Option not to train the AI?

Thanks @mira_l_b - appreciate this. Okay, will do.

Hi again @mira_l_b - thanks again, just reading through the notes on “Feature Requests” - it says:

“We are likely to not implement any large changes that are not already on that agenda until next year, so it would be best if you started a brainstorming topic in General rather than make a Feature Request.”

Seems there are two issues coming out of this thread:

  1. Settings option to “not use for AI”

  2. Settings option to “filter out Group Accounts” (and to remove group accounts from leaderboard)

Would these be considered “large” issues? Or easy to implement?

Do you think it’s best to brainstorm here first?

Thanks again.

not sure why you think this should be a topic of conversation. unless something has fundamentally changed since the post below, seems like you really should be working directly with iNaturalist staff to troubleshoot and address the real issue. until you make some headway on that, seems like everything else is just a distraction.

8 Likes

Yes @pisum had several conversations with Scott and Carrie about this and we’ve done our best - and have made it clear the issue is closed from our point of view unless or until iNat - the staff? the community? - wants to discuss other options (iNat can always delete the account if it wishes).

Not sure, however, what any of this has to do with the original thread - but anyway, will try to conduct a more positive discussion elsewhere; and if that doesn’t work, will just shut up about anything that Charlie might disagree with. :-)

ok. i think i get it. it’s not that you can’t do the oauth flow that iNaturalist proposed, you just don’t like the oath flow that iNaturalist proposed. maybe you don’t want your users to be aware that their data is being sent over to iNaturalist for some reason? maybe you think that if they have to create and sign into an iNaturalist account that that (complexity?) will scare off some users? maybe you just don’t want your users to even be aware of iNaturalist, perhaps for fear that they may see it and decide they don’t need your app anymore?

what i don’t understand is why you even decided to send data over to iNaturalist in the first place. if you were simply trying to allow your users who were also iNaturalist users to send their observations from Questagame to iNaturalist, then why would you object to iNaturalist’s proposed solution?

if you’re primarily trying to silently leverage iNaturalist’s community for identification power, then that seems to be just sneaky (and not in a good way). (isn’t that what your BEE thing is supposed to be for? isn’t your whole thesis that offering a monetary incentive for identifications is the best way to get identifications? maybe instead of spending a whole lot of effort on sending observations to iNaturalist to be identified, you should simply increase your bounty in BEE?)

frankly, i don’t understand how iNaturalist and the community here benefit from the observations sent over to the questagame “group” account. but if you can justify the value somehow, then maybe my next thought makes sense…

if you’re not against having your users sign up for an iNaturalist account, but you just don’t want to have to have them sign in each time manually, then i wonder if you’ve proposed setting up some sort of alternative (assertion or bounce) oauth flow with iNaturalist? seems like that could be a compromise. your users would just have to either set up or already have an iNaturalist account, and then they would just have to link the accounts once instead of having to sign in each session.

2 Likes

i’ve never seen any nonprofit organizations with investor relations pages. i’m not going to post the link here, but it’s easy enough to find, if you really want to see it. i wonder what they’re monetizing?

1 Like

A question that is probably only answered by QG themselves…

That’s a good question. On their investor relations page, they do say that they’re making money,

QuestaGame is the first company of its kind to establish profitability in biodiversity conservation. It’s demonstrated a solid product-market fit, while creating a global market for reliable ecological data and expertise.

So, it sounds like they’re selling “reliable ecological data and expertise.”

1 Like

It sounds like they have vague fears about AI and feel like it’s something that should be considered dangerous. However, I find their discourse about AI on their “Terms of Service” page somewhat incoherent. They assert, for example, that

What we’re trying to say is there are ethical considerations when it comes to training AI systems in the environmental sciences, especially if, at the same time, we neglect to provide opportunities for the amazing, organic supercomputers that exist in people’s skulls.

So as part of these Terms, let’s just say that you agree to be a thinking person, and we agree to do our best to prevent AI from infringing up your right to think and learn and be rewarded for your knowledge.

But, I have no idea how they imagine that an AI learning to recognize plants and animals could in any way diminish a person’s own efforts to learn those things, any more than my learning to recognize plants could diminish their ability, or pleasure, in doing the same.

2 Likes

My impression was that QuestaGame receives payment from users in the same way as many games do, and then channels that revenue onto various nominated recipients based on the identifiers settings. When I signed up to QuestaGame to check the platform out, I quickly decided it wasn’t for me, as I want to SIMPLIFY the observation/identification process rather than make it more COMPLEX, but for many the game aspect would be appreciated. I was given the choice of many recipients, and I chose iNaturalist to receive the proceeds of my identifications. Just because there is a form of accounting in the process doesn’t neccessarily mean they are “for profit”… many “not for profit” organisations exist with mechanisms to reward desired behaviour or significant contributions.

Observations are clearly marked as coming from QG, so if anyone is against what they are doing, they can simply hit “Next” or even filter out from the Identify query with the account name. For me, there are many interesting observations that come from QuestaGame users, and I for one think it’s a good thing that there is encouragement to involve that “market sector” into environmental awareness. Compare to the analogous “Pokemon Go” or whatever it’s called, where people run around “catching” make believe critters… :/

My main concern with QG is that there is very limited engagement between the two communities. For me, the engagement and community is the most powerful component of iNaturalist, and throttling that back seems very akin to the idea of making computer chips with advanced capabilities, but disabling parts of that and selling them cheaper…

Hey everyone, thanks for the comments. Just noticed these. Appreciate the feedback and will try to quickly answer each of the questions raised.

QGame is a “social enterprise.” It operates in the middle space between non-profit and a private enterprise. The company’s primary measure of success is biodiversity mapping and conservation, and is motivated by profit only as a means to increase its impact on these things.

QuestaGame is free to play, but we do receive grants and have “products” (revenue) and we do “pay” experts (donations to organisations of their choice). Our goal is to increase this pay as quickly and significantly as we can. These models are innovative and, we feel, align with the recent IPBES report on biodiversity which calls for a paradigm shift in humanity’s approach to biodiversity conservation (we think including the power of market forces should to be part of the equation).

The only real AI “fear” we have is the lack of reasoned discussion about its design and purpose; a lack of transparency; an inability to opt out. We’re actually big proponents of AI tech and see terrific promise in using AI to help accelerate iNat’s mission.

Re the ability for QG and iNat users to communicate - this seems a legitimate goal, and something we’ve been wanting to implement for a while. We think we have some relatively easy solutions for this and hope to trial them asap.

I should add that I’m writing as Andrew Robinson. I’m one of the co-founders, chief engineer, and a director of QuestaGame. I’m speaking here on behalf of the company, but please feel free to contact me directly if you’d like to discuss any these issues in more detail. I really like hearing new perspectives.

On twitter I’m “QuestaDrew” (I don’t use it often, but will reply to direct messages and kick off email conversations there).

Thanks again!

it looks like you guys are continuing to do things without consultation: https://forum.inaturalist.org/t/new-questagame-feature/4869. i get that sometimes it’s better to ask for forgiveness than permission, but it seems like you guys have overplayed that card here.

cross out “AI” from the above quote, and this fairly well summarizes how i personally feel about how QuestaGame has decided to latch onto iNaturalist.

1 Like

@pisum Remember the “assume good intentions” rule!

@questagame

I’m guessing there was a stuff-up in pushing through the wrong data, as the few I saw were largely not “comments” but were values, such as who had made the QG-side ID.

Firstly, I think a large scale test on a live system with users that have already shown a sensitivity to such things was probably not the best move here. Easy in hindsight of course!

Secondly, I think it should have been discussed with iNat developers more before running it. Easy in hindsight of course!

Lastly, what you were doing… it doesn’t really address what I was getting at. Discussion, where questions can be asked and answers given, views expressed and ideas shared. 1:1, duplex, intelligence to intelligence. I make IDs and comments on lots of observations, and many of those observers I have conversations with about what they are seeing. This forms a fundamental part of what iNat is about. When I see QG observations, yes, interesting data points… but I can’t ask questions, hold a conversation with them, or teach or learn from the connection! I am in iNat a) to share what I know, b) to add to what I know, and c) to connect with other naturalists! As the QG interface is operating at present, at best it is only as good as reading a book about what it’s participants have seen, which I can do down at the local library where they have thousands of books with nature pictures in.

What’s needed is a communication channel between iNatters and QGers, not a way to just show what the other side is saying. If QG were to have an in-game “identification bureau”, and iNatters were portrayed as staff at that bureau, and the staff at that bureau would be then asking the QGer questions about what they saw… and the QGer would reply back to the iNatter bureau staffer… “we iNatters” would appear gamified on the QG side of the conversation, and the QGers comments and answers would appear as normal comments in the iNat observations… but most importantly of all, a dialog can take place!

I’m imagining a lot of eyebrows going up at the thought of being gamified, but as a sufferer of “avoidant personality disorder” I can totally relate to a group of people that might need that in order to engage with nature and naturalists. Just consider the discussions we have had around “nope” species and gateway critters, of illustrations as a pathway to breaking down phobias on spiders etc… Like it or not, this is the world we live in!

One last point"
I’m just one “user” of iNat speaking out on how I see things… I am not staff or development, and my ideas and thoughts should not be acted on as if I were :)

4 Likes

Hi @kiwifergus - thanks for this. Yes, agree with your points and appreciate the feedback and ideas. Yes, we were developing a two-way channel (no, it wouldn’t just show the other side) in which members of each could communicate directly, back and forth. However, the best solution to resolve this - which would allow a safe testing space - would be the “group accounts” feature which separates organisations from individuals (we have this on our system); and then we could develop comms from there. But looks like that’s been a non-starter for iNat for whatever reason.

I think it is a fundamental part of the iNat concept, that the observer is very much a part of the observation. It is not explicitly stated anywhere that I know of (though it might be), but there are many places where it is alluded to, such as the discussions around duplicate observations and the validity of the same plant to be observed on the same day (and obviously place) by two different people.

Group accounts are discouraged, but do exist and are tolerated to a point, but most certainly not when they are large and impactful. This of course is my perception, I’m just an iNatter though… not in any way an official spokesperson.

1 Like

Why not respect what the iNaturalist staff discussed with you and just go down that path, even if it’s not the ideal path for you?

Even if they were being totally unreasonable, and it’s not just a case where you just refuse to really listen to and understand answers you don’t like, what is the point of making a fuss about it here?

3 Likes