In general, I like shape of Academia.SE. And I am still being surprised, that with so many subjective questions, this site is a nice resource.
However, in my personal opinion:
- Too much of life-story/life-choices/coaching/etc
- often associated with personal-advice) - just please, no (I understand that newcomers may treat SE as a forum, but it is our job to help them shaping their issues into questions which work well in SE system)
In this site I have seen questions that sound like "what should I do with my career" with little or no "question-ness".(@Thanatos)
- Should a Ph.D. be done with a low h-indexed professor (typical, not particularly bad)
- https://academia.stackexchange.com/questions/17775/going-to-gap-a-year-before-reapplying-for-cs-phd-any-advice (a typical bad one)
- why it is a problem?
- hard to generalize,
- hard to answer (as there are many treads, which is the one most important to the OP?),
- hard to compare answers,
- not much re-use value,
- even hard to read the original question,
- it is harder to find answer (as someone can ask question with the same title, but different - background)
- Too much of soft answers
- I mean, if someone asks a question about chances of being admitted somewhere (or anything else) I think that we, as the community, should put more value on at least trying to use any data, objective references, links to other materials, etc. Sure, sometimes answer is "yes" (or "no"), which is obvious for any insider; but it many cases it isn't.
- Is it more difficult to score a Tenure Track position in the US when applying from outside? (answer is fine, but without any data or third-party insight it may be "calming, yet uninformative")
IMHO we should have much stricter comments and moderation for questions:
- capturing one's life story,
- asking a few questions at the same time,
- too long (IMHO they can be as long as one wishes, but the question, or the overview, should in in the first paragraphs).
And for answers (just comment-bugging may suffice), when:
- it seems that some data, research papers or essays can be linked,
- the answer seems to be specific for a given region or discipline.
(And less strict for the comments :).)