HulloThere

This is a good one

Something about this cover image and title is giving real "Les Cousins Dangereux" energy to me

They look like how reef fish look in 18th century scientific illustrations!

Aha good job - there were lateral photos that show them off better but I love the facial expression on this fella

A fish with three pairs of horns along its head and back and an intricate striped pattern on its bulbous lumpy body swims among seaweed. Credit: Alan Travers via Wikimedia.

A male Aracana boxfish looks at the camera. He has a long face with a small mouth and bright, garish turquoise stripes on an orange body, with a matching loopin pattern on his tail and hexagonal spots elsewhere. Julian K. Finn via Fishes of Australia

What the fuuuuck how am I just learning about Aracana. Look at these whimsical little dudes!!

Rhizanthes lowii "flower" photo by Ch'ien Lee, a terrible plant structure that looks like a free-range rectum, positively swarming with delighted flies

"flower" of Mitrastemon yamamoti from Suetsugu (2018), another stinky plant structure that attracts a VAST array of insects (including flies, beetles, cockroaches, orthopterans, etc) to come nibble its horrid pink flesh

reading up on holoparasitic plants, regretting it instantly

WW2 airmen named their inflatable life jackets after Mae West

There is a single Benthosema glaciale floating around somewhere off Mauritania that for a split second today visualised the face of god then immediately forgot it when a particularly tasty euphausiid swam past.

Speaking of which happy Casimir Pulaski Day:

www.youtube.com/watch?v=TfEk...

Lying back and kickin your feet up

Yeah for marine species WoRMS is the place to go. Maybe GBIF will have it for terrestrial fauna?

Fucking soy right strikes again

Yeah, the correct answer (according to my taxonomist supervisor) is “you should cite all of them” but no one does. I think in my work I’ve tried to make sure any in-text references get authorities but I’ve not bothered with tables etc.

The metal trap is definitely real - I have three elder brothers and the oldest is a big metalhead so it basically took until I got to uni for me to develop any of my own taste

This guy thinks there wasn’t any blood in There Will Be Blood

This could’ve been a core part of my music intake as a teen… instead I just listened to Minutes to Midnight 1 billion times… a wasted life…

There must be a term (probs German) for the unique melancholy of discovering something really good about 20 years after it was contemporary. Like, who am I meant to talk to about Sufjan Stevens’ Illinois now, it’s 2026!

Ahh man I liked the EUS, gonna sit alongside Shuttle Centaur in my head forever as a “what could’ve been”

Oh hey that’s a gurnard, probs grey gurnard (Eutrigla gurnardus) but they’re hard to ID.

happy 14th anniversary to all who observe

A silver fish hovering head down over flat sandy grey

For clarities sake I am still working on the PhD (but I have been working on the BDFP for longer…)

Bathyclupea is such an odd little family. What you up to fella?

Just over a hundred black digital pencil drawings of bony fish on a white background- this image mostly consists of catfishes, eels, and freshwater fishes

Just over a hundred black digital pencil drawings of bony fish on a white background- this image contains lots of deep sea fishes and gobies

Just over a hundred black digital pencil drawings of bony fish on a white background- this has a lot of spiny fish, many quite normal but also seahorses and flatfish

Still working on my Big Dumb Fish Project*! Just hit Family No.350, the Bathyclupeidae - only 128 families to go 🫠

*no, not the PhD

AI refers to several different technologies and has only come to mean LLMs in popular use in like the last 4 years.

Not at all, always happy to try and communicate my understanding of how science works behind-the-scenes. Anyway as I stress, not my field - I work with caught specimens so I’m very aware that I’m evil lol.

Not an issue unique to tech and AI to be fair. I know someone who once got in trouble for saying they were sampling with an Otter Trawl (which, I have to stress, is not a trawl for catching otters…)

Tbf I do think it’s worthwhile for people working in these spheres to keep this kind of public image stuff in mind. Five years ago using language about AI to discuss this kind of work wouldn’t elicit much of an emotional response in either direction from non-experts.

(Hoping that Kakani or someone else will jump in if I’m getting this egregiously wrong! Again I’m only good at identifying dead animals in a jar in front of me lol)

It’s only the same thing as programs like iNaturalist and Merlin, but specifically trained to be good at your target ecosystem, be that seafloor or water column.

So again NAE but as I understand it the imagery is fed through algorithms to identify and count the animals (same done with ROV transect footage) - the point of things like FathomNet is to train these algorithms. Again I believe these are quite computationally cheap.

We’re not the bear though and it feels pretty rough to be lumped in with the bear because of branding