Agree button on Identify thumbnails

I find this idea somewhat attractive, and would be curious to know from other identifiers how much this would cramp their style. I’m thinking for anything with either (1) a comment, or (2) multiple photographs, the thumbnail agree should be disabled. But maybe there are overriding good reasons for keeping it there, despite the potential for sloppy IDs.

3 Likes

Several people (including me) suggested getting rid of the “agree” buttons on thumbnails back in the google group, but there are high-volume identifiers who use them for straightforward groups (or one, at least!)

One of the things that came out of that discussion was that Scott ran a blind ID study to look at where the most errors were creeping in. Are most “agree” errors coming from new users unfamiliar with the site, or high-volume users who make an occasional error? If that question was resolved, I missed it.

Apologies if this discussion is starting to stray from the feature request. Perhaps we need a separate discussion of the “agree” button.

3 Likes

Good point. I moved this to its own topic.

2 Likes

As a somewhat high-volume identifier, I rarely use the thumbnail agreement. The only time I do is when it’s a string of observations from the same user, and I probably shouldn’t even then. I’ve found that I can find possible location or date errors, am more likely to annotate and more likely to provide a comment when clicking the observation. It wouldn’t bother me at all to see the thumbnail agreement disappear.

5 Likes

I think this would be a reasonable compromise, if there is anyone who feels strongly about keeping the agree button.

3 Likes

If it is kept as it is now, I would be in favor of a separate DQA for duplicate observations and a separate DQA for observations containing photos of different organisms, so that these observations become “casual” and no longer appear in the Identify thumbnails. Maybe this could be a separate topic once a decision is made about the agree button.

I think the “Agree” button on Identify thumbnails should be removed if:

  • there are any comments
  • the observation has a description
  • there is more than one photo or audio file
  • there are no photos or audio files
  • an issue (besides captive/cultivated) is marked in the DQA
  • the observation is flagged
  • something else; I commented below
  • I don’t think the agree button should be removed in any of these cases

0 voters

6 Likes

… it’s an observation.

(The agree button on thumbnails should be removed.)

6 Likes

At least 2 staff have said they frequently thumbnail, so that’s why I didn’t include it as an option, since I think it’s unlikely to be removed completely. But yeah, I never thumbnail myself.

2 Likes

I can agree there are some observations that are just so easy to ID off one photo even as a thumbnail, but then I also find it just as easy to open thumbnail one, ID / agree / comment / flag / annotate, and then right arrow onto the next one. But if people use it as a preference then we shouldn’t get rid of it outright!

1 Like

Having the agree button on thumbnails encourages actions which often lead to bad outcomes. The staff are more than experienced enough to avoid the pitfalls, but I still think it would be a net improvement to remove them, since there are so many more less-experienced users. So, one more vote for that.

5 Likes

It is already conditional on whether it displays or not, so I think changing the conditions for it to appear are better than getting rid of it outright!

Perhaps only staff, curators, and maybe high volume IDers with low error rates could have the thumbnail agree button? This would require different permissions, which may be a coding issue. But, I’m assuming at least staff and curators already have different permissions and UIs than normal accounts. For an added coding challenge, make so users that have it available can turn the feature off.

1 Like

I used to think that until I realized that I might id an easy bird on the thumbnail, but then next pic I couldn’t see was just a flower, and the third one a tree. Or, easy bird was already observed in a companion record, and the observer was asking for the tree it was sitting on this time.

I see the latter regularly- almost have research grade, and then someone comes along and id’s something else in the pic. Gosh, why didn’t they just read the specific instructions in the comment? …Oh, maybe a thumbnail agreement.

4 Likes

Um, no… that comment is about single photo easy observations, and in particular how some identifiers will not want the “agree” button taken away from the thumbnails. Hence we were discussing removing it only for multiple photos etc.

This won’t be from hitting “agree” on a thumnail, because to agree, there has to be an identification in the first instance, which would presumably be of the subject organism!

1 Like

Oh duh on the last point! So, I have no explanation why someone would just ignore what the observer is saying.

1 Like

i kinda don’t understand why people do it, i have a super slow internet connection and even for me paging through the observations the normal way is super fast at least if you aren’t enlarging photos

3 Likes

I just noticed that my dashboard notifications have “agree” buttons for Research Grade observations, but not for “needs ID.” Is that relevant to this discussion?"

I think observations at rank genus and coarser don’t have Agree buttons.

2 Likes

So it seems like some of the same issues we’re voting on here (identifying without being aware of comments and photos of multiple organisms, etc.) would be applicable there.

1 Like