In cryptography, we've had eprint.iacr.org for a while now and if you want your paper to get noticed, an eprint version will do more for you than a talk at a se

In cryptography, we've had eprint.iacr.org for a while now and if you want your paper to get noticed, an eprint version will do more for you than a talk at a second-tier conference. Even at first-tier conferences like Crypto, people will sometimes have "eprint 20xx/yyy" on their last slide to encourage people to go and read the paper. Peer review still has direct value, and peer reviewed publications on your CV have indirect value so the IACR conferences (and possibly Springer/LNCS) aren't going away any time soon. But everyone I know of in the field looks on eprint first when they actually want to read papers.