@JohnRennie "No offense meant". None taken. Just that I felt a need to clarify, and didn't know about the change. The last time any suggested edit of mine was rejected, (well ...) was probably before these changes were introduced.
Of course, this wasn't the first time I had to get into a timeline reconstruction :P
The tag wiki for reference-request contains the following sentences:
Contrast this with the reference-request tag, which is for questions that ask for a specific resource, instead of any resource explaining a particular topic. The tags resource-recommendations and reference-request are
mut...
@Qmechanic Are the tag synonyms on references not bolted on backwards? I thought the arrangement was to map books into resource-recommendations and reference-request into specific-reference.
@EmilioPisanty : Ha. After today's edit it seems correct now, cf. today's meta post. The two current tag names were created Dec 9, 2013, cf. this meta post. Let me also ping @David Z, since he was involved in the res recom and spec ref wiki tags.
Ah, now I see what you mean. Well, after moving all ref req to spec ref, I created a NEW ref req tag. This is because new users tend to write ref req when they mean res. recom. This is confusing to people familiar with the past terminology, but it is intended to satisfy the needs of running the site in the future. The previous meaning of ref req is now obsolete and not in use.
So yes, it is by intention/design. We get 10 times as many res recom question than spec ref questions. So when one uses the ref req tag one gets mapped to the res recom tag in order to minimize need for retagging.
@EmilioPisanty yeah, what @Qmechanic said makes sense. In practice, 90% of the time when people type reference-request into the tag box, what they actually mean is resource-recommendations. (As evidence, we used to have to retag reference-request to books all the time)
If someone is using the reference-request tag correctly (according to its original intent), then yes, it should be converted to specific-reference, but hopefully people who want to do that will see specific-reference in the list that pops up and know to choose it instead. Those questions are fairly rare anyway.
@DavidZ That's fair enough. But then the tag wikis should be altered accordingly, and a correcting answer should be added to meta.physics.stackexchange.com/questions/5281/… to document the change.
I dunno, I'm the one who posted the question in the first place ;-) so I would encourage someone else to post the answer. It helps show community involvement or some such thing.
I know that there are a few different options to sort questions based on popularity, views, tags etc etc. These options give the same list to anyone who uses them at that moment.
But is there any option which learns from my past activities (views, votes cast, votes received etc etc) to curate a...
Has anybody noticed how this user has a tendency to write off-topic answers which end in recommendations of some book or paper (also linked in his profile)?
That user is trying to leverage the high up-vote to downvote rep ratio. Answering enough easy questions with actual physics for one or two votes each can generate more than enough rep to off-set the hammering from pushing some nonsense theory elsewhere.
If gets worse if misguided people upvote the nonsense because "downvotes are mean".
2
The evidence on Stack Overflow is that such compensatory voting is not common, but I don't have clear data of Physics SE.
You can look for statistical patterns. The study Shog9 used for stack overflow was to consider only those posts with exactly 2 votes and zero score (i.e. one up and one down). Then he asks were there more ordered up,down than down,up. The null hypothesis is that it should be random (i.e. even to within statistics).
There is a simple grammatical error in an answer that I was trying to fix. Unfortunately, it only involved modifying 3 characters. So, I quickly lightly copy-edited the remainder of the answer making 2 other minor changes.
The edit suggestion was than quickly rejected. Is there anyway to just...
@dmckee So, if I understand your procedure correctly, you test against the assumption that among questions with +1/-1, it should be equally likely that the +1 came first as that the -1 came first. And you find that on SO, for questions it is more common that the -1 was cast first, but for answers, that the +1 was cast first, with a significance that rules out randomness.
@dmckee Yeah, the statisticians will be calling for my head for that one.
It's a bit weird to assume counting statistics because "letting people post answers on SO for years" is not exactly a repeatable experiment that we would expect to be governed by Gaussian noise, yes