FTC and JD Holding Hearings on IP
hondo77 writes "The U.S. Federal Trade Commission and the Justice Department are holding hearings on intellectual property laws over the next few weeks (the first one was Feb 6). They're looking at the balance between IP rights and the free market."
I'd tend to agree on abolishing many of the copyrights, IP, etc... on the internet, but the fact still remains that someone somewhere must be paid for something to be developed or innovated and that particular person/company will want people to know and maybe even pay for something that has taken them so much time to develop. Think about the music industry, if someone said, OK you can copy the music as much as you like then the recording industry would simply stop releasing music, then there would be nothing to copy! maybe im on completely the wrong track here, but the way id understand this particular article is that everything should be free, I just dont know if that could be, as i said, someone somewhere has to foot the bill to pay someones wages to develop whatever is being trade marked.
http://www.webhostingtalk.com
Make it idiot proof and someone will make a better idiot.
Stupid patent examiners are a problem.
There are certainly some ideas which are sufficiently new and non-obvious that they deserve patent protection. I think the Fast Fourier Transform would have been one of them. But right now there's a huge number of patents being issued for stuff which is neither new nor non-obvious... and that is where the problem lies.
Let's take an example... searching for patents which include the phrase "hash table" in their title reveals ten patents.
The first patent (Dec 2001) is on a hash table which uses key mod N as an index and stores key div N inside the hash bucket (instead of storing the complete key). Hello set-associative content addressable memory. Every major cpu manufacturer has prior art on this one.
I can't make any sense out of the second patent.
The third patent is on using a hash table inside a switch to speed up finding a MAC address/port combination. Obvious to anyone with a background in algorithms: If you want to find something quickly, stick it in a hash table.
The fourth patent is on using two hash tables, and placing records into the second if they encounter a collision in the first. Prior art: Any 1st year data structures & algorithms textbook.
I can't make any sense out of the fifth patent.
The sixth patent is on inserting data into a hash table by writing the data first and the key last, in order to maintain thread safeness. Obvious to anyone who has written multi-threaded code.
The seventh patent is on growing and shrinking a hash table when it gets too full (or empty). Prior art: Any 1st year data structures & algorithms textbook.
The eighth patent actually looks like something intelligent; the ninth patent seems to be a duplicate.
I can't make any sense out of the tenth patent.
Ok, so out of nine distinct patents, we have five which should clearly have never been granted based on prior art or obviousness; three which I can't understand; and one which looks to be worthy of patent protection.
Here's an idea: If the USPTO grants a patent, and someone later demonstrates prior art or obviousness, the person who invalidates the patent should get to claim all the fees paid by the patent filer. I have a feeling that if this happened, we'd see a very rapid deflation in the number of dumb patents on the books.
Tarsnap: Online backups for the truly paranoid
The Great Rogerborgio will use his mysterious powers of prediction to determine what will happen in this debate:
Flame away, but far better if you get over to WIPOUT and actually write it down where someone other than the /. regulars might read it.
If you were blocking sigs, you wouldn't have to read this.