search appears to be skippig over common stop words about source HOT 4 CLOSED

The stopwords list is all managed on the backend, and I believe we're now pointing the dev server at a common elasticsearch instance that serves several Mozilla projects. (Hopefully Ross can confirm here?)

Not sure if there's a custom stopwords filter set up there, but by default elasticsearch just uses a standard English set. If we do want to change this, details and a couple relevant links here: http://stackoverflow.com/questions/4927629/can-i-customize-elastic-search-to-use-my-own-stop-word-list

I think current behavior is probably OK though? Anyone else want to weigh in?

from source.

I feel like skipping things like "in the" is just fine behavior. Certainly far better than the alternative (any instance of the word "the" pops a search result). Can't quite come up with a reason that I'd want any behavior that is counter to that.

from source.

Safe to close this one?

from source.

Those are reasonable justifications. It seems that some more operative words are not considered stop words, so it's unlikely to lead to searching missing out on relevant results. Things like "in the" are kind of borderline; they consist of stop words, but the combination is meaningful and far less common than the individual stop words. That said, keeping it this way it is now and having a look at the search logs at some point to see if people were actually searching using stop word phrases is less work, with the same net effect =)

from source.