To ensure the survival of the site, it seems that the primary priority should be keeping the questions coming. One problem here is that many questions are poorly formulated, and our responses are sometimes more dismissive than dialectic. I've found your efforts to rephrase questions in clearer and more generally useful ways very helpful in this regard. While it entails modifying semantic content in sometimes major ways, I think you've been very tactful in attending to, preserving, and emphasizing the intended, often implicit questions. I haven't seen any objections to your edits from OPs yet (whereas I have seen some endorsements), and rollbacks are always an option, so I don't see any harm in the practice so far, and I advocate continuing it liberally.
I've also found many of your "sock puppet" questions with self-provided answers helpful. (By this, I mean "sock puppet questions" in the sense of the Winter Bash hat, not in the sense of using multiple accounts.) I'm wary of meddling with the proportion of questions that are asked and answered by one and the same, but at least in the beta phase, this seems like a potentially beneficial pursuit as well. Asking and answering our own questions (while maintaining fairly high standards for what's worthwhile) could be a good way to supplement the rate of content creation here.
A secondary priority seems to be seeing to it that most questions have at least one answer, and preferably two or more. I've revisited some less recent, difficult (i.e., vague and rather unscientific) questions today to post answers of the "as good as it's going to get" variety myself. For old questions with few or no answers and little hope of improvement by the OP (and little direction for editing by the rest of us), lowered standards for the conclusiveness of an answer may be acceptable.
I also advocate spending a little time with the unanswered questions list, and looking for highly upvoted questions with less than three answers to see if there's anything worth adding to those discussions, as any contributions to these kinds of questions is especially likely to be appreciated...except in as much as old OPers may have already moved on, and upvoted questions may be gaining popularity from especially complete answers. I should also admit that these suggestions are also "notes to self," as I haven't really spent any time carrying them out myself.
Otherwise, we seem to be on track, which is very encouraging! I'm a big fan of this site, and feel it's contributing a much-needed, optimally efficient format for global collaboration and communication between the experts and consumers of our sciences. Cheers to all my fellow participants!
I'm starting to get into the habit of referring to this answer when I feel like a question shouldn't be closed, because we need questions, but closed questions count toward questions per day...However, I've asked a related question about percent answered and answer ratio. At the time of this edit, the question is fresh, so I'd like to edit again when/if I receive an answer, but as I've argued in the question itself, I think the matter of closing questions is likely to remain relevant either way:
If closed questions don't count, closing questions that won't get answers should be good for these statistics.
If closed questions do count, closing a question that seems off-topic / broad / duplicative / etc. might prevent the rare answer from someone who doesn't mind so much, and thus drag down our statistics.