Finally, I have a Science paper after 25 years of research in theoretical physics.(http://www.sciencemag.org/content/338/6114/1604.abstract )
(A pre-Science version: http://dao.mit.edu/~wen/pub/dDSPTsht.pdf )

(1) This may be my fourth or fifth trial. So my rate of success = 20-25%

(2) I wrote the first short version based on the long version (http://arxiv.org/abs/1106.4772 ) . It was rejected due to a negative review of one of the referees, that the paper is too hard to understand ("I don't understand at all what the result actually is." according to the referee ). Then we tried very hard to improve the paper to explain things as clear as possible in 4 pages, but that was rejected again. I felt that I could not do any better, so I gave up. But Xie and Zheng-Cheng did not.   Xie (and Zheng-Cheng) wrote a totally new version, which (I feel) is about as hard to understand as the previous version. But their version got accepted!   This makes me to realize that I have no common sense.

(3) We spend so much time for the Science paper, and I wonder if it is worth the effort.

(4) Science seems like the papers that are news worthy. So it is hard to publish truly new results in Science. This is because truly new results are usually about something that do not even have a name (so it is really new). It is hard for something that does not even have a name to be news worthy.  For our case,  the first SPT paper was published in PRB 84, 235141 (2011).

(5) I love PRB, which publish all my new results.

So what is lesson learned after my first Science paper: it is not worth it.