David Eather <eather@[EMAIL PROTECTED]
> writes:
>Ertugrul Soeylemez wrote:
>> Unruh <unruh-spam@[EMAIL PROTECTED]
> (07-04-14 17:02:29):
>>
>>>> Feel free to make fun of it, but you may ask yourself the question
>>>> why people spend so much time, money and power for factoring
>>>> numbers, which nobody uses? ;)
>>> Because people DO use it-- factoring has been im****tant in may ways
>>> for thousands of years. And if general techniques for factoring are
>>> found, one of the most popular cryptosystems in the world fails.
>>> So your comparison is, shall we say, flawed.
>>
>> You don't find new factoring algorithms. You use existing ones to
>> factor numbers, which nobody uses. You don't do this to know how long
>> it takes, because you can roughly estimate it beforehand.
That is a theorist talking. Experiments can often deliver surprizes. And
demonstrations are believed much more than estimates are ( Recall that
Rivest estimeated that the the factoring of a 128 digit number would take
longer than the age of the universe when RSA was described in SciAm.
)
>>
>> So the scientific value of doing this is pretty near to zero, besides
>> some knowledge about how well a cluster or grid performs, which you can
>> as well estimate, too.
Again, estimates vs experiments. And factoring has long been seen as an
im****tant problem. Cracking CryptoSMS has not. Who cares if you do?
People want some reward for their efforts.
>>
>> I do agree that attacking CryptoSMS is pretty pointless, but not
>> completely.
>>
>>
>> Regards,
>> Ertugrul Söylemez.
>>
>>
>No, completely useless. In places it is massive overkill, so
>inefficient that it precludes widespread use hence, "why bother?" In
>other places it is flawed in very well known ways that there is nothing
>to learn.
>If CryptoSMS is listening the book "Secrets and Lies" has some very good
>parts dealing with exactly the problem areas. Those parts made the
>author so depressed he stuck it in a draw for a year.
|
