Key Length Correlation

I've read Blaze and diffie's paper on Semetric Key Length.  I'm currious
how semetric key length correlates to public key cryptography length.  Does
public key length have to be much longer, (assuming the algorithms are strong),
to provide the same level of security as a shorter semetric key?  TIA   
-- 
According to John Perry Barlow:                       *What is EFF?* 
"Jeff Davis is a truly gifted trouble-maker." *email <info@eff.org>*
*** O U T L A W S  On The  E L E C T R O N I C  F R O N T I E R ****
US Out Of Cyberspace!!! Join EFF Today! *email <membership@eff.org>*

Jeff Davis said:
> I've read Blaze and diffie's paper on Semetric Key Length.  I'm currious
> how semetric key length correlates to public key cryptography length.  Does
> public key length have to be much longer, (assuming the algorithms are strong),
> to provide the same level of security as a shorter semetric key?  TIA   
> -- 

It's been calculated that in the case of PGP, to equal the work factor 
to brute force the 128 bit IDEA key would require a RSA key > 3000 bits.

Someone probably has a reference.

-- 
Kevin L. Prigge                     | "I rarely saw people sitting at
Systems Software Programmer         |  computers producing real code
Internet Enterprise - OIT           |  wearing ties." - Philippe Kahn
University of Minnesota             | (speech at Software Development '90)