Prof. Scott :

Thank you again for the opportunity to revise this paper.  We have
spent time discussing the reviewers and AEs comments, and feel that we
have put together a version that is acceptable.  While we wish that we
had the time available to spend on making this a more ambitious paper,
it isn't clear that the cost would merit the gains.  Therefore, we
have focused on refining the current approach for the paper, and
respectfully decline to take the potentially more useful and ambitious
approaches.

Addressing the points raised by the AE:  

1. we have tried to clarify in the introduction that data analysis is
programming, which was muddled in the prior version.

2. the parts on remote execution have been combined into a single
section.

3. We have attempted to resolve and clarify issues raised with
documentation.  The AEs issue of documentation in general merits a
second paper, which is in progress.

4. We have substantially removed the detailed and pointless discussion
of major/minor modes, and have left the more general paragraph (with
some editing to hopefully aid in presentation).

5. (Source code control/Ediff)

6. the Section 3 intro has been rewritten and moved to section 2.  We
think that putting ESS first has presentation advantages, and may keep
the reader interested (whereas moving it later runs the risk of losing
attention).  So, Sections 2 and 3 (Emacs and ESS) have been switched. 

7. Again, we've tried to address the nature of statistical data
analysis coding as programming.

8. The section on interactive programming has been modified.

9. Changed as suggested by the AE.


Addressing the points raised by Reviewer #1:

We'd first like to thank reviewer #1 for extremely thoughtful reviews.
If there was a solid benefit to taking the more ambitious options, we
would definitely consider it.  That, unfortunately, isn't clear at
this stage in the process.

There are not any particular bullet points to address, more general
suggestions, but we feel that in both changing the paper after
re-reviewing, as well as in addressing the similar concerns from the
AE and Reviewer #2, that we have tried to satisfy the most pressing
concerns raised. 


Addressing the points raised by Reviewer #2:

1. the discussions on remote and multiple processes have been
unified.  Source code control and ediff issues have been clarified
with respect to usage.

2. We've tried to remove Emacs jargon when possible.  In particular,
the section on simplifying keymaps has been somewhat clarified. 


Again, we'd like to thank you for reviewing the article, and hope that
in its present form, constitutes a work that is publishable by JCGS.