[Clascomment] OPT-IN: gamma n --> pi- p Differential Cross Section Measurements with CLAS

[David Ireland]

Hi Paul, et al., 

This is a very well written paper with nice looking results, and plenty of phenomenology. There is a lot of detail of the experiment, which is probably OK, but for future publications (CLAS12), one could see this level of detail being reported in the first paper from a run period, with subsequent ones referring back.

I have just a couple of comments:
- line 340: The analysis review will possibly have been through this, but there is no justification for using 4.5sigma. Why not 3sigma or 5sigma? It won't change the results, but this number seems arbitrary.

- line 680: How was the maximum J=10 chosen? It may be good to justify this on the basis of a plot of residuals.

- equation 11: This is likely established technology, but the introduction of the normalization constant X seems arbitrary. Chi^2 is not necessarily the best way to judge goodness of fit (as opposed to parameter estimation, which is also begin done here). Is the idea to show that the systematic uncertainty represented by X is smaller than some statistically expected number (\epsilon_X)?

- I particularly like the figures 2-14; clearly annotated, with a much better color palette than the ROOT standard - in fact I would advocate this as a collaboration standard! However figures 15-19 are not in the same style. Probably a minor point, and I realise that this will be because there are different people producing different plots for a paper of this length, but if there is any spare time to give all plots the same look and feel, that would improve the look of the paper. For instance, I personally don't like spaces between subplots, so e.g. figure 11 or 12 - very good, 17-19 could be changed. If this is too picky, I won't insist on changes!

Best wishes,

Dave