Everybody is listing concerns here because they are invested in the site and want to see it continue to do well. On meta, one of the things we talk about is the community dynamics as a whole. To say that the dynamics of a site like this are complicated would be an understatement.
We need to try and keep our existing users. While we can't keep everyone, we should try to maintain an environment where the community doesn't drive long-term users away. Part of this effort is to limit the questions that are frustrating to answer (aka "bad questions"). Questions that are incomplete, questions that change drastically, and questions with a wide scope generally fall into this category. Closing these questions is our primary method for dealing with these issues. This issue is very important to Olin, and is reflected in many of his answers regarding the treatment of questions and users who ask such questions. This is an important issue for long-term site health.
We also need to grow the user base for this site. If we aren't increasing the number of users, then the community will eventually stagnate and die out. Users also make content for the site (questions and answers) so the more users we have the more content will be available. However, new users often aren't familiar with the rules we place on questions, or what information is required to get a good answer, and may unintentionally ask a question that we would consider "bad". This, combined with a different model from a traditional forum, can result in a bad user experience which will result in new users who quit. This issue is very important to Russell, and it is reflected in his meta posts. This is an important issue for long-term site health.
Ultimately, both of these issues are important, but aggressively pursuing one goal will adversely affect the other. I can find anecdotes for both issues, but the plural of anecdotes is not data. It would be nice to actually have data on what is happening so that we can have a balanced approach and adjust our response to new users. Scott is calling for more data so that we as a community can understand what is happening overall and have some numbers to base community policy. To paraphrase, we have issues, but we aren't sure if they are serious issues or minor issues, and more data would help us assess that.