[Cryptography] New attacks on discrete logs?

[Jerry Leichter]

On May 22, 2014, at 6:17 PM, Dave Horsfall <dave at horsfall.org> wrote:
>> Certain special curves were proposed, were patented, were investigated.  
>> Turns out those curves suck.
> 
> How can one patent a mathematical expression?  It's a fact, after all, or
> are you saying that the application itself was patented?
I know nothing about the details of this particular patent - not even that it exists - but it's certainly possible to patent the use of a particular mathematical object for a particular real-world purpose.  This is not significantly different from patenting the use of a particular molecule as a drug.  (Drugs tend to be protected in other ways as well; in particular, the steps used to synthesize them are usually patented.  But such a patent wouldn't stop someone else from finding a different way of making the same molecule.  The patent on the use of that molecule would, however, keep them from selling it as a drug.)
                                                        -- Jerry