Picking a second discipline on publications returns nothing

Why does picking two disciplines on publications return almost nothing?

On the publications search page, adding a second discipline to the filter empties the results. Not a smaller set, an empty one. Nothing is wrong with your search terms and nothing is wrong with your account. Filter by one discipline at a time on publications, then narrow with the search box or the other filters.

The identical looking control on journals does the opposite. Adding a second discipline there widens the feed, because a journal article matches if it carries any one of the disciplines you picked. Same dropdown, same multi select behaviour on screen, opposite maths underneath, and nothing on either page tells you which one you are using.

Publications asks for every discipline at once

The publications feed reads your discipline picks as a list of requirements rather than a list of alternatives. A publication has to satisfy all of them to appear. Journals reads the same picks as alternatives, so a journal article that satisfies one of them appears.

That difference on its own would already make publications the far narrower filter, because most publications carry a single discipline. What happens in practice is stronger than that.

A publication carrying both disciplines is excluded too

The requirement is applied in a way that no publication can meet. Instead of checking that a publication carries the first discipline and separately that it carries the second, the feed looks for one tag that is both disciplines at the same time, and no tag is two things at once.

So the rule to work with is this: two or more disciplines on publications returns an empty feed every time, whatever the pair, whatever is actually tagged. Publications that genuinely carry both of your disciplines are dropped along with everything else. Picking three behaves the same way.

This is a fault rather than a deliberate rule, and it is on our list to fix. Until then, treat the discipline dropdown on publications as though it allowed one choice.

What to do instead

  1. Pick one discipline and read those results. A single discipline behaves correctly and returns everything tagged with it.
  2. Narrow with the search box rather than a second discipline. Search terms combine with a discipline filter without emptying the feed, and a term is usually closer to what you were looking for anyway.
  3. Use the other filters without worrying. Publication type, license, author, typology and date range all match on any of the values you select, so adding options there broadens the results rather than wiping them.
  4. Compare two disciplines by running them one after the other. Filter to the first, note what you find, then swap it for the second.

If you have already selected two disciplines and are looking at an empty page, remove one of them. The feed fills again as soon as a single discipline is left, so you do not need to clear the whole filter set or start a new search.

Telling the two feeds apart

There is no marker on either page for this, so go by the feed you are on. On journals, each discipline you add can only bring more articles in. On publications, the second one takes everything away. If you use both feeds in a session, that reversal is the thing to keep in mind.

Related