Glen_b's answers
[I copied and started working on the questions before Alecos' and Nick's edits above. I'm not going back to change it now. It shouldn't make any substantive difference.]
Before starting I apologize for the length of this post -- there were many questions, some quite searching. I suggest you look for the questions you're most interested in rather than try to read the whole thing (in any case I come back to similar themes in several of my answers).
What other sites on SE system are you active on, if any? Compared to these sites, how would you characterize the atmosphere and style on CV, if it is any different? How are you going to handle questions and topics that are borderline between CV and these complementary areas of expertise?
I've been asking and answering questions on StackOverflow since before CrossValidated existed - just under 5 years; I've been active enough there to have four Yearling badges on that site, with (at the time I write this) 82 posts in all. Some of my answers have been quite well received - my highest rated answer there is currently 3 votes shy of a 'Good Answer' badge. I've asked three questions. All but a few of the 82 posts I've made there are related to R (the others being of a statistical nature related to things like C and SPSS for example).
On math.stackexchange I've been answering questions there about 2.5 years - long enough to have a couple of "Yearling" badges (and 33 answers). About 80% of my answers relate to statistics or probability, and the remainder are just other mathematics questions of various kinds. My highest ranked answer is a series of leading questions for a homework question involving linear algebra and calculus.
I think both of those sites have a greater fraction of short-answer "routine bookwork" questions which tend to produce a "fastest gun in the west" mentality (and the greater volume in both places tends to encourage that approach further). Answerers in those places often tend to just answer such questions plainly. CrossValidated has a carefully cultivated approach of giving thoughtful guidance on such questions. This involves considerably more effort on the part of answerers, moderators and other participants -- including askers. While the effort is higher, I think the outcome is generally better, since it helps people come to understand the questions and the answers. Many askers are able to arrive at their own solutions, which is far more valuable. This approach also demystifies statistics, helping to reveal the underlying logic.
More generally, StackOverflow's questions are (necessarily) focused to a larger extent on the mechanics of code, and math.stackexchange tends to largely concentrate on very tightly focused, specific mathematical problems (I think variation on "how do I do this question?" would characterize most of the posts there), and 'fastest gun in the west' raises its head again. CrossValidated seems to have a larger fraction of relatively subtle questions about how to get from some vaguely expressed question of interest to a question that can be answered with data and issues with the resulting analysis -- and in which different experts can have quite different, often illuminating takes not only on how to approach it, but even on what might make a good answer. The questions often involve a lot of effort at clarification before the underlying issue is revealed – askers of statistics questions understandably often don't even know what it is they want to know! A question ostensibly about t-tests eventually turns out to need an analysis involving mixed-effects logistic regression, for example. Perhaps unsurprisingly, comment threads are often very long, and the resulting answers tend to be a lot longer on average than the other sites.
I think one difference is a relatively more collegial atmosphere here. [This is not to suggest collegiality is absent from the other two, nor in any way to imply that they're in any way 'doing it wrong' -- the differences in style (such as they are) are to a large extent a product of the kinds of questions that are asked, and also the much larger volume of questions.]
I'd handle questions at the borderline with these groups exactly as I have so far. If a question is posted to any site that is on topic for both that site and a second site, there's no problem to solve: it should stay where it was posted. If it's off topic here but on there, it should move. If I am in doubt, I lean toward being conservative - give the original poster the benefit of the doubt and assume they chose to post it where they did for a reason. There’s also the opinions of other voters when questions are nominated for closure as on topic elsewhere, so if I’m unsure, there will be other opinions - usually quite soon. I've dealt with many questions that have moved, often both answering a question and voting to close it as being on topic elsewhere. Sometimes where a post is probably on topic but could do better elsewhere, I instead encourage the original poster to consider flagging their own post to ask a moderator to move it.
Do you have a vision for the future of the site that is different from how it is now? In other words, do you hope to move the site in a new direction in any respect? If so, what, and how would that be an improvement?
I think any major change should not come from me, but from the community more generally - via open discussions on our meta site. That's how site-specific policies are discussed, informally voted on and occasionally modified, though usually we're talking about a minor change in emphasis rather than anything big.
Bigger changes, especially any that require programming, tend to be discussed on meta.SE and again, that's for the broader community to discuss.
I think the site does what it does very well, so I wouldn't want to change it too much. I think the challenge will be trying to maintain its best aspects in the face of rapid growth, and that's a community effort.
I would like to see a minor change in emphasis on a couple of points:
occasionally I think some posts are put on hold more quickly than necessary; where feasible I'd like to encourage change-before-closure.
I think many new users need guidance - someone to explain how to find out what the conventions and expectations are. I try to do that when I can, and I'd like to encourage more of that.
However, as the site keeps growing those kinds of encouragement will need to come from ordinary users of the site -- moderators will tend to have their hands full, and will likely need to lean more on the tools provided rather than extended discussion. We have lots of new users every day, and at best a handful of moderators can only discuss things with a very small percentage of them.
So the changes I'd like to see are ones that come from more participation by ordinary users. I'll try to facilitate that, and where possible to lead by example.
The multiple objectives of any SE site create an inherent tension, forcing us to make decisions that trade off between accommodating askers, creating canonical resources, establishing discoverability of content, and rewarding community members for their participation. With many questions (and answers), there is no one best action to take. As a general moderation strategy, how do you believe we should act? Where do you plan to strike a balance among these objectives?
As with several parts of the questions here, this is not entirely a theoretical question for me -- my response largely represent what I already try to do*, and what I will aim to do as a moderator where I can.
*(except that some of the things I presently can only vote for, or raise in some other capacity – comments to the poster, flags, and so on, I could actually do as a moderator)
The various choices of action should be based on the particular characteristics of the situation - the specific issues with the post, and in some cases the history of similar posts by the same user. I don’t have a fixed and unchanging set of rules -- the number of variables is quite large, but I will explain what I can for some situations.
In many cases, where time permits and signs are encouraging, I feel we should try to guide new users - for many users whose question is not so far from suitable, a few words of suggestion (or a quick edit) are often sufficient to convey the right idea. In those cases, putting a question on hold can be held in reserve, for cases where some guidance proves inadequate.
In some other cases, the question is relatively more poor, and for those questions it makes more sense to put the question on hold until the question improves. In some cases we can help with that, but (for example) where the user hasn't learned from guidance, I think it's sometimes necessary to ask that they fix (or help to fix) the post themselves.
The choice of approach is partly subjective, but often the indicators of problem posts become clear with a little experience helping new users with their questions.
As a frequent participant in review queues, we see posts where such choices can be difficult sometimes. One thing I frequently do after voting on such choices is to look back and see how other people voted, which helps me see how I stand in relation to other voters -- particularly the more experienced ones. This helps to give a more consistent style across questions.
Encouraging good use of tags and titles is something I think can often be done by example (if you add tags to a post it helps users understand more about how tags work), but sometimes it makes more sense to encourage the original poster to fix it themselves.
Regular removal of duplicates and occasional merging of posts is one important part of encouraging canonical answers.
Detecting bad posts, on the other hand, is something that moderators and many high reputation users do all the time -- but almost anyone can help: flagging a post with serious problems is something almost any user can do.
In your opinion, is there a serious threat/challenge that Cross Validated currently faces? If yes what is it? Do you believe that as a moderator you could contribute more towards dealing with it, and if yes, how?
I see a number of possible challenges. One is that with rapid growth we've seen a drop in the proportion of questions getting answered. Both questions and answers per day are growing, but questions grow more quickly than answers.
A second challenge is that as the sheer volume of questions grows, it becomes harder for ten or twenty active people to curate them all - to fix mistakes in questions or answers, to identify duplicates and off-topic questions, to ask for clarification of issues, and so on.
To some extent yes, though mostly indirectly. We're beyond the point where one or two more moderators will affect those particular challenges much -- but fortunately these things don't need moderator privilege to deal with. Let's consider a few actions and the reputation you need:
Essentially anyone who is registered can offer an answer or suggest an edit and it only takes 15 reputation to vote up or flag a post. Those four things are the lion's share of what we need more of. We have thousands of users that can do this -- and this is where I think we can best encourage more participation - answers, suggested edits, votes and flags. Editing to improve a post is part of curating the site. If you would like a healthy site, find one question you can answer and start typing, or one post that needs some work to be a clear, useful post and click that "edit" link at the bottom. Even if you can't provide a full answer to a question, you can encourage someone who does, by upvoting good answers. Flag posts that need moderator attention. Moderators can't be everywhere - tell them what needs fixing.
At 50 reputation you can comment everywhere. Is a question ambiguous or unclear? Ask for clarification.
At 500 reputation you can start accessing review queues. If you review one late post or one new user's first post, you're actually helping to moderate the site. It's not just the people with the diamond who can moderate. We have over 600 users who can review posts. (My own first reviews were in September 2012; my reputation was below 1000 at the time.)
Everyone should take a look at what they can do. You might be surprised how may ways you have to make the site better.
So even if I were a moderator, I see my ability to deal with these challenges as mostly trying to encourage ordinary users to help out more -- those challenges are really challenges for all of us who want the site to continue to be successful and useful.
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?
There's a series of steps that can be taken.
The first is simply to discuss the issue with the user (there's the possibility of private chat for example), and conversation is always the first avenue. Most people will respond, if approached in a reasonable way.
Occasionally stronger action is required, and moderators have several options if stronger action becomes necessary. These have been discussed on our meta site, for example, and can even include things like suspension.
Someone who generates a large number of arguments and flags is disruptive of the wider community, and harmful to the long term health of the site; giving lots of good answers doesn't change that assessment. It may be a tragedy to lose an active participant, even temporarily, but the alternative - an increasingly dysfunctional site that leads to the loss of even more-valuable members - would be worse.
How would you handle a situation where another mod closed/deleted/etc a question that you feel shouldn't have been?
For me this is not a theoretical question, because it's something I've already dealt with in my capacity as an involved ordinary user.
So I'd do pretty much what I've done already when that's happened - simply raise the issue and discuss it.
For example, there are several meta posts where I did exactly that (please, go ahead and review my meta questions to see this), and one or two minor queries raised on chat or in comments. These issues are nearly always resolved satisfactorily one way or another. However, moderators have the ability to have that discussion in private where necessary -- my preference would be to discuss in public whenever that's reasonable, as now, because I think moderation should try to be as open as possible, but I have no problem doing it privately, especially if there's any question of sensitivity or privacy.
In your opinion, what do moderators do?
In the shallow sense, what moderators is not a matter of opinion: the privileges and duties are laid out; I won't labor the point by reiterating them.
In the broader sense, moderators keep the site effective - on topic, clear of spam, reasonably polite, and so on. There are a number of effectively 'janitorial' administrative tasks that the site can't take care of on its own, and now and then some "guidance of people" is needed that occasionally requires more than ordinary peer pressure, again with the aim of keeping the site effective.
StackExchange sites are intended to be long term repositories of good questions with good answers; a lot of what moderators do is, in one way or another, ultimately based around keeping it that way.
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?
As a general principle I think this isn't a good thing, since it retrospectively alters the light in which those things will be seen (in fact, I see this as a design flaw).
Even as a moderator, I think we should be able to 'take off the badge' when not acting in the capacity of a moderator (presumably this would necessitate something like instituting a second 'moderator' account for moderators, attached to but separate from their normal one). When I answer a question, or ask one, or comment on a statistical problem with an answer (rather than things like formatting, layout or style), I'm just being an ordinary user and my words shouldn't have any 'official' weight attached to them.
However, in terms of my own use of the site it's not that great a concern for me. There's a couple of my less polite comments in the past that I'd rather not see the diamond on, for example and with things like those, when I encounter them I may well delete some (but this is not unusual, I delete some of my old comments when I see them already).
In what way do you feel that being a moderator will make you more effective as opposed to simply reaching 10k or 20k rep?
This depends on the particular capacities in which we're judging effectiveness.
My stated aim in my nomination was to help the moderators more than I presently can. While I have more than 20K reputation, and access to moderator tools (and use them almost every day), moderators can still do numerous things that I cannot. Some jobs require a moderator.
Certainly in that sense, access to those additional privileges that come with the diamond will give me greater scope to give aid where it's needed. In particular, I'd be able to help out with things like moving posts to other SE sites, removing spam and various of the other moderator tasks that I can't presently do.
Some tools that moderators have access to may also make me a little faster at things I can presently do. Since time is finite, faster means doing more.
In other senses, I will actually be less effective. I presently answer a lot of questions on the site. I expect that with the additional duties I'll be taking on, I'll be answering a lot fewer questions.
