Discrete logarithms weaker than supposed?

Discrete logarithms weaker than supposed?  I can't access the original paper (and I'm not sure I could understand it even if I did!), but if this news report is to be believed, researchers have cast some doubt on the strength of cryptography based on certain variants of the discrete logarithm problem.  Assuming that they are correct, further assuming that this theoretical result could be reduced to practice, and even further assuming that the variant in question is used in common cryptographic building blocks – then this could be a big problem.  If so, I'd expect to hear much noise from the crypto community, pretty quickly...