*What is your personal view of the role of CSTheory: in particular, please post an answer to Gil Kalai's question here.*

I think there are many roles for CSTheory.  Some students might want to browse the site to decide if TCS is right for them.  Some researchers might use it as an occasional resource.  Others feel the site gives them a community they otherwise wouldn't have.  All of these seem to be valid roles for the site.

*How would you deal with a user who produced a steady stream of valuable answers, but tends to generate a large number of arguments/flags from comments?*

I would resolve the flags as appropriate and comment on their inappropriate posts explaining why they are against the policy/spirit of the site.  If that doesn't help, my next steps would very much depend on the reason for the large number of flags.  If their intentions seemed genuine, I would be patient. But if it were clear, for instance, that the user is just insulting people for fun or to troll, I would consider suspending them for a period of time.  

*How would you handle a situation where another mod closed/deleted/etc a question that you feel shouldn't have been?*

I would contact the mod privately and discuss with them.  I think such problems will not be frequent here.

*What is your personal view of the role that CSTheory should/will/does play in the greater TCS community?*

I think it could be (and already is) a great resource for the community.  It would be great if most TCS researchers joined this site and participated.  However, the site's role in the community may evolve organically, as these things often do.

*Practically, one of the most important roles of a moderator, is to decide which questions to close, if at all. There are several attitudes toward this problem. Do you think that referring specifically to the user who asks the question and his or her perceived capabilities (e.g., "lack of mathematical maturity", as expressed by some moderators) is a legitimate reason (or part of a reason) for closing a question? Or should you judge a question only by its own merits?*

Mostly by its own merits.  If you know that the person asking the question is, for example, a famous researcher, then it makes sense to re-evaluate the question with this in mind.  But a bad question is a bad question.

*In your opinion, what do moderators do?*

Moderators basically should try to keep the site running smoothly with as little intervention as possible.  This involves deleting content that doesn't belong, stopping arguments that get out of hand, guiding new users, etc.

*A diamond will be attached to everything you say and have said in the past, including questions, answers and comments. Everything you will do will be seen under a different light. How do you feel about that?*

That's fine with me.  (Though if I were running the stackexchange site, I might have chosen a different design decision.)

*In what way do you feel that being a moderator will make you more effective as opposed to simply reaching 10k or 20k rep?*

The software makes moderator votes immediately binding, so this lets them resolve egregious issues more quickly.  Of course moderators also have access even 20k users do not.  However, more importantly, because moderators are elected by the community, they can speak for the community in a way high rep users cannot.  This also means that moderators should be more careful in what they say and how they use their privileges.

*We often get low quality or obsolete or chatty flags. Based on early discussions the moderators normally don't take any action and dismiss them as not requiring moderator intervention. Do you think we should continue this way or change the policy?*

This seems to make sense as a policy.  But I don't have experience dealing with flags, so if Suresh or the other mods think the policy needs to be changes, I'd defer to them.

*This is one of my concerns about long term health of the site. Do you consider the primary goal of this Q&A site to be helping researchers in theoretical computer science and related fields to get help with their research?
If you do how should we deal with persistent trouble making users who try to abuse the site for their own wishes? E.g. users who are not researchers in theoretical computer science or any related field and insist on posting questions/answers/comments that are off-topic by the site's policies, e.g. cranks that want to use the site to verify or draw attention to their attempts on famous open problems?
How can we make sure the site will not be overridden by users who are not researchers so cstheory does not suffer the faith of other online open resources for researchers like Usenet groups? (Keep in mind that with the current amount of community participation in closing we are going to have very serious trouble if the number of such users with 3k+ reputation reaches 5).*

The bigger such problems become, the more active moderators have to be.  So far, we've been pretty good at closing crank questions.  I would strongly prefer questions be closed by the community, and I would try to cast close votes only as fifth votes.  However, if the site gets over-run by attention-seekers and it becomes evident that the community is ill-equipped to deal with them, then I (in consultation with the other mods) would become more active in closing and deleting content.  If certain people are persistently abusing the site, they can be suspended.  If the scope of this site is to change, it should be by community discussion, not by spammers.