User blog comment:Wythagoras/All my stuff/@comment-10429372-20130715145943/@comment-5529393-20130715172759

LittlePeng9:  Joel David Hamkins argues otherwise. You may be interested in the following talk:

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&cad=rja&ved=0CEsQFjAD&url=http%3A%2F%2Fjdh.hamkins.org%2Ffiles%2F2012%2F04%2FBristol-2012-Pointwise-Definability-Talk.pdf&ei=sS_kUeLTDe66yAHQs4GABw&usg=AFQjCNEMMqFcjg7u-WaoSJzWShmzm89SZw&sig2=Lg_ze82WZbi_wlaQ49aoCg&bvm=bv.48705608,d.aWc

In it he talks about pointwise definable models, where every element of the model is definable, so for example all real numbers and all ordinals are definable! While I cannot argue with his mathematics, my philosophy is different;  while pointwise definable models exist, they don't describe the "real" universe, and even if the "real universe" can be extended to a pointwise definable model, that model is not "real". I guess I am somewhat of a Platonist. :)