Duplicate prevention: Notify observers if their image checksums match others on the site.

I’m afraid the checksum solution can’t be workable, because it is * legitimate* to post the same photos for multiple organisms within the photo. One often has to do so, for example, with a pond full of ducks, who do not cooperate nicely when asked to segregate themselves by species. Or, to name another example, a flower with an insect on it.

2 Likes

The request doesn’t say that users wouldn’t be allowed to upload the same image more than once, it only says to

(I would also point out that cropping your duck or insect would be a good idea regardless of whether this request is implemented or not, and the cropped photos would no longer be identical.)

8 Likes

Personally, I find it preferable to use the duplicate feature for these cases. The reasons is that when duplicating the observation, it doesn’t upload a second photo but instead links multiple observations to the same photo in the database. As a result of that, it is possible to get a list of all the observations/species in the picture by clicking the info button on the image. I find that a very convenient feature. I suspect a lot of people aren’t even aware of it though.

7 Likes

I’m certainly not…

2 Likes

I think it’s a really neat feature and I wished it was more obvious. Here’s an example of what it looks like. Notice the list of associated observations to the right of the picture. The only way I know how to get more than one observation associated with an image like this it to use the duplicate function under the edit menu on the observation page. There is no way currently that I’m aware of to merge observations with separate uploads of the same image.

3 Likes

It’s much faster to post all at once though, there’re many ways to link observations together, cause link like that has no actual benefits for observer or ider.

Well, personally I prefer having the links on the photos over other ways of cross-linking observations because not every photo may have multiple species (e.g. first one cropped to just one individual). This method provides a list on a per-picture basis rather than per-observation basis, which seems much more accurate than e.g. an observation field. It does take a little bit more time to set it up this way, but my workflow is rather slow anyway so speeding things up is not necessarily a priority for me.

It is better not to upload the same image several times, but rather to upload the image once, and then duplicate the observation. You will then have all links to the different observations on the same image, e.g.:

There is literally a “duplicate” button built into iNat. Clearly the devs do not find any issue with people duplicating observations in order to make different observations out of a single one, and each observation still has a different address, so it doesn’t really make much, if any, difference, if someone duplicates or uploads more than once.

People don’t always have the time (or want) to mess around making multiple edits of the same photo, but with duplication you can’t easily specify why you’re making the change, without extra text.

I will often take the time to do so, and add a maker indicating what the observation is of or an inset image to images that need extra detail, but that’s only when I’m editing photos back at my computer. When I’m in the field I’ll just duplicate the observation, as in these oyster and barnacle observations.

I don’t hold it against anyone if they decide to upload twice, or to simply duplicate an observation, as long as they add an initial ID or at least a clarification as to what the difference is meant to be, nor do I think anyone else should.

4 Likes

For globalbioticinteractions it is useful to duplicate the picture, but probably this is not the problem.
https://github.com/globalbioticinteractions/globalbioticinteractions/issues/822#issuecomment-1259645925

1 Like

There is a huge difference between duplicating the observation (by using the duplicate button) and uploading the same photo several times :

  • if you upload one photo and duplicate the observations: all observations have a solid link between them: one photo in the database (and also, one photo on the server);
  • if you upload the same photo several times, the observations have no link between them, unless you add a comment or an annotation which is just a more complicated way to do what a duplication can easily do (and also, more space used on the server, unnecessarily).
1 Like

So what? That hardly matters if observations are linked or not.

Many people have interest in interactions between e.g. pollinators and plants, predators and preys, pests and plants, etc.
Link between observation is therefore very useful!

2 Likes

You can do all that using observation fields, as many traditional projects do. Having the observations automatically linked is nice, but not at all necessary.

1 Like

It’s easier to find linked observations if they share the same photo (with duplicated observations) rather than manually look for a duplicate observation and add observation fields to the photos.

And those people can do it via myriad of other ways, observers already do quite a lot, it’s pretty inconvinient to duplicate observations on iNat.

It’s very easy: you just have to click the button « Duplicate »!

Hm, if I say it’s not easy then I know what I’m talking about, unless you have 100% the same data for both observations, it’s not as fast as a click.

i think this is wandering away from the original topic.

4 Likes

It is actuallly not wandering away from the original topic. The question is about duplicate prevention and not uploading twice the same photo.
Some people argue that they upload twice the same photo for two different observations (two different species on the same photo), and other that you shouldn’t upload twice the photo, but duplicate the observation instead.
In a well built database, you should never have the same information at two different places, so you shouldn’t have the same photo uploaded twice.

1 Like