Gordon Conferences get it right in any number of ways. In my own subfield they're pretty much the perfect conference, and repeated attendance provides a network that has been very helpful for me (my amenability to the drinking culture doesn't hurt). There's occasional some cliquishness due to the Harvard/MIT set sending Every Possible Student and Postdoc, but it's not so bad.
I'm on the much-hyped 70th birthday weekend. We ended up somewhere not hot, just like Mom wanted: Orlando.
I like the idea. At APS conferences I always burn out too quickly.
I don't understand the OP. I thought there was a generally accepted rule that attendance at more than one session per day was prohibited (except for PHD students or people on the market) at most conferences.
2: At least the family's all together? Or is it still setup with one family one weekend and another family later?