Thank you, polarkernel, for all the work you invested into this site; I am sure that it will pay off in the long run.
I don't think that PO is dead; it just needs time since the level of the questions, and hence the effort needed to write good answers, means that answering is slower and fewer people spend their time on it. Indeed, contributing significantly to PO is very time-consuming, and I just had a break of 3 months because other work didn't allow me to free enough time to contribute.
On the other hand, I also think that PO needs better rules for handling rudeness. Professional scientists are used to an academic atmosphere that is polite, friendly and inviting, and few are willing to actively support an institution where these attributes are missing. But we need them to give answers to the many questions with currently 0 answers and to give the review section more weight.
I therefore strongly support Prathyush's view that the nonscientific part of all contributions should be edited whenever it can be improved in tone without affecting the scientific content. This has nothing to do with scientific censorship which is the suppression of the right to openly discuss scientific opinions.
Therefore I do not agree with Ron Maimon on the extent the user should be protected. I don't want contributors to be banned for rudeness, but I want them to be corrected, so that all content visible to the casual user has a polite, friendly and inviting appearance, and any rudeness that may have been imported by users should be hidden in the revision history for the few who want to search in it, and to be able to clear up user complaits about poor editing (which are hopefully rare). Moreover, editors or moderators responsible for an occasional inappropriate edit should be protected to the same extent as users who write contributions in an inappropriate tone.
Since Ron Maimon was very influential in establishing the no-rudeness rules of the site I'll take his posts as example. I very much dislike his tone in several of his contributions; for example this review, which is far below scientific standards as regards the tone of the review. In my opinion, this sort of rudeness chases away many potential contributors, and if the rudeness is not far outweighed by scientific content, it is detrimental to a scientific discussion and to the reputation of the site in the outside world. (I left StackExchange after his ban there not because I supported his rudeness but because I found the SE action inappropriate compared to what he had contributed there.)
It is necessary for a healthy site to suppress rudeness. The contributor's individual words are not sacred and should not be treated as sacred, only the scientific content they contribute is. I do not at all support the expression of any sort of rudeness - I am just willing to temporarily tolerate it if it is compensated for by the high scientific quality of the contributions. But the contribution should then be edited such that all traces of rudeness are purged out, hidden in the revision history. I would have done such editing in a number of cases myself if the rules had allowed it, in the same way as I routinely correct for mistakes in spelling or grammar. The quality of the site increases not only with the scientific content (which is of course central) but also with the quality of the form in which this content is presented, and for newcomers the latter plays an even larger role than for regular members.
The voting mechanism is not a suitable regulator for the quality of form. I never downvote because of the form, since downvoting should be for scientific content. (I also don't upvote a post just because its form is perfect, if the content is not also good.) Moreover voting on comments (where a lot of the rudeness happens) has no effect on user reputation, hence exerts no real influence at all.
Thus users who are allowed to edit should actively improve the outer form; any scientific content including the most severe criticism can be cast into a polite form at the level acceptable in any professional scientific paper or referee's report. The rules of PO should explicitly say that this is expected and will be achieved primarily through contributions by users who care, and if care is lacking, through editing by competent and trusted members. (Moderation means creating a moderate atmosphere, and this includes creating politeness.)
There is nothing ethically dubious about such a procedure. The contributors know what to expect; if they think they are corrected without sufficient reason, they can complain, and everyone can check the editor's actions, and other competent and trusted members can reedit poor edits if necessary.
On the other hand, the rules should say explicitly that all scientific content - i.e., facts or beliefs about scientific matters, including criticism, reformulated in a polite way if necessary (even unjustified beliefs or criticism) - will be preserved in the main text, while all matters of forms may be subject to correction, in which case the original form will still be available in the original form in the revision history. Thus all users will be able to thrive in a polite and healthy atmosphere, and contributors must not fear to be banned because of secondary reasons such as lack of politeness.
I hope such changes to the PO policy are feasible and will be made quickly, as this is necessary to attract all those who like to contribute only in an atmosphere comparable to that existing in the traditional academic institutions.