5
$\begingroup$

If we vote a question to migrate away to a different SE site, there is also a non-trivial and lesser known requirement (documented here):

The question needs to have a tag, which exists also on the target site.

If it doesn't, the migration will be automatically rejected.

This rule is one of the numerous obstacles with what the SE tries to minimize the number of the question migrations, on unclear reasons.

Currently we can vote to migrate only to the MathSE and to the Physics Meta (that we can select only from a short, pre-defined list of the target sites, and only mods can migrate a question to anywhere, is another mysterious obstacle).

There are 99 tags existing on both the MathSE and the PSE, these are (generated by this query):

Tag                        #Math #Physics
-----------------------------------------
algebraic-geometry         13316       27
algorithm                   6544       58
algorithms                  6544       58
analyticity                  664       41
asymptotics                 4335       37
bifurcation                  150        5
big-list                    1056       38
biology                      176      538
books                      11135     1178
brownian-motion             1776       88
calculus                   68438      248
category-theory             6188       37
celestial-mechanics           14      305
cellular-automata             58       36
chaos-theory                 262      163
chemistry                     90      473
classical-mechanics          837     3473
compactification              42      126
complex-numbers             8816      278
coordinate-systems          1622      592
covariance                   716      216
curvature                    728      537
data-analysis                536      190
definition                  7545      517
derivatives                14310      424
differential-equations     21151      267
differential-geometry      13936     1418
differentiation            14310      424
dimensional-analysis         138      390
dynamical-systems           2790      112
education                   2043      568
electromagnetism              62     7173
entropy                      575     1017
error-propagation            188      322
estimation                   797      337
faq                           73       27
field-theory                6397      793
fluid-dynamics               465     2903
foundations                  575       71
fourier-transform            909      538
fractals                     596       34
gauge-theory                  91      838
general-relativity           289     5079
geometric-topology           737      376
geometry                   23204      422
graph-theory               11449       28
group-theory               24648      680
image-processing             358       33
induction                   4999      502
integrable-systems            41       73
integration                35037      430
kinematics                   159     1751
laplace-transform           2014      540
lie-algebra                 2895      392
linear-algebra             59666      316
mathematical-physics        1902     1420
metric                      7230      884
mnemonic                      23        7
network                      142       23
noise                         96      111
notation                    6208      499
optimization                9070       40
parity                       121      161
pedagogy                    2043      570
percolation                   61       20
perturbation-theory          334      389
probability                47334      487
quantum-computation           99      413
quantum-field-theory         161     5627
quantum-information           38     1274
quantum-mechanics            578    12135
reference-request          11135     1178
reflection                   230      707
regularization               137      200
representation-theory       4333      422
signal-processing           1100      148
simulation                   317      571
singularity                  185      342
soft-question               7188     1091
special-functions           2700       17
special-relativity             7     4174
spherical-harmonics          152       83
spinor                        71      349
statistical-mechanics        132     2294
statistics                 17114      248
steady-state                  17       17
stochastic-processes        6932       67
string-theory                 40     1510
superalgebra                  19       39
symmetry                     566     1006
teaching                    2043      568
tensors                     1360      885
terminology                 4235     1015
tqft                           3      229
trace                        584       57
vector-fields                580      358
vectors                     4779      807
visualization                446       69
volume                      1379      146

If you vote a question to migrate to the MathSE, please check if it has at least one of these tags, and add one of them if it has not.

$\endgroup$
  • 1
    $\begingroup$ An important caveat: the tag added should be one that actually fits the question. $\endgroup$ – David Z May 19 '17 at 6:59
  • $\begingroup$ @DavidZ Yes. This is why I ordered the tags by their counts (more exactly, by the multiple of their counts on the two sites). A fitting tag can be found most likely in the top of the table. $\endgroup$ – peterh May 19 '17 at 7:08
  • $\begingroup$ There is an exception of sorts to this requirement. Every site has a list of so-called "intrinsic" tags. These tags don't really exist on the site because it is assumed that any question on the site is on that subject. A question from site A tagged with an intrinsic tag of site B can be migrated to site B, regardless of other tags, provided it satisfies the other criteria for migration. The main intrinsic tags on Mathematics are, unsurprisingly, (math) and (mathematics). Physics does have a mathematics tag, so any question here with that tag can be migrated to Mathematics. $\endgroup$ – user29205 May 19 '17 at 10:04
  • $\begingroup$ @arjafi Ok, thanks. What happens if a question tagged only with an intrinsic tag, in our case, having only the "mathematics" tag, is being migrated? There is no (visible) "mathematics" tag on the MathSE, how will this question be tagged? $\endgroup$ – peterh May 19 '17 at 10:31
  • 1
    $\begingroup$ @peterh In that case it will be tagged untagged. $\endgroup$ – user29205 May 19 '17 at 10:32
  • 2
    $\begingroup$ I don't understand this question. What is it you're actually proposing, and how would it impact the real world? Can you point to any question migrations that have been rejected for this reason? (hint: the three-month window in the 10k tools has precisely zero occurrences of this). Failing that, can you provide an example of a feasible set of tags that would get the question migrated but which do not exist on MSE? The tag requirement is there for a reason: to prevent migrations that would be off-topic on the target site. When automatic rejections happen, humans can easily step in. $\endgroup$ – Emilio Pisanty May 19 '17 at 13:46
  • $\begingroup$ Put another way, unless you can demonstrate real-world examples, I am unable to distinguish this question from histrionics that are unconnected with what actually goes on in that migration pathway. $\endgroup$ – Emilio Pisanty May 19 '17 at 13:48
  • 1
    $\begingroup$ @EmilioPisanty Yes, migration rejection happens on this reason occasionally. I've seen it multiple times. If you wish and it has a trace in the data SE, I would be happy to collect their whole list for you. | Typically, if I find a migration vote, I also always check the tags, to avoid this problem. Here I am proposing the community to do the same, to avoid that a question ends up closed because of incompatible cross-site tagging. $\endgroup$ – peterh May 19 '17 at 15:20
  • $\begingroup$ @EmilioPisanty About the rationality of the migration rejection: the optimal tagging should be decided by the target site, because it is obviously their area of responsibility. Furthermore, there is software support even for untagged questions. Thus, my opinion is that rejecting the migrations on this reason is an irrational obstacle to lower their number, but it is not an essential part of the post. The essence of the post is a notice to watch the tags, and the list of the tags what can be used to avoid the problem. $\endgroup$ – peterh May 19 '17 at 15:24
  • $\begingroup$ @EmilioPisanty Realworld examples demonstration is on the way. $\endgroup$ – peterh May 19 '17 at 15:27
  • $\begingroup$ @EmilioPisanty About the ongoing close vote: I could convert this notice/suggestion to a Q&A format, in this it would make very clear what does it ask. Would it be a good idea? $\endgroup$ – peterh May 19 '17 at 15:30
  • 2
    $\begingroup$ If you can show actual examples, then yes, but until there's proof of actual harm, I'm going to call this for what it is: pure hypotheticals. And even then, sorry, but I just don't see it. If a question is vote-migrated, it's because it's off-topic here, and an overwhelming fraction of that is people who couldn't be bothered to figure out that maths questions should go to the maths site. I don't particularly see why we as a community should go out of our way to accommodate people who can't even tag their off-topic questions correctly. $\endgroup$ – Emilio Pisanty May 19 '17 at 15:59
  • 2
    $\begingroup$ @EmilioPisanty As I've seen it happening multiple times with my own eyes, it may be "pure hypothetical" in your comments, but not in my eyes. Fortunately, there is a way to collect them [check my Meta SE post], and you will soon get a nice list. My reason to suggest this is that I want to make the migration system better, which can be a more important and more positive goal as to close salvageable questions on the PSE because they weren't tagged properly for a MathSE migration. $\endgroup$ – peterh May 19 '17 at 16:05
  • 1
    $\begingroup$ I am with Emilio in this one. The data needs to be presented in order for us to accept there is an issue worth raising. Until then, this is not particularly worthwhile. $\endgroup$ – Kyle Kanos May 20 '17 at 15:19

You must log in to answer this question.

Browse other questions tagged .