Handling assoc. editor's report

The first referee gives a negative judgement on the paper. The second 
referee is more enthusiastic, but points out serious mathematical flaws in 
the paper.

My own opinion on the importance of the topic of the paper is as follows:

Although the abstract says the paper extends the Q method from SDP to 
symmetric cones, really this means extending it to second-order cone 
programming, since the method is certainly not new in the LP case, where Q 
is simply the identity matrix. Unfortunately, the Q method was never 
competitive for SDP, and there are several methods that are very fast for 
second order cone programming, so it's not likely to be competitive.  That 
means the paper's contribution is likely only to be a theoretical one.

In view of these assessments I regrettably feel obliged to reject the paper.

The Handling Editor