Publications detector should report a summary instead of an empty flag

[Janet Davis]

Story: As a user, I want to see the following as a summary instead of a flag, so it's not confusing.

[0-0]: Publications: This document does not mention many publications. (Try referencing more concrete publications or work byproducts, if possible.)

I discussed this with @j6k4m8 on Slack and he agreed!

The web team has this on their Product Backlog.

[Janet Davis]

@molliem , we have started work on addressing this issue and we have a question for you.

If there are not enough publications, then it's pretty clear the summary text should be the same feedback and suggestion that previously appeared in the Issue.

But what if there are enough publications? Should the summary be blank, or should the summary contain something like one of the following?

1. The text appears to mention an adequate number of publications.
2. The text appears to mention [X] publications.
3. The text appears to mention at least [MIN] publications.
4. The text appears to mention [X] publications, which is greater than the minimum threshold of [MIN].

I wrote "appears to mention" because we can't be certain that the letter refers to the recommendee's publications, or that the mentions all refer to different publications. (There could be mentions of 3 different publications, or 3 mentions of the same publications, and the Detector wouldn't know the difference.)

[Mollie Marr]

I like the idea of returning what the text appears to mention (3.) since the minimum threshold may vary by recommendation type and I don't have a way of estimating what is expected at each level.

…

On Thu, Jun 25, 2020 at 1:26 PM Janet Davis ***@***.***> wrote: @molliem <https://github.com/molliem> , we have started work on addressing this issue and we have a question for you. If there are not enough publications, then it's pretty clear the summary text should be the same feedback and suggestion that previously appeared in the Issue. But what if there *are* enough publications? Should the summary be blank, or should the summary contain something like one of the following? 1. The text appears to mention an adequate number of publications. 2. The text appears to mention [X] publications. 3. The text appears to mention at least [MIN] publications. 4. The text appears to mention [X] publications, which is greater than the minimum threshold of [MIN]. I wrote "appears to mention" because we can't be certain that the letter refers to the recommendee's publications, or that the mentions all refer to different publications. (There could be mentions of 3 different publications, or 3 mentions of the same publications and the Detector wouldn't know the difference.) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#62 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGNPQY4PWDM3COBSLMXUZVLRYOXHVANCNFSM4ODBW3LA> .

-- *Mollie Marr* MD/PhD Candidate | GS3 Developmental Cognition And Neuroimaging Lab <https://www.ohsu.edu/school-of-medicine/developmental-cognition-and-neuroimaging-lab> Department of Behavioral Neuroscience Oregon Health & Science University Immediate Past President, National Student Division American Medical Women's Association (AMWA) [e-mail] | [e-mail]

[Janet Davis]

You got it!

[Janet Davis]

Resolved in #67; waiting for @j6k4m8 to merge