Johan Löfberg

yes, but I've always tried to avoid adding overloading of doubles. Don't like that as it is such a universal class, and I bet that would lead to nameclash in...

Fix does not work with ``` sdpvar x y t = x^2+y^2 intvar u v z = u^2+v^2 Model = [t == 1, z

thanks for reporting. For the underlying modelling issue, please post a question on the google groups forum for suggestions

This one too (test_tobust_10) ``` yalmip('clear') w = sdpvar(2,1); sdpvar t W1 = [norm(w,1)

Essentially blocked by the fact that penlab requires hessians. Might be possible to create a hack which returns internally generated hessian approximations

Support implemented, but I think I post-pone making this default behaviour

Na, just syntactic sugar, no problems really

Absolutely, but unfortunately not possible using the framework currently implemented

Gridding feature in plot could be a future feature

At the moment a bit of a mess to accomplish, as far as I can recall. I'll keep this open though