Spam accounts - enhance users "Identifications" page to manage deleted content

TLDR: Identifications can end up in limbo. User is only presented w/ option to withdraw their ID for a removed post, currently no way to delete that ID since the link to manage the ID no longer works.

Background: I identified an observation as Human, then flagged the content as spam. The observation, iirc had used a fairly obvious stock photo and used the description field to add links to a spam, pirate TV subreddit.
Recently, going through my identifications page, I saw that my ID was still considered Leading. Odd I thought, and clicking through to the observation led me to see the account was suspened. Which was great news, the system works. But going back to my list of Identifications it dawned on me that now I had a Leading observation that would never proceed to the coveted status of Improving.
Seeking a way to delete my ID, I found I could not. Hovering my mouse only showed the option to withdraw it, certainly not the same as deleting it, which is what I was seeking to do.

Feature Request: When next redesigning userā€™s Identifications page, consider whether their IDā€™s should remain listed, deleted, or allowed to be managed in some way, when another userā€™s observation or account is removed.

Thoughts: Thereā€™s an interesting conversation to be had here. Options include iNaturalist deleting the observation and Identifications wholesale. A notification could be sent, saying one of their identifications was removed because the observers account was deleted. I donā€™t think the current state is optimal. When users identify content that is latter removed, those Identifications end up in a limbo state where the thumbnail remains, and the user is unaware that the rest of iNaturalist can no longer see that content.

Below: Screenshot of how that observation looks currently in my Identifications page, and the non working link to the since deleted observation.


https://www.inaturalist.org/observations/35404961

Epilogue: I also reported the users and subreddit in question to the mods of Reddit. They and the subreddit have since been removed from the site. Based on a true story. Names of those involved have been changed to protect the innocent.

2 Likes

Perhaps this problem could be addressed by one person coveting less (instead of the community being tasked to change an existing infrastructure).

ā€œImprovingā€ isnā€™t ā€œbetter,ā€ and ā€œleadingā€ isnā€™t a mistake that needs to be expunged.

8 Likes

Thanks @schoenitz for picking up on that. The gamification of identification on the site rises with each enhanncement that encourages inter-user behavior.

one person

The system is designed to encourage users to make a species level identification. I can see how my comment implied, ā€œIā€™m frustrated, I have an explicit goal I want to reach, and I need this feature to better my scoreā€. But in truth, with some cheekiness, I was just writing from the perspective of the majority of users on this site as I see it currently. I believe that with the exponential growth of iNaturalist, and the long tail of one-off, short term users -more than 50% of them are aiming for a species level identification. For better or for worse.

If I edited my post, deleted that line about coveting, could you see a benefit to discussing how deleted posts are handled? In short: Why take away the ability for users to delete or modify their own identifications in these circumstances, given that the IDā€™s are still preserved and counted when iNatualist presents them on their Identifications page?

1 Like

Not sure what the issue is here. If you add an ID to someone elseā€™s record and then decide it was incorrect or ā€“ for whatever reason ā€“ you want to recant, you can withdraw it. Most of us have done that for various reasons, although usually because we believed our ID was wrong or weā€™re uncertain. Yes, it appears as a withdrawn ID and doesnā€™t actually disappear. Is that a problem?

4 Likes

Iā€™m having a hard time distinguishing between the actual concern here and the ā€œcheekyā€ tone of the complaint. I donā€™t think itā€™s as common as you might think for most users to worry much about identification withdrawals, species-level identifications, leaderboards, etc. I agree that the ā€œgamificationā€ of iNaturalist could use a rethinking, to the extent that it encourages worry over those features.

6 Likes

@jnstuart

Problem is a strong word, I think itā€™s amatter of clarification/notification.
The user can normally chose three actions [edit,withdraw, delete]. When the second user has their account suspended, those actions are deminished to [withdraw]. That might not be a problem, but itā€™s an odd behavior to take some (minor) autonomy from the user.
A compounding oddity is that the thumbnail is still preserved.
Compare this to how iNaturalist handles it when an observation has their photo flagged as copywritted -the observation still alows the user to edit their ID, but the thumbnail disappears from their identifications page

Simply, thereā€™s work to be done consolidating how users manage their identifications, and what notifications they get regarding actions to posts theyā€™ve submitted identifications to.

Well out of the gate, I am on a back footing here. My thinking for making a feature request, was to create a space to talk about the particularities of what control users have from their identifications page.
I agree that a minority of users arenā€™t combing over their past identifications, but at the same time I was confused why I couldnā€™t delete an ID, I was also surprised that there was no notification that that user was suspended at all.
I image even ā€˜greying outā€™ an identification like the one I posted might suffice. Greying out, would preserve the ID, but visually let the user know that the account has since been deleted or suspended.

I do hope iNaturalist closes this thread if they donā€™t think changes need to be made. But I saw a place where there could be an imrovement made to consolidate some of the functions that are available to users from that (identifications) page.

4 Likes

Users do worry about species-level notifications (without them no data is useful anyway), and yes, some of us would like to delete an id, e.g. wouldnā€™t want to get notifications which you do get if id is withdrawn (little out of post theme). Leaderboards help people do more work, and many do care about them too, lots of people who do a lot of identifications do.

1 Like

Somehow off-topic but let me say that it is good to read that sometimes we succeed in getting rid of these users who are (yes they are) nothing but spammers.

2 Likes

I have some confusions over thisā€¦

what does TLDR; mean?

Why do you need to change your ID in this situation? The suspended account and/or flagged observations wonā€™t be RG, and in fact the Homo sapiens ID puts it to casual by default. The observation is still there (curators can still see the hidden content) and there is every chance of a mis-behaving user correcting their behaviour and/or whatever it is that got them suspended, so it would be prudent to let the ID stand. The observation itself doesnā€™t have spammy content, and I presume it is just the Account profile that has that, hence the suspensionā€¦

I can remember being quite excited the first time I saw one of my IDs marked ā€œmaverickā€, but that might be because I was picturing ā€œTop Gunā€ in my mind :) The Leading and Improving are labels that to my mind are more about how the IDs relate to each other, than any status that I might be trying to achieve with my IDs. That relationship between IDs is in turn more about how the Community ID calculatesā€¦

1 Like

ā€œToo long; donā€™t readā€ meaning ā€œthereā€™s a bunch of text below, hereā€™s a summary.ā€

3 Likes

I guess itā€™s up to you, but if you know itā€™s spam, why ID it? Or did you realize it was spam only after you added the ID?

In general we donā€™t delete content unless it is obscene, offensive, etc. Spam is simply hidden from public view, although staff and curators can see it. But I agree that a user should be able to delete their own ID if they so desire. So would your request be to have that option for otherwise inaccessible IDs?

Also, this is under the Bug Reports category, not Feature Requests. Did you originally create the topic in Bug Reports, @KC1LQM?

A separate discussion, but I would push back on this. A lot of taxa canā€™t be IDed to the species level from photos, so a coarser ID is best. Doesnā€™t mean the data arenā€™t useful, or that teaching/learning isnā€™t happening. Although yes, itā€™s good to aim for a species-level ID if the evidence warrants it.

2 Likes

For such taxas higher levels are somewhat similar to species in terms of identification, so people aim for genuses or families if thatā€™s the best they can get.

1 Like

Pedantā€™s Corner: ā€œToo long; didnā€™t readā€.

6 Likes

ultra pedants view: it would be ā€œdidnā€™tā€ if after or in reply (past tense), prior would be a warning (future tense) hence ā€œdonā€™tā€.

This is what confused me, it didnā€™t seem to be a reply to anything, so I thought maybe my understanding of what it meant was wrong.

2 Likes

This is a bit of a technical question, but is hidden content somehow made invisible to web spiders and crawlers on the site? If not, then by not deleting, are we not simply allowing the spammers to get away with their goal which is not really to get any inat users to click (although they will take it if it happens), but rather search engine result manipulation?

Or does the spam ID tool used by the site require you to keep them ? Can they not be read into it and then deleted?

2 Likes

I agree we should be able to delete or change our IDs on observations that are ā€œlockedā€, such as suspended accounts. Iā€™m not a fan of deleting IDs generally and prefer withdrawing them, because I think the progression of IDs tells a storyā€¦ at the very least it can alert others to the difficulty of IDing the particular observation/taxa and encourage them to take more time on those in future observations.
When I change my ID with a withdrawal, I am likely to supply a reason to explain the change of mindā€¦ but with a delete and re-ID, i might not feel the need to explain the change of mind because others wonā€™t see it as a change of mind, if that makes sense. I think anything that encourages discussion and promotes sharing of knowledge/expertise is a good thing!

3 Likes

didnā€™t is the accepted meaning for that string of 4 letters.

The acceptance
https://www.urbandictionary.com/define.php?term=tl%3Bdr

4 Likes

addedā€¦
https://forum.inaturalist.org/t/glossary-inatforum/1966
:)

2 Likes