Looking for tips on how to increase life list

General
,
21

At the first, you don’t need to be a scientist to love nature and to investigate the world around you. Even more, I can tell you as a failed biologist, often scientists don’t like nature and even don’t know the world around them.
Especially if you are 15. You have a lot of time to understand what really you want to do in your life and how to learn a nature (if you still want to do it as an adult).

About species, don’t forget that you live in nothern hemisphere, and winter is coming. More species will be at spring and summer.

3 Likes
22

My experience with Seek is that the Algorithm is quite bad at IDing photos already taken, even if it should be totally clear from the picture (would reach RG in iNat).
What really helps when I do IDing with Seek is “3D scanning” around the observation. That seems to help the AI. Of course that will be easier to do on something that cannot run away (plants and fungi, slow animals) than animals that can run away (like your deer or other animals).
So probably won’t help with the deer (unless it will stand still while you slowly walk around it…), but could help increase your life list when dealing with plants, fungi or slow insects etc. Or dead animals, since they also can’t run away :laughing:

23

The iNaturalist app/website still has a life list feature that is pretty easy to access, and you’ll rapidly get far more species on your life list if you use iNat than Seek, because it supports far more species to begin with.

5 Likes
24

I suggest trying to get photos of deer scat or tracks and see if it gets the species based on that. The algorithm sees and is trained on lots of those.

I had some fun with Pokemon Go but it got repetitive after a while, it wasn’t as novel based on place as I hoped (like there weren’t really special pokemon consistently near things that corresponded to the real world), it’s skewed against small towns and rural areas, doesn’t work without cell service, and the battle system is really rudimentary and arbitrary. Oh yeah and the statistics of the Pokemon are confusing and don’t always match their actual abilities. Whereas iNat is great for rural areas, works fine without cell service (except less so with the Next app), and has a more interactive community. On the other hand, there aren’t taxonomic splitters making 44 cryptic Eevie evolutions no one can tell apart :D

25

That’s awesome, I also collect animal bones!

26

I totally get the Pokémon connection! I grew up with earlier Pokémon games, so the concept of iNaturalist as a Pokédex for our world was a very easy and fun one to grasp when I was first getting into the platform.

I would echo some of the comments here; you should consider using the iNaturalist app (it’s on both iOS and Android), instead of or in addition to Seek. They’re very different apps. That way you yourself can confirm that you have in fact seen a while-tailed deer; and between community IDs and the computer vision suggestions list (rather than the single suggestion from Seek), you’ll get better feedback and higher accuracy on what you’ve found.

As for actually increasing your life list, I think plants and invertebrates are great ways to increase your species count. Plants sit still, which is very convenient; and looking very closely at plants from different angles, as you often have to do to identify a species, will often help you spot other smaller things that you can also add to your life list, like insects, spiders, snails, springtails, lichen, mushrooms, and so on.

If you’re able to find one, a macro lens clip-on for your phone can be great for these!

If you use iNaturalist itself, you can also check which species other people have seen in your area that you haven’t, and that can give you goals for the next time you’re outside. It’s not made clear on the web page, but if you search your area and then add &unobserved_by_user_id=username (replacing “username” with your username) to the end of the URL for that search, you’ll see which species others have seen that you haven’t observed yet. This can be a great source of inspiration.

2 Likes
27

One of my favourite things about associating with naturalists (professional and casual) is the level of excitement over wildlife. I’ve seen even elderly naturalists literally squealing over cool finds before:)

1 Like
28

This made me laugh (in a good way). I’m one of those elderly naturalists that squeal. I think I amuse my friends. I know I amuse my husband.

I found an insect that was the first (identified) observation for my state - and the closest obs is 350 miles away. I found a spider that our state’s premier spider identifier said ‘Great find! These are tiny and cryptic’. There are only 63 (identified) obs of this spider on iNat. And a gall that got the comment ‘Nice find, a fairly rare one, in my experience.’

Yeah, I squealed for all of those. This has not been a great year for me in birdwatching. But it has been a spectacular year for me in insects, spiders, and galls.

5 Likes
29

My sister and I have a running joke where she reports to me all the ridiculous things that Seek suggests when she tries to identify a deer. It is a whole thing. The iNaturalist app, on the other hand, is so good at identifying distant photos of deer that sometimes it will suggest a deer even if there are no animals visible. :sweat_smile:

I highly recommend switching to the iNaturalist app. It takes a little more investment in creating observations to go with your photos, but that’s just slightly more complicated gameplay, isn’t it? You can still use the computer suggestions to make a preliminary ID (and they will be better suggestions), but also you can connect to the wider community for help. There is also a new version of the app on the way that will make the suggestions even easier to access.

If Seek is like a single-player game of “Identify All the Things” then iNaturalist is the cooperative multi-player version, and like many multi-player games, it opens up gameplay in whole new directions. I’ve identified so many new things since I switched over to iNaturalist. The combination of computer suggestions and direct human knowledge just can’t be beat by computer suggestions alone!

5 Likes
30

I can relate – and I say this as someone who holds an M.S. in biology.

My thesis advisor’s approach was, “First come up with an interesting question and only then choose a study organism.” Little did I know how difficult that is for me. I ended up asking my advisor what research studies they had been intending to do and hadn’t gotten to yet, and then I picked one. Nothing wrong with that as a grad student – it served the purpose of teaching me the skillset of designing and carrying out a study, including the analyses.

The upshot, though, is that after my thesis and the journal paper based on it, which were heavy on experiments and statistical tests, everything I have published since then has been observational studies. I’m just so enamored of taxa as entities, it is difficult for me to formulate more general ecological questions. So, surveying the butterfly fauna of a given location, or elucidaing the life history of a given lepidopteran species comes much more easily to me than, say, formulating a question about the relative importance of different mortality factors under different environmental conditions.

All those ANOVAs, regression analyses, Spearman’s rank correlations, and what not seem so removed from the actual organisms that got me interested in biology in the first place.

Anyway, getting back to the OP’s question, my impression of Seek is that it was mainly created as an alternative for kids too young for the main iNaturalist. If there are any older Seek users on the Forum, they can set me straight as to what they prefer about it. The minimum age to use iNatualist is 13, so at 15, you are old enough. But to get to the more immediate question: you mentioned

If this is the case, the solution would seem to lie in the developers providing Seek with a more complete set of training data. Now, someone who knows the backend please weigh in if I’m wrong here. Presumably, because Seek is not connected to the iNaturalist database, each instance of the app would have a self-contained instance of the AI. A standalone AI. That would mean that more Seek users collecting pictures of deer would not help to train your instance, but a downloadable update to the app with an updated AI would. Am I on the right track?

1 Like
31

I think SEEK uses an older pre-Geomodel version of iNat’s CV. Something like that.

32

Once long ago i took a photo of a dusty looking field with some walnut trees, and tried to add it to iNat as a walnut observation. The algorithm insisted it was a white tailed deer. I was laughing at the algorithm with my friend and started walking off only to have a deer jump up and bound off. it had been hiding in the grass. I don’t think the algorithm could actually see it, but maybe it just detects heavy deer browse or something. Or there are just deer everywhere.

It’s not for me either. Maybe some aspect of neurodivergence… i’ve got some ADHD traits sprinkled amidst my autistic ones and those repetitive super narrow studies just break my brain, even though i am very good at hyperfocusing on other things. I’ve got no desire to ever do that type of research, even though people seem to think i should. I also can not deal with the cultural, social, and political environment around universities. I really can’t stand it and piss people off very quickly, same as i piss academics off on this forum…

33

I think it’s as simple as the computer being very good at detecting patterns, but it can’t always distinguish which patterns matter. So the computer is detecting “this is the kind of photo that people often claim has a deer in it”, not so much “there is a deer in this photo”. The computer also commonly suggests woodpeckers when looking at photos of tree trunks, and warblers when looking at photos of leaves seen from below.

Have you considered theoretical physics? Of course you have to really like doing math. :sweat_smile: (This is a joke about theoretical physics, not a practical suggestion. If I were making a practical suggestion, it would be: Doing research is not the only way of doing science, surely? For example, science communication is critically important!)

I think it’s just a matter of how frequently they are able to update the various apps. In theory, there is no reason why Seek cannot use the latest model, but in practice there are limited resources to devote to making the necessary updates.