"Crypto shocker": well, yes, it is shocking and it is about crypto.
"four of every 1,000": 27,000 out of 7.1 million from the researchers' current collection. (Most of the paper is based on an older version of their collection, for which the figure is more like 2.7 out of every 1,000.)
"public keys": this is the least accurate bit; it applies only to RSA keys, not to all public keys. That's pretty bad -- though, according to the paper, somewhat over half of the public keys in their sample are in fact RSA rather than ElGamal or DSA or anything else.
"provide no security": the paper says "... that offer no security". Perhaps "no security" is slightly overstated -- the keys are vulnerable only to attackers who can be bothered to do much the same work as these researchers did -- but they make it clear that doing that work is far from being rocket science, and that there are "people who know how to do the computation quickly".
What about this justifies saying that "even by the usual linkbait standards, this is egregious"?
