Automated/Consistent Guidance Based on DQA Assessment Votes

The Data Quality Assessment can be used to indicate whether something in particular is “lacking” within an observation in terms of the qualities needed for the observation to reach Research Grade.

Sometimes, identifiers adding a DQA vote enter a comment containing a previously (or quickly) drafted response that aims to guide the user’s future actions, so that that both users can have a more meaningful experience with iNaturalist. Sometimes those comments are kind, diplomatic, and helpful; and sometimes they are less than kind, diplomatic, and/or helpful.

It would be really nice if the providing of some of this very common feedback could be automated as much as possible on behalf of the identifiers doing it now, and presented to the observer by the site itself instead, with a consistently kind, diplomatic, and helpful message.

For example, when the number of “No” votes for “Evidence related to a single subject” exceeds the number of “Yes” votes within a single observation, a message could be presented to (at least) the observer somehow that says something like “This observation has been assessed by one or more iNaturalists as containing separate evidence of multiple subjects. Here’s some context about what that means. If you disagree, you can take steps ABC (add your own vote, add a comment with clarifying information, etc.). If you agree, you can take steps XYZ (split your observation, add a comment letting people know they can remove their vote, etc.).”

As for how exactly this message presents, I can imagine a number of options:

  • some iNat bot user automatically adds a comment when certain conditions are met (probably actually not a great idea logistically, but throwing it out there to get the juices flowing)
  • some banner or message appears somewhere on the observation (ideally somewhere prominent, but maybe easier to put in the DQA area somewhere) with canned text when the page is viewed by the observer (maybe even when viewed by others too)
  • turn the DQA attribute text into hyperlinks, or put hyperlinked, superscript question marks next to them, that point to help articles/videos that provide guidance about that particular attribute (or about all of the attributes)
  • (if no real in-observation automation) some page with more detailed information on how to think about and resolve the DQA assessment attributes that identifiers can simply paste instead of the varied responses they paste now. Right now the pertinent information is spread across multiple different places (old help page, one of the many individual pages on the new help “page”, tutorial videos, etc.

This would provide value to all users of iNaturalist. It’d be tricky probably to nail down the exact logistics, but boy would it save me a lot of time and burnout.

I approved this because it doesn’t seem to directly propose changes to Notifications (which we’re not accepting feature requests for currently), but proposes a variety of actions apart from notifications that could occur. So for anyone, responding - please don’t propose a direction that involves changes to notifications and focus on the many non-notification aspects of this proposal!

I like the general idea of this. An iNat bot might be a nice way to do onboarding for newer users when they first encountered action on the DQA in their observations (though I would want a way for users to turn that off…)

I think the easiest to implement action and that with the clearest benefit would be to consolidate DQA documentation and link to it as clearly as possible across the website and apps. We do get a fair amount of posts on the forum asking about DQA definitions, even though that info is out there. So I think considering ways to make that info more obvious/easily found would be good and help - like the hyperlinks proposed above. I think some of this is probably under consideration already with the new FreshDesk-based Help, so maybe it is in the pipeline.

8 Likes

Since I add the DQA vote, I add a Text Expander copypasta. I link to the new Help - which is - helpful, kind and diplomatic.

I’d also reccommend the use of “:)” or “:D” to anyone commenting such things. It makes the whole message seem instantly more friendly. :D

Jokes aside, I’d appreciate a feature like this. I frequently ID unknowns and there are a bunch of the observations containing pictures of multiple different organisms.
As I want to maximise my ID-output, I usually just write “Please upload photos of different organisms as separate observations” or something similar and move on. It’s not unfriendly, but I wouldn’t call it kind either, but anything else would just take up too much time.

1 Like

I think this is fine - everyone interprets comments differently, and many people speak languages other than English.

is polite, to the point, and translates easily. Job done!

2 Likes