p2pnet.net News:- Entertainment lawyer Jay Flemma doesn’t believe Altnet’s circular email campaign to p2p companies in a bid to get them to license the TrueNames ‘hash’ patent will work.
In fact, “As I understand the lay of the land in this case, I believe they are grossly over-reaching in attempting to turn the world of IP into the wild, wild west and effectively mug these companies by trying to make them pay for something for which they do not have the rights to defend or prosecute,” he told p2pnet.
Flemma, who specialises in music, film, tv and book law with particular emphasis on the confluence of the media with the Net, is consulting with companies who’ve received the Altnet patent letter.
“We’re having discussions as to whether or not Altnet really has a leg to stand on,” says Flemma, an expert in the legalities of Net distribution media.
But, “I think their attorney’s claims in the Washington Post that a jury found that their patent was valid is woefully inaccurate because it is not a jury question – or what’s called a question of fact, whether or not a patent is valid,” he says. “That is a question of law,”
“I can tell you this: juries do not rule on the question of whether or not a patent is valid.”
The EFF (Electronic Frontier Foundation) is taking an interest in events, but hasn’t yet decided whether or not to become actively involved, the EFF’s Jason Schultz says.
As p2pnet was the first to report on Monday, Altnet has fired off a round of identical letters to companies it believes use hashes (think ‘links’) for a digital file.
How2Share Technologies, a small Canadian company based in Victoria on Vancouver Island, British Columbia, and which markets PiXPO software for a picture sharing network, is one of the more recent victims.
“I don’t think the patent has any legs,” managing director Jim Wallace told p2pnet.
Without saying it in so many words, the Altnet letters imply that if firms it approaches don’t license the patent, they’ll be sued.
In the meanwhile, in case you’re wondering what all the fuss is about, “A Hash as unique identifier was the whole idea behind hashing algorithms,” says Exo in a p2pnet comment.
Read on >>>>>>>>>>>>>>>>>>>>>>>>
Earliest example I can recall was the Hash sorting technique. In essence IBM’s punch card sorting machines (pre WWII) are an early example of hash sorting. In the punch card machine the hash was simply a nibble from a specific character column that is used to steer the card to a specific bin. This hash is only perfect in the since that all cards having the same character in the selected column will be steered to the same bin. To sort on a wider field, you simply start with the least significant column of the sort field; run the cards, re-stack first bin on top of second bin, and so on, repeating for each column in the field.
In data communications the CRC is a hash guaranteed to be unique over a specific number of bits. A CRC-16 is unique for files up to 2^16 bits in length, CRC-32 for up to 2^32 bits. Different CRC algorithms use different bits to generate the feedback used to digest the data into a hash. (early 1970’s?) CRC’s were first used to detect data errors over serial data links. Early example: IBM mainframe to terminal equipment communications using SNA and SDLC protocols. TCP/IP protocol uses a CRC-32 to detect packet errors. (Mid 19080’s)
The field of cryptography, specifically public key cryptography, needed secure hashing algorithms. (Only secure in the sense that the estimated length of time to brute force content that will generate a specific hash value will take a very long time on the order of many CPU years. RSA patented various Message Digest functions (in the 1970′s, several RSA patents recently expired and are now public domain).
Most of us are familiar with the MD5 algorithm in which several P2P applications use to generate file Hash values. The whole Idea of using a Message Digest function in cryptography was to generate a hash on a plain text document (file) such that it could be used to detect if that document changes in any way. If any character in the document (file) is different then the hash will be different. These hashes are used when digitally signing a document, to verify that a document is the exact same (uniquely identified) document you viewed before signing. To prevent someone from changing the hash, it is encrypted using your private key so that others can verify using your public key. The results of processing the document using the MD function should exactly match the hash decoded using your public key.
Databases have used such hashing algorithms to generate unique keys for locating data in a database. Software has a good example in the C++ language standard template library in the implementation of the std map object. The map object stores a key, the key can be plain text, but is usually a hash to minimize the number of characters that need to be compared when dealing with long strings. The key is used in a binary search to locate the mapped data. This technique was introduced in the STL library in the mid 1980′s when C++ first appeared, but the general technique in software originates in Data structure text books well before that.
Earliest reference I can cite from the top of my head is in Donald Knuth’s “the art of computer programming” from the early 1970′s.
first to report – Altnet tries TrueNames on p2p ops, p2pnet, January 10, 2005