Well, I had time now to read the linked article and comments by "Mad Scientist" and decided to make a own answer now, as the upvotes to my question can mean "agreement" or "this is a important question for yielding high-quality and attracting experts on this site and we should set up a guideline". I think it's rather the first (people agreeig with the points in the question), but no one really wants to make very critical comments concerning other user for obvious reasons. But this is no personal accusing of the author here, as he can direct to the blog post for this style.
quote from "Mad scientist"'s linked blog article:
The guidelines for private beta, when sites are empty and in the formative stages, are a bit different than the guidelines for a mature, launched site. In general I am referring to launched sites here, or sites that have at least been in public beta for a while.
Probably "a while" here means more some months, not few weeks, as you need that time to attract outsiders and newbies to this place. Also Atwood is referring mainly to solutions in answers of problems of programming, he generalizes this in the comment for all launched sites, arguable if this is a good generalization for all kind of questions and topics imho.
While I agree that answering with a solution is ok (but not very often of general interest and lowering the signal-noise-ratio of interesting content), that you found out yourself when facing a programming or math problem and you know the solution is correct, as it worked, I still don't think questions like the one on "synthetic life" have a single correct answer, Venters achievements were discussed strongly and put into perspective in the blogosphere. They are issue of different scientific views, what the term life means. In the current accepted answer, maybe this information and set of difficulties are covered in the given links, but I'm not going to read all those links, this info should be given in the text of the answer, otherwise comparing answers and voting for correct ones becomes impossible. If this is popular science level, what does from scratch mean? I think most laymen don't know, I'm not sure what actually the answer is, it looks like read all those links for exact definitions, then it should be community wiki imho, the answer mainly consists of links anyway. The purpose is probably to back up the answer, but I miss more the info in the answer itself. How do hundreds of papers to artificial life and abiogenesis really help me on this specific question? I have no problem if people disagree and vote this up, as they did, for good starting points or info, but the author accepting his own answer because of more upvotes (and here the temporal advance plays a huge role as explained above) is really problematic to me. There was no fair competition which would imho have generated even better answerS and more discussion in the comments. Compared to the discussion in the blogosphere the answer looks amazing simple here.
One of the commentators points out in the blog, often user answering their own question give very little context, what the motivation of the question is or why it is interesting. That's actually something I like the most on SE, questions give more context and are often closed for being unspecific, while on Quora and Reddit you often read very poor questions more asking a expert to make the work for you.
So from these points, one can deduce what may be a better kind of self-authoring and what type of questions are suited for it. Well, obviously Atwood likes users to seed his network with questions on their own (I would too ;) ), will raise the number of google-hits. But more important to me is, how will this affect quality of a site or community dynamics at early stages of a beta, that's probably why he sees here a exception to his rule. Is it productive, attracting experts? Quality of accepted answers comes from competition between diff. answers, and with this big temporal and voting advance and many interested voters on the current site (often voting for effort people put into Q&A), this competition pretty much vanishes. I can't argument vs. links given in the already accepted answer nor will this question attract much of the former voters anymore, I read max. 2-3 answers of a question. So why waste my time? I kills competition, for tricky scientific questions, this is the worst case, seems Atwood is not aware of this or amount of content is more important to him. But it's anyway more the question what the community of commiters wants to achieve here, they invest their time here. If you want popular science level, fine, I will rather read blog, news articles for such content or a text book and more frequent quora. But then really try to create competition, so the answers of popular science questions will be correct or not arguable as lacking detail, deeper insights and exact definitions. Otherwise I fear your top voted Q&A and user will rather scare away people than make them joining this place. That's the main problem to me, it kills a lot competition, for math/prog. questions this is no problem, as the answer is often a result being true or false, while many questions on physics, biology, chemistry really need discussion and agreement in definitions for a correct explanation. Look up some of the discussion pages on wikipedia, if you don't think so. My aim would be insightful detailed posts with links, as you find them sometimes on Quora with a big info/links ratio in the answer itself and well explained questions that really help you understand why the answer is good and important. This seems unlikely for self-authored Q&A. What I miss on this question is discussion and explanations what synthetics life is and the problems of producing it are. And it imho mainly developed to a list of links to lectures and videos because of this self-authoring non-competitve way of asking it. I would make the answer CW or even close the question for being to unspecific or not really explaining the problem/misunderstanding. Most of current links are very googable.
But this is of course a matter of finding a common denominator in a community, I'm not versus popular science at all, but we should really try reaching a level higher than google link collections or newspaper articles for the top questions to keep this site alive. Every average student (should) know how to find keywords and use search operators.
(Sorry this got so long, but I rather do write complete analysis on meta than breaking it up in dozens of comments where everybody loses overview and the depth of discussions is limited, a discussion format would be better here. Alot of meta and area51 Q&A seems rather redundant and pointless because of this)