I asked a question on Computer Science.
It got migrated to Mathematica (after agreement from the Community).
The question on the new site got lots of attention, but was then closed by a moderator as off-topic.
I feel like I'm stuck between worlds.
My question is: A community voted to migrate my question and then it got closed - what do I do?
You are mistaken that your question was closed by a moderator. On the contrary, one of our mods J.M. encouraged you that the migration is a good idea. It would therefore be very unlikely that your question is closed by another moderator since we are a rather friendly community and if in doubt, we try to be open.
What happened is that users of the community voted to close your question, none of which is a moderator. For this, you need 5 users that vote to close. A moderators vote would always be fatal instantly. This is another reason why mods rarely close questions, because they want that the community decides.
Your question does indeed not really fit here, but I still find it interesting and I would have liked to see it open.
@halirutan has already answered the OP's actual question, but I think some discussion is warranted about whether this question fits this site.
Normally questions here are expected to stem from an actual problem you are having, and they should be objectively answerable.
This question is very much a borderline case. I did not vote to close it because there was already a reasonable answer when I saw the question and because I did notice when the CS moderators were asking whether migration would be okay. Without these two factors I may easily have voted to close as well.
On the surface, the question really doesn't belong. It is asking why the designer of Mathematica chose a specific design. It doesn't give a context and it doesn't make references to any specific applications of Mathematica (e.g. what are the relative merits of term-rewriting when applied to computer algebra?). Even if it did, there are computer algebra systems using very different paradigms (Maple, Maxima, Axiom), so the answer would easily contain many subjective arguments. (It is interesting to note that both Wolfram and Maplesoft market their choice of programming paradigm as an advantage over the other!)
On the surface, the question reads like: What was the designer of Mathematica thinking? I'm sure many interpreted it this way (see the comments under the question) and this is obviously not a question that can be answered by anyone else than the designer himself. Indeed, the answer links a talk by Stephen Wolfram where he talks about the creation of Mathematica.
I think the only way this question makes sense is if we interpret it as being about the history of Mathematica. If we view it from this angle, it suddenly makes sense. It is not a question which stems from a practical problem, but arguably it does fit the StackEchange format in general. There is a History StackExchange after all.
There are no precedents for such questions (that I am aware of). But perhaps we can have a discussion whether such questions are welcome. The history tag is so far about things like In, Out, $HistoryLength.
Personally I don't mind allowing history questions, as I don't expect that it will get out of hand. But I am not sure if most such questions would get good answers.
My reason for closure was a little more complex than the choices provide for. I do view it as a question that cannot really be answered by anything but opinion, but that is in large part because it was not accurately framed. I tried to get at this in a comment. (Due to the specific history of originating in a different SE forum and being closed in MSE, commentary is now locked, so nobody can give further explanation there. Oh well.)
Anyway, my claim about not being accurately framed starts with the notion that term rewriting is a form of syntactic substitution. This is definitely true in its design and implementation within Mathematica, and everything I have seen about the topic elsewhere also suggests it is a strictly syntactic beast. Mathematica certainly supports term rewriting but it is not the heart of the program. What Mathematica does do, at the heart of its computations, is an emulation of infinite evaluation. To respond to a question raised in comments by @AlbertRetey, this differs from term rewriting in that is is semantic in nature (in the sense of computer language semantics). That is to say, something much more powerful than expression parsing and substitution is involved in the process.
The question is undoubtably interesting. My reason for voting to close it was that I believe only Stephen Wolfram can give an answer that is not hearsay. The quote from a talk by Wolfram was not posted at the time I voted. If I had seen it I might have acted differently. Or perhaps not: the residual issue of whether or not discussing the history of Mathematica on this site still troubles me.
