Should we link to the PDF or to the abstract or source page for scholarly references? Does it matter which?

Question

We strongly discourage link-only answers for several reasons, one of which is that if/when the link breaks or rots the answer becomes useless and future readers are left with a non-answer.

In this Space SE answer (using because it's a good and recent example) there are three scholarly articles cited. The OP linked to PDFs, I modified the links to make them the source/abstract page rather than the PDF, and the OP changed them back to the PDF links again.

I'm asking this question here because there are more users familiar with linking to scholarly papers here than there, and where this is done in answers a lot more frequently.

Question: Should we link to the PDF or to the abstract or source page for scholarly references?

---

Original iinks:

1. http://adsabs.harvard.edu/pdf/1986CeMec..39..365C
2. http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1988CeMec..43..127J&defaultprint=YES&filetype=.pdf
3. https://iopscience.iop.org/article/10.1088/1742-6596/911/1/012018/pdf

My suggested (and rejected) alternatives:

1. https://ui.adsabs.harvard.edu/abs/1986CeMec..39..365C/abstract
2. http://adsabs.harvard.edu/full/1988CeMec..43..127J
3. https://iopscience.iop.org/article/10.1088/1742-6596/911/1/012018

Answer 1

I'm usually in favor of linking to the abstract rather than the pdf. On Stack Exchange, chances are good that unless someone is looking for a figure (which should arguably just be in the answer), they're unlikely to read the entire paper; therefore, sending them directly there seems like overkill. Plus, abstracts can provide additional useful information about the paper without effectively shoving an entire wall of text in your face - links to other versions, papers that cited the paper, other ways to view it, etc.

For what it's worth, this is considered good practice among the academics I work with for basically those reasons - there was recently a micro discussion about the same issue. Etiquette might differ among other groups - this is only my corner of a corner of the scientific community - but I was raised, so to speak, on linking to abstracts rather than pdfs.