:unsolicited usually commercial messages (such as emails, text messages, or Internet postings) sent to a large number of recipients or posted in a large number of places https://www.merriam-webster.com/dictionary/spam
iNaturalist follows said definition and applies the term as so:
While using ‘spam’ in this way is correct, many iNaturalist users tend to think of the word in more general sense, referring to repeat, unwanted actions. As such, countless spam flags don’t actually refer to our meaning of spam, instead they involve duplicate observations or copy-and-paste comments. This all creates additional work for curators and risks new users being suspended due to false flags.
I’m wondering what can be done to reduce the number of false spam flags. Should the word ‘spam’ be replaced by a term with a more narrow definition, such as ‘soliciting’? Or perhaps the definition of spam needs to be more visible to iNat users?
An interesting point! I looked over the recently resolved false spam flags and many of the flaggers appear to be english-speakers, though a few of them also seemed to be bilingual.
Thank you for the screenshot too, where on iNat can I find it?
When we flag a comment, spam is the first option on the list.
Since iNat uses a ‘narrower’ definition, they could list it after the Other options.
Also I just checked – Spam is already clicked. Which seems, odd.
Offensive
Copyright
Spam
AI (since hopefully there is, not yet, much of that)
Other
That’s a good idea. I think for photos, Spam should maybe be just above or below Other, and Copyright Infringement should be on top, since I believe it’s probably the most common.
If there’s any way to not have one of the radio buttons already selected, as Diana said upthread, it would probably lead to more people giving thought to what the reason should be and making a conscious choice.
I think this is a good option. Quite a few flags I’ve recently viewed were mistakes; the flagger not even aware they were making a flag. Not having a button automatically selected could help mitigate errors.
I asked our devs, and my understanding is that it may be possible, but once a button is selected it couldn’t be unselected. A workaround would be to have a “None” option selected, but I think we don’t want flags to have no reason cited. I don’t think either option would be an improvement. I do think moving the “Spam” option closer to the bottom and moving the most common reason at the top and selected by default would be an improvement. So for example for photos I’d make Copyright Infringement the default at the top, and for comments I’d make Offensive/Inappropriate the top default choice.
Are you able to message me a few examples of these?
The flag description definitely needs improvement:
Commercial solicitation, links to nowhere, etc.
Taken at face value, “links to nowhere” could include any 404. Also, that “etc” at the end is an open invitation to misinterpretation - especially given that there’s no accompanying link to the full help article.
Spam is a serious problem, so it’s absolutely crucial to have a clear definition right at the point of entry. For the same reason, I think the opening sentence of the help article needs work:
Like other things beginning with the letter “s”, spam happens. If you see something spammy, please flag it as spam.
I don’t wish to be a fun-sponge, but opening with a slightly off-colour joke and casual phrasing like “something spammy” gives entirely the wrong impression. Anyone reading this could be forgiven for thinking that anything slightly iffy is flag-worthy. I highly doubt most people will follow the wikipedia link and slog through a quite lengthy article trying to work out which of the multiple definitions are relevant.
First impressions are very important! The rest of the help article is much better, but we all know that many people don’t get past the first few sentences…
Having all options initially unselected would certainly be an improvement. It doesn’t matter that the button cannot be unselected, so long as a definite choice must be made first. This seems to be a quite common approach, which can be found on other websites that face similar issues. For example, here’s the flagging dialog from StackOverflow:
I agree, the description shouldn’t be this open-ended. I’m reading at the iNat Help page as a base to create a more specific description, though am having trouble understanding what exactly ‘spurious sites’ entails (@tiwane?). This is a very important distinction since links to other sites are posted to iNat all the time.
Maybe description of spam being ‘commercial solicitation’ is enough?
