Tom Saunders

Would love to see researchers properly incentivised and rewarded by funders/institutions to publish fewer papers. One of the many interventions needed to untable the web of problems with academic publishing.

If you're a researcher in NZ please do our very short survey on the Vision Mātauranga policy! We are keen to hear your opinions. DM for the survey link

YouTube video by Team Candiru Beautiful wasp zombifies cockroach

Great article! One of my favourite non-Aotearoa examples is Ampulex compressa (emerald cockroach wasp). Pretty amazing: www.youtube.com/watch?v=-ySw...

You've run workshops for media orgs before, right? Why don't these points about CC in particular make it into the handbooks or documentation used by the journos?

South Island Jurassic Park? Peter Jackson backs project to bring back moa Groundbreaking de-extinction project already under way, with a short timeframe revealed.

This is absolute shameful bullshit from all parties concerned. Peter Jackson is being scammed, Colossal are total frauds (5–8 years? Give me a break), and I have no idea what Ngāi Tahu think they’re doing. There are clear and simple reasons why this is impossible. www.nzherald.co.nz/nz/south-isl...

Work with us! Kaimanaaki Pāpāho | Media Advisor – Science Media Centre The Science Media Centre is looking for a media advisor to join our dedicated team working to improve the reach, relevance and accuracy of news media reporting on science and related topics in…

THE SMC IS HIRING!

We're looking for someone to join our dedicated team working to improve the reach, relevance and accuracy of news media reporting on science and related topics in Aotearoa New Zealand.

More info and apply here: www.sciencemediacentre.co.nz/2025/06/26/w...

FYI iNaturalist have added an update to the post, including: "We are not giving Google special access to your iNat data, and we have no obligation to use Google’s infrastructure as part of this grant. Google is providing funding and advice on how to potentially leverage AI."

🗺️ Ever wander to a new place and think, "Hmm I wonder what birds/plants/[insert favourite taxonomic group] I can find here?" 🔍

We built {infinitylists} 📋 to help nature lovers create personalised, location-based taxon lists! 🌏

Yes - retraction should be encouraged when issues surface, not punished. Surely one of the biggest bang-for-buck moves would be for institutions/funders to lower the # of pubs they judge/evaluate/reward researchers on each year. Many of these integrity issues are symptoms of demanding quantity.

Most concerning to me is the attempt by the VC to frame it as a win for 'academic freedom', even though University lawyers repeatedly argued that academic freedom was a 'privelege' and that Wiles should stop talking about her expertise.

Nice one, congrats on the new role Fonti!

Some great examples here of how researchers can partner with Māori to co-develop research projects in ways that give effect to CARE principles alongside FAIR.

I couldn't have done it without the amazing support and guidance from my supervisors Greg Holwell & Gonzalo Avila, and our collaborators Kye Chung Park & Lee-Anne Manning. I'm truely grateful to have had the opportunity to work with and learn from all of them.

...the moment of attack means we're examining ecological host range (the hosts that are actually attacked in the field, as opposed to physiological host range, or the hosts that support development of the agent, often tested in a lab setting).

...and this is where chemistry becomes important (for insects in particular). The stink bug egg parasitoids in this work use a variety of cues to hone in on their targets, and these cues become more and more reliable as they get closer. Testing these cues, while moving further out from ...

This is important because regulators often rely solely on the emergence work, where agents are confined with potential hosts in tight quarters to see if they *can* develop in a non-target host. This is crucial data, but a more accurate assessment of risk requires moving from *can* to *will* ...

In our case study, our two parasitoid wasps emerged from the two hosts at similar rates, but showed markedly different motivations to search for each host in arrestment experiments.

In particular, arrestment bioassays in open arenas are relatively simple/inexpesnive but provide much-needed behavioural context. A substrate is contaminated with potential hosts and how long the agent searches is a proxy for motivation to find that host. These times can then be compared.

A close-up image of a small black parasitoid wasp backed onto a stink bug egg, which is the same size as the wasp. She is laying an egg into the stink bug egg through a tube on her abdomen.

My final PhD chapter is now published #openaccess in the Journal of Chemical Ecology: doi.org/10.1007/s108...

Main idea: Chemical-ecological methods can complement oviposition work in host-specificity testing of classical #biocontrol agents by teasing details related to ecological host range. 1/

Two academic (lecturer) positions available at the University of Otago in Aotearoa/New Zealand! One in Conservation Biology, another in Evolutionary Ecology. Links for positions below:
otago.taleo.net/careersectio...
otago.taleo.net/careersectio...

Where do Walker and Lee propose people live, if not on K Rd close to new train stations and cycleways? Does this mean the development on the gravel pit can finally go ahead?

Yeah definitely, but you've got a theory of programming in your head already - when I said beginners I meant people new to programming entirely. User needs to understand certain concepts in order to generate code they can then evaluate and troubleshoot, if necessary.

100% - Rather than generating code, IMO the main use for beginners is asking conceptual questions, eg "I want to do X task in Y language - what packages/functions should I be looking at" or "how would a programmer approach X kind of problem".

LLMs can't stop making up software dependencies and sabotaging everything : Hallucinated package names fuel 'slopsquatting'

Bad actors are creating packages with names that mimic hallucinated packages in code generated by coding assistants. When the code runs it installs the malware-riddled packages. Another reason why learners should not use coding assistants to generate code (there are other good uses for learners).

Wow this is excellent, thanks so much for creating this resource.

@brieferyet.bsky.social
Always lots of interesting commentary. And even if I don't always agree with their conclusions, I always find their arguments thoughtful and well reasoned.

Wikipedia Faces Flood of AI Bots That Are Eating Bandwidth, Raising Costs 'Our content is free, our infrastructure is not,' the nonprofit that runs Wikipedia says, warning that AI scrapers are wasting the site's bandwidth.

The rise of AI bots scraping websites, mostly for openly licenced images to feed generative AI models, is increasing infrastructure costs for sites like Wikipedia - this will become an issue for OA publishers too. Rate-limiting and enforced authentication might need to be imposed.

It's a shame Colossal spoiled it by gunning for hype instead of leaning in to the more grounded version. Could've been a nice educational moment about extinction, conservation, and gene-editing. But now the focus is on the exaggeration which taints the achievement.

It might be a cat with 8 inch fangs, and it might in some ways resemble a sabre tooth, but it ain't a sabre tooth tiger (Smilodon spp.). Colossal's argument is they have indeed ressurrected an extinct species, which is laughable in this case.

"If they look like this animal then they are this animal" is definitely not how the morphological species concept works.

Genetically altering grey wolves to make them look a bit like dire wolves does not suddenly make them dire wolves.