreThe standard response to e.g. k3b on UserLinux is: "we won't include it, instead we'll rewrite it for gtk !" But they don't deny to include apps using other toolkits ... OOo, Mozilla, ... Yeah, ... a united linux community.... Bye
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 15:01 UTC (Fri) by ber (subscriber, #2142) [Link] In the U.S., 50% to 95% of software patents should not be grantedThis is a bad thing to say. The system is fundamentally flawed as it opposes innovation. A finding which the federal trade commission suppored in their report from last october. No patent on software should be granted. If the summary is true, then Mr. Perens believes in the software patent system.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 17:46 UTC (Fri) by JoeBuck (subscriber, #2330) [Link] I agree that software patents are a big problem, and almost all software patents are socially harmful. However, the problem with saying that there should be no software patents is that, because of the way the world is going, you are essentially saying that there should be no patents on digital circuit production either (at least up to the point where the design is handed over to the fab). Digital logic is produced by software techniques (starting from Verilog or VHDL, which can be thought of as parallel programming languages); embedded software burned into ROM is really just hardware; an FPGA (field-programmable gate array) is programmed simply by placing the correct bits into RAM, so it's really a software-programmable device. But FPGAs are commonly used to prototype ASICs (application-specific integrated circuits), and the ASIC is then typically built by taking the exact same starting RTL (register transfer level) Verilog that was used to produce the FPGA, and taking it through logic synthesis, place and route, and making a mask and sending it to a fab. So, is an FPGA software and an ASIC hardware? If you create a legal advantage for software, I'm sure that this will make the stockholders of Xilinx and Altera happy. Similar arguments could be made in microbiology: DNA is basically a digital bit stream, and a lot of the DNA-processing patents are basically just software patents in disguise.
Bruce Perens expounds 'Open Source State of the Union' (NewsForge) Posted Jan 23, 2004 15:59 UTC (Fri) by ewan (subscriber, #5533) [Link] BBC News Online have a mini interview with Bruce covering this sort of ground.
|
|||||||||||