Optimize the user interface for annotating

Feature Requests
1

This is a request to make an interface more efficient. It’s not a feature request per se but I don’t know where else to make this suggestion.

After manually annotating thousands of observations, I have a concrete suggestion that will make the bulk annotation process more usable. I use “Identify” via the web interface where each annotation requires two clicks plus a scroll to the arrow that advances the frame. My suggestion reduces the two clicks to one.

Every time I annotate an observation, I have to activate the appropriate select menu (one click) and then select the desired menu item (two clicks). This two-click procedure is not strictly necessary. Once selected, the menu items can remain exposed in the frame, eliminating the need to activate the select menu each time an annotation is made.

The select menu is the culprit. Simply eliminate the select menu. In its place, expose a sequence of checkboxes or radio buttons (depending on whether the choices are mutually exclusive).

It’s not necessary to expose the checkboxes (or radio buttons) of all annotations, all the time. Using submenus, the checkboxes (or radio buttons) of each annotation could be exposed or not, at the user’s discretion. When the frame is advanced, the annotation interface is preserved, which essentially reduces the number of clicks to one per annotation. That doesn’t sound like much, but over a large number of annotations, the savings in user effort would be substantial.

Thanks,

Tom

2

Hi Tom, it’s helpful if possible if you can use screenshots and examples in your requests.

Is there any reason you don’t use the keyboard shortcuts instead? For example, when I’m annotating, here’s a typical workflow:

  • filter to show observations that need annotations, open identify modal, and click the Annotations tab
  • look at photo of obs1 = flowering
  • pressp key, then l key
  • advance to next observation by pressing right arrow key
  • obs 2 = flowering and fruiting
  • press pand l, then p and r
  • right arrow next

Keyboard shortcuts make this a pretty quick task, and I can add annotations much faster than iNat can keep up with me.

To summarize based on the above: this request is to eliminate the requirement to press the p key at all, right?

2 Likes
3

Thanks for following up. Here’s a screen shot of the GUI I’m referring to:

iNat%20identify%20modal%20dialog
iNat identify modal dialog.jpg2386×1290 338 KB

To be honest, I haven’t used the keyboard shortcuts since I didn’t know they existed (until just now). I must have missed something along the way.

I’ll try the keyboard shortcuts but that’s a completely different interface. It may turn out to be more efficient than the GUI but that’s because the GUI is suboptimal. FWIW, this feature request is for the GUI, not the keyboard shortcuts.

I’ve figured out a way to reduce user actions to one click per observation, which is just about as efficient as you can get. In general terms, it boils down to the following modifications:

  1. Lose the select menu, which is inefficient.
  2. Implement a new feature that auto-advances the frame.

I don’t mean to imply that this is trivial but that’s it at a high level. Hope that makes sense.

Thanks,

Tom

2 Likes