All rights reserved observations

frousseu · February 18, 2021, 3:53am

I just recently learned that all rights reserved observations are not getting out to external platforms such as GBIF, where, presumably, a lot of the science that is done with iNat data will be going. I am also noticing more and more that a lot of users seem to have their observation copyrights set to all right reserved. It also seems to happen a lot with new users or users with few observations. I can’t remember where I read that, but apparently about 65% of observations right now have a license that allows them to be exported to GBIF. I find this really unfortunate as it leaves out a significant portion of the data that could be used for science or conservation.

My feeling is that this is mostly a byproduct of people choosing to retain all rights over their photos instead of people consciously choosing not to share their observations. If you’d ask any iNatter whether they would like to share their observations (species, date, location, but not photos or recordings) with scientists or biologists so that they can use the data for conservation or for better understanding the distribution of species in space and time, I would expect that the vast majority of people would say yes, of course!

My questioning is whether the current way of choosing licenses should be improved so that observations don’t inherit from the same licensing chosen for the photos or the recordings. There is actually a feature request here Provide separate checkboxes for media and observation data licensing upon sign up for separating the observations from photos when subscribing to the site, but it apparently did not attract much attention (ah, this is where I got the % mentioned earlier!). To me, this request makes a lot of sense. I actually don’t remember what is asked when one subscribes to the site (or through the app). Could it also be a possibility to have a default for observations (CC0, Public Domain) different from the default for photos or observations? I bring up the subject here and the feature request, because I think a bit more discussion might be needed around this.

Another minor point is that right now, we can filter for licenses when IDing, but it seems that we can do it only for photo licenses. Am I right? Or perhaps, we could filter for observation licenses through urls? Anyway, my point here is that I may not want to spend my time IDing observations that people choose to retain their rights over. But of course, I may keep doing it since it is probably more a question of people not realizing that they chose to make their observation not usable (related Cannot filter for observations with © (All Rights Reserved) photos).

earthknight · February 18, 2021, 4:51am

It might be worth looking through this thread too:

https://forum.inaturalist.org/t/can-someone-explain-licensing-to-me-please/2691/10

melodi_96 · February 18, 2021, 5:03am

Owner can change licence later and anyway iNat is about users, not data, so, we need more awareness, but also we need to take it easily that many are not ready to give up “all rights reserved” same way as people are obscuring locations of all their observations, near home or not, it’s their right. As ider you may consider that adding any id helps science, those observations are on maps anyway, anyone else can go and look up what is there, taxon of it can be very “needed” to be found in an area. And having most of observations with more opened licences helps it too. So, don’t focus too much on it, all rights reserved doesn’t mean it’s a lost case, iNat itself can be a great tool and default now is not a closed licence. In fact I’d say having duplicates and all the messy stuff with opened licences showing up on GBIF is a problem too and one may think a lot before allowing new users to get stuff to GBIF even easier than now.

murphyslab · February 18, 2021, 5:14am

Some very good points, @frousseu. I’m particularly sympathetic to your sentiment:

I may not want to spend my time IDing observations that people choose to retain their rights over.

Even with filtering only for photo licenses, it’s frustratingly limited. The options available are:

Any
Attribution
Attibution-NoDerivs
Attibution-NonCommercial
Attribution-NonCommercial-NoDerivs
Attribution-NonCommercial-ShareAlike
Attribution-ShareAlike
No Copyright (CC0)

My own preference would be to have search for photos that fall under an “Any-CC” or “Exclude-Unlicensed” criterion.

And in terms of making more photos and observations available for re-use, it would be beneficial to have dual licensing available on iNat. Consider that Wikipedia does not allow CC-BY-NC contributions, while GBIF does not allow CC-BY-SA contributions. CC-BY-NC-SA of course works for neither, however a user may not be comfortable with eliminating all restrictions in favour of CC-BY or CC0. Why not allow a user to directly apply a dual license of CC-BY-SA and CC-BY-NC? Dual licenses like that could also be useful for limiting the impacts of license changes.

prokhozhyj · February 18, 2021, 7:31am

Very many users do not look at their licenses at all, but simply click “yes” on the proposed by system when registering. So it would be good if the system in all cases and for all registration paths would offer GBIF-friendly licenses by default, and if anyone wants to have “all rights reserved”, then a person concerned about this issue can set such a level manually.

fabienpiednoir · February 18, 2021, 8:59am

100% agree. Creative Commons licences should be the default choice rather, with a possibility to opt to copyright.

wouterkoch · February 18, 2021, 9:03am

Claiming rights on an observation (which is basically a non-copyrightable factual statement) is always a bit dodgy. In fact, the license options GBIF offers besides CC0 were a compromise to address some unrelated concerns in the discussion.

In my opinion it would be more correct and intuitive to make a single distinction: do you want your observations shared with GBIF, or not? It would be great to be able to filter for this and have it clearly visible. I would even go so far as to say that “Research grade” loses its meaning if an observation is not being shared.

cmcheatle · February 18, 2021, 1:40pm

Not sure if you are suggesting this default be for just observations or both them and photos.

This was covered in a very long thread here
https://forum.inaturalist.org/t/a-case-for-changing-the-default-license-to-not-include-a-nc-clause/18690/84

For observations I really have no opinion since you can’t copyright a fact anyways, the issue here is people cutting and pasting already copyrighted text and then ‘relicensing’ it.

For photos, I 100 percent disagree. Too many users would be somewhere between upset, angry or surprised to find out they have licensed away rights to their photos via a open license. If a user wishes to do so, that’s great, but they should be the ones making that decision and actively implementing it. Not having it made for them in something few users review or read.

trscavo · February 18, 2021, 2:08pm

I changed my default license when I found out that Wikipedia does not allow CC-BY-NC. I didn’t know the other fact about GBIF until just now. If I had known that, I would have changed my default license to accommodate both Wikipedia and GBIF.

That’s a great idea. Be sure to ask about Wikipedia separately:

Do you want to share your photos with Wikipedia?
Do you want to share your observations with GBIF?

That’s much better than asking a naive user about actual licenses.

cthawley · February 18, 2021, 3:20pm

I am definitely on board with having plainer language and a more permissive default license for observation data. I think most people do want to share their data when they come on the site, and I’ve heard of folks who thought they were sharing data when they weren’t (they didn’t realize they had their default observation license to all rights reserved which they picked for their photos). Hcek, I didn’t understand the difference for a long time.

It’s a complicated topic that we shouldn’t assume the average or beginning user will understand.

I think changing this default is probably one of the single biggest changes that iNat could make to increase data-sharing (and I think it would be broadly supported).

I understand a lot of the issues for and against the current default photo license (in threads cited by @earthknight and @cmcheatle above), and my sense is that, while the current situation is a compromise that leaves no one perfectly happy, there isn’t going to be a strong consensus to change it.

So leaving the photo license issue aside, I think having an option that says something like that proposed by @wouterkoch is a good idea:
“Do you want to share your observation data with scientists?” (and then one of the little question hover icons that would explain this means your data will be shared to GBIF and what license it would have).

I’d also be in favor of having the option to filter for observations (not just photos) by licensing, as I’d prefer to concentrate on observations being shared and not restricted. Being able to filter for those could also address issues like those in this thread, with questions about sharing/citing maps from iNat which show both observations with licenses that allow sharing and those that are all rights reserved.

prokhozhyj · February 18, 2021, 4:00pm

O’k, but then it shall be better to make this stage of registration less “automatic”, say, not to issue a default license at all, but ask to choose (as a mandatory stage of registration). Because really too many people do not think at all about what licenses they have and what follows from this.

edanko · February 18, 2021, 4:34pm

Maybe the observation can be added to GBIF from iNat once research grade, just without the images, if the images are the only problem?

Also, I really don’t think anybody has GBIF in mind when they reserve rights on their images. Couldn’t this be added to the fine text somewhere on the site?

cmcheatle · February 18, 2021, 4:41pm

This already happens assuming the observation is licensed appropriately.
ie https://www.gbif.org/occurrence/2826450921

All my RG records are in GBIF without the photos (my photos are licensed All rights reserved - I’m not a professional photographer or anything, I just don’t think it is unreasonable to wish people to ask to use my property).

melodi_96 · February 18, 2021, 5:01pm

cthawley · February 18, 2021, 7:00pm

As I understand it, the observation data will go to GBIF if the observation license is set correctly, regardless of the photo license. So users can set their photo licenses to all right reserved and still have data go to GBIF.

frousseu · February 18, 2021, 11:19pm

This sums up what I think quite well!

frousseu · February 18, 2021, 11:30pm

Although GBIF is probably the main source for people wanting to use iNaturalist observations, I think that any questions that could be used to determine whether a person wants to share their observations or not should not be restricted to GBIF as observations could be accessed through other platforms. For example, it is also possible to download observations directly from iNat.

gcwarbler · February 18, 2021, 11:58pm

Here’s just one example from me as an author: I’ve prepared a number of papers for publication which have utilized images downloaded from iNaturalist. But I never even bother to check the license status of the individual observer because I explicitly contact each observer in advance regarding the use of their image(s). If I get permission (which has never been refused), I use it; if not, I don’t. It’s just that simple. That bypasses any worries or tedium about checking for licensing limitations. It’s always, “Used by permission”, and I always acknowledge the contributing observers in the published paper. When I contact an observer (via iNat or by email), I’ll always include a brief description of the research, the intended target publication, and may even explain how I would use the image.

frousseu · February 19, 2021, 1:39am

Contacting each observer could be possible if you worked on a single species with few observations or in a restricted geographical area. But, suppose you are working on a group of species throughout their whole range or you are working on a large scale biodiversity assessment across several taxa. It would be absolutely impossible to contact each and every observer contributing to the datasets needed for such questions. Also, the concerns here are for the copyright of observations, not images. I completely understand why someone would retain complete rights over their images.

frousseu · February 19, 2021, 3:26am

I started looking at the other topic suggested by @earthknight and it generated another question/concern related to the licensing of observations. I actually had already read part of that discussion, but as a lot of the discussions in this forum, it is rather long and complex and a bit of a garden of forking paths… Anyway, from what I understand, observations may not be copyrightable, at least in the US, but there seems to be some uncertainty around this and it might vary across countries. Suppose that they are and an observation has a CC license. All CC licenses seem to require that the licenser be credited for their work. Suppose one wants to publish a paper making use of the observations of several thousands of observers with CC licenses. The CC license seems to imply that every observer should be credited in the paper for the use of their work. Is that right? If so, wouldn’t it make a bit impossible to make such a use of observations? Thousands of observers cannot easily be credited at once. Is my understanding correct? I’m new to licenses and I find this rather complicated!