RSS Board >> ac_instructions

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
=== The Area Chair process ===

  * By January 23, I will have assigned papers to each of you. 
  * By January 30, please assign papers to reviewers. Each paper will be assigned to 2 ACs. You can see all the reviewers. Each of you should assign 2 reviewers to each paper. 
  * If you have reviewers who have not downloaded their papers, please contact them ASAP. The list of delinquents is here: 
  * Reviews due February 28th. 
  * Verify reviews are complete, and moderately consistent between March 1 and March 7.

=== Managing the review process ===

  * Author names: remember that as area chairs, you can see the author names but the reviewers can't. The reviewers may speculate to authorship -- we should try and discourage this. 
  * Reviewer names: as area chairs, you can see reviewer names, but the reviewers can't see each others names. And, reviewers can't easily see their own ID number. While it's better not to refer to reviewers, it may be the only way to identify a specific reviewer. One reviewer's comments may carry significant weight with another (e.g., "Reviewer A is so famous, she must be right.", or "I have so little respect for reviewer B, I just want to contradict him"), thus influencing the discussion.

=== Primary and Secondary area chairs ===
  * The primary area chair has the following responsibilities:
    * Chasing down the 2 reviewers they assigned to each paper.
    * Leading the discussion on each paper, attempting to achieve consensus across all 4 reviewers, or at least a reasonable decision. 
    * Presenting the paper at the area chair meeting.
    * Leading the discussion at the area chair meeting to a decision. 
    * Writing the review summary. 
  * The secondary area chair has the following responsibilities: 
    * Chasing down the 2 reviewers they assigned to each paper. 
    * Assisting the primary chair in the discussion, and attempting to achieve consensus across all 4 reviewers, or at least a reasonable decision. 

=== Prior to the author response period ===

  * Please ensure your reviewers have provided reasonable initial reviews. 
  * Look for issues such as inappropriately short reviews, inappropriate language or other things you would not want to see in a review of your own paper.  
  * Also, if you see wildly divergent reviews, please see if there's any way to get the reviews to come to some kind of middle ground. They don't have to agree, but the authors are likely to react to the disagreement, and not add anything useful.

=== Assigning papers to reviewers ===
  * One key issue: I am asking you to assign papers to the reviewers that you yourselves recruited. If you feel that you have a paper that really should be assigned to a reviewer that you did not recruit, **please send me email first**, and I will let the recruiting area chair know. Since everyone did a great job of recruiting reviewers, this should not be a problem. Also, if you want to recruit additional reviewers once you see the papers, this should be no problem. The list of who recruited whom is.
  * A second key issue: remember not to give review assignments to other Area Chairs. This will not work out well for you. 

  * I recommend you download your papers first. It may be easiest to work through the papers offline, assigning reviewers to papers, and then entering the information into CMT afterwards.

  * The easiest way to assign reviewers seems to be as a two-step process. First, give yourself some suggested reviewers using the "Manage Reviewer Candidate Suggestions" link. The reason for this is that you get to see the reviewer keywords easily. Note that if the other AC on this paper has already set up suggested reviewers, the list of suggested reviewers will already be populated **with his or her reviewers**. You need to add your own. 
  * Once you've given yourself a set of candidate reviewers for each paper, go back to the main screen (probably easiest to click on "Select your role: Meta-Reviewer" and click "Manage Assignments, Bids and Conflicts". 
  * Alongside each paper it will say on the far right, "Edit Bids / Edit Conflicts / Edit Assignments".
  * Clicking on "Edit Assignments" will bring up a list of your suggested reviewers. You finish the assignment by clicking the box along the right-hand side and then clicking "Save". **REMEMBER TO CLICK SAVE** otherwise you will lose your assignments. 
  * The reviewers will need to be notified that they have received their assignments and need to be let loose. CMT does not do this automatically. The best way is to mail them directly. (CMT provides a way for you to email your reviewers and fill in their papers, etc. But honestly, here be dragons.) 

=== Shepherding Conditional Accepts ===

  * Please go ahead and contact your authors. 
  * The final call on the paper will primarily be yours, so be sure you know what you want. Be explicit with the authors as to what needs to happen. If you aren't sure yet, please take a few minutes to figure it out. I think everyone is clear, but if the details have become fuzzy in your mind since the AC meeting, give me a call and we can sort it out. 
  * Please set a timetable with your authors. It would be a mistake to get the next revision of the paper on May 25th. You probably will need to negotiate the exact timetable with your authors, but I would recommend getting a new version of the paper in early- to mid-May at the latest. 
  * Expect to iterate.
  * If you think you need to, please feel free to get on the phone with the authors and talk them through the proposed changes. 
  * Make sure the authors know to include a summary of the changes at the front of their final revision. If you aren't happy with these changes, you should let me know.