To begin, a related question exists to evaluate claims' authority, any input here would very likely inform any decisions over there.
As to 'should answers need sources?', the answer is a mixed one. Gardening is a blend of prior knowledge mixed with the experience gained through improvising. In some cases, the personal experimentation is what will give the edge to a great answer. In other cases, it can result in unfortunate consequences. Here I would draw an analogy with cooking a fantastic tasting meal, but inadvertantly downing everyone at the table with food poisoning.
There are three ways to break this up:
- Questions requiring expertise and experience, but that are unlikely to carry unintended consequences
- Questions requiring expertise and experience, but whose unforeseen/unlikely unintended consequences are not incredibly substantial
- Questions requiring expertise and experience, and are likely to carry unintended, serious consequences
In the first case, I would cite my response to how to keep critters away; put up a statue of an owl. This has worked for me, maybe it won't for you or the pests you're dealing with, but at the end of the day no plants or persons will be any more or less harmed as a result of following the advice. Anecdotal answers are fine, backing them up is a plus.
In the second case, I would cite my response to how to keep critters off your irrigation; spray hot pepper oil on the equipment. In most cases I cannot see how this would carry consequences, but perhaps the oil eats through some cheap plastic tubing. No plants or people are really that harmed, no expensive equipment was really damaged. Anecdotal answers work, but some fact-checking and thinking about the possible unintended consequences (and probably voicing them in the answer) is a bit requisite for a thorough answer.
In the last case where the consequences might be severe, I would cite the question about making compost tea while avoiding e. coli. While I am glad it has one source. As a question-asker, I personally would not trust that source or consider the answer to be authoritative (particularly, because it cites only one source to assuage the concerns of someone worried about food poisoning). It has nothing to do with their credentials, it is because these questions voice a concern about bacteria and other pathogens. Similarly, while losing a random basil sprout isn't a big deal, losing an incredibly rare flower or heirloom variety of tomato (perhaps imported by your great grandfather from Italy) would certainly change the weight of any given answer. [Here I would refer to the list offered through Aaronut] on Cooking.se3.
- While I don't believe it is necessary to make every answer conform to the authority criteria, I think it is typically in everyone's best interest to first evaluate the type of question (and the possible consequences of implementing a given answer).
- If the question falls into that third category, and I asked it, I am unlikely to accept any answer that doesn't at least try to show it's work.
- If you have been making compost tea the same way for 30 years and have never contracted a food-bourne illness, that's fantastic; but please be very thorough in explaining what you do, and reference other guides to really explain it so that I don't try to follow your method and screw up.
The only problem is trying to diagnose where exactly the question falls. I think another question should be asked to address the scope of the site in this regard. For instance, although food safety guidelines are an important aspect of Cooking.se, asking about nutrition is considered out of scope. I am not sure this exact concept comports, but a similar consideration might be fruitful.