Attracting new editors
I think the meta point to take from this is that we should acknowledge the problem and experiment with ways to solve them. Wikia has done some great work with this. Here's my report from an informal conversation with Danny Horn on that work.
The fact that new users are less likely to edit short articles is a very interesting finding. It's also a bad sign: longer articles are more likely to have settled into a "status quo". Shorter articles NEED those extra sections to be written. (for video game articles, the typical article needed a section about the plot, a section about the gameplay, a section about reception, and a section about development.) But apparently editors are less likely to start a section than they are to expand one that's already there. The fact that editors are more likely to edit a section than click the top "edit" tab only confirms this.
Perhaps part of the WYSIWYG improvements could be some way to show what sections are missing? Or maybe we can drive newbies towards editing stubs, along with some suggested "sections" they can add, and how to make a new section.
How about this: With a probability of approximately 1/1000 (or let it be 1/10,000), when an anonymous user wants to read a small article that is not a disambiguation page (important!), the edit link is underlayed in red, and a flashy message says something like: "Did you know that you can change the content of (almost) any article by clicking the edit link above? If you can add anything to this particular article, why not try it out right now?"
Of course if we activate readers in this way there is a huge chance that they will add unsourced information and will be bitten by newbie-headhunters.
Either way, this problem is too important for us not to try it out. It's a overall idea. I somehow suspect that those stubs will be less protected, and that almost any edit to a stub will be perceived as a good one. So it's a win in the "good experience" column for new users, and it's also a win for content.
I like this idea immensely, but second the concern as well.
If we're trying to get 1st time newcomers the main thing is to get them to make edits that are likely to stick and not be reverted or deleted or get warnings. For example, original research tends to be removed but accurate information with poor balance tends to be fixed and improved upon by others.
What's missing is interfacing that catches newcomers and guides them to a quick (user friendly) summary of what we look for in an article - "encyclopedicity" (a kind of topic that should exist), notability (evidence the world takes notice), at least one source, and no original research or promotion, should stand a good chance of covering it.
Also biographical articles have especially high standards and start with something else if not sure. We may want to exclude BLP stubs and stubs in certain controversial areas where it's hard to edit well without experience.
Thats the direction, I would go. Something like: "Articles of this kind typically have the following sections (aaa,bbb,ccc) which are missing here still. You can create one automatically by clicking on the titles." as a drop down option list right beside the edit button maybe.
So the idea is: having more different editing options. Even something as simple as an automatic "add an image" where you do not have to place it, but is sorted in a standard gallery beneath the articels' text and is then sorted out later by other editors. This way adding images or creating sections in articels could be made more easy.
The other idea about advertisement for the edit button. I am not sure, if people really do not know that they can edit every page in Wikipedia. But its worth a try.