Again there was a question about what $dx$ is. I know that this question has been up several times and that there has been a bit of disagreement on how to best answer the question. The natural thing, of course, is to close this as a duplicate of one of the other such questions.
Fortunately though Robin Goodfellow wrote a blog post giving a nice answer. Therefore I was wondering if it would make sense if it was possible to close a question as a duplicate of a blog post.
This is not likely to be implemented, nor is it a desirable feature. First, recall that even posts on another SE site (e.g., MathOverflow) cannot be used as duplicate targets, even though they have the same Q&A format. If cross-site duplicates were implemented, they would be implemented between different Q&A sites first; but that is not happening.
Second, I think this is not a desirable feature. One of fundamental principles of SE is that everyone can answer a given (non-closed) question, and the best answers rise to the top; additionally, the question author is able to select one answer as most helpful. None of this is available on a blog. Robin's explanation of differential forms on manifolds may be great, but a calculus student confused by $\Delta x$ and $dx$ in "thin slice" integration problems might find it hard to follow, especially since they likely had no multivariable calculus yet. On a Q&A site, there could be other, more accessible answers. Not on a blog.
If the blog post was to be used, then I would prefer simply referring to the post as a source, and writing an answer specific to the question. The reason is fairly simple: the intended audience and purpose was very different from those of the OP.
Indeed, my blog post was mostly me venting after reading a string of "differential geometry" questions that should have just been filed under multivariate calculus. It was certainly not meant for students just starting calculus. For a new student, I would suggest introducing $\mathrm{d}x$ simply as a special kind of vector that is, in some sense, complementary to the derivative operator $\frac{d}{dx}$. I would probably avoid delving deeper into this definition with the student until (at least) they understand the concept of directional derivatives.
Presumably the blog posts are supposed to be more permanent than other external links, so are more valid to link as substantive answers. This depends on any significant change in the blog infrastructure being amended for all the links which depend on it.
