John Trant

@trantteam.bsky.social

Assoc Prof UWindsor, Faculty of Science Research Chair; married to @shufflersunite.bsky.social he/him/you bastard Probably writing a grant. Big Ottawa Senators Fan. www.trantteam.ca AND www.binarystarchem.ca. Total scientific saturation at an epochal pace

Binary Star Research Services is proud to announce that we plan on doing YOUR comp chem on Europa and Io! We will power our servers using Io's INEXHAUSTABLE INTERNAL FIRES (TM) and cool them using Europa's INFINITE ICE (TM). This will cut our power costs by a factor of INFINITY. Details to follow.

Writing a book is hard. It’s so fucking hard. Just this week I’ve reached out to so many people for help to get this book over the finish line. And so many people have helped me. Yesterday one of my professors from undergrad hopped on a zoom call to talk through Raman spectroscopy.

Okay genuinely asking— what does accountability look like for Holden Thorp and Hank Green? What I’d like to see: I’d love if @holdenthorp.bsky.social set a clear journalistic standard at SCIENCE and prohibited the use of LLMs for researching, writing, and editing publications at SCIENCE.

Debbie Gale Mitchell@heydebigale.bsky.social · 4d ago

So in the last two weeks EIC of Science and a famous science communicator have lost massive amounts of credibility because of their reliance on LLMs... Some examples of how to lose trust with your audience to share with my students I guess.

Civic Holiday is one of the happiest holidays of the year in (most of) Canada. In Ontario, it is Simcoe Day (no one calls it that). The daycare is open, but work is closed. For parents with no family...this is incredible.

Look, it handles my invoices! Its ok if it only right 98% of the time! I am sure the tax man will understand that it isnt tax fraud if I didn't mean it. Like when I agreed that all data was true and accurate and that I am responsible for it, they didnt mean I was responsible for the AI stuff. Right?

The Economist@economist.com · 3d ago

An economy powered by individual entrepreneurship, with AI handling tedious back-end tasks, could prove fulfilling and lucrative for many people. Register for free to learn why

Some comments on this are WILD (about Wild!). "The AI tools are good and mostly right, so you should use them" is nuts. If, like Alex you are writing science, mostly right is the same as completely fucking useless. If mostly right is good enough...you are not doing something important.

Alex Wild@alexwild.bsky.social · 4d ago

The short version is: after a month the AI has not given me any good reasons to trust it, and I am honestly freaked out by the confidence with which it weaves completely wrong facts into prose, and the confidence with which it writes and professionally formats entire sections of bullshit.

I apologize for interrupting your regular consumption of shitposts and angry retweets from this supposedly serious scientific account of a VERY SERIOUS SCIENTIST to post something important. A former MSc grad from my lab, Dr. Xiao Yu from my group is running a network for Asian women in STEM.

“That could be me" event RSVP

That Could Be Me is a virtual conversation series that brings together Asian women in STEM across different career stages through meaningful 1:1 conversations. Whether you're a student exploring what'...

docs.google.com

I've gotten into watching Super League rugby while exercising. Realizing that it is like a bunch of teams from right around Leeds and then two from Southern France beating up on each other. From what I've heard about Leeds...this is consistent. But I expect better from the French.

Extremely niche post alert: Anyone else REALLY careful with their mouse use when approving submissions on "Editorial Manager"? So easy to accidentally click "edit" instead of "view" or "approve" and have to go through a random 15 minute waiting period again.

If you fear that state-run grocery stores paying living wages to all employees will undercut the pricing of private sector grocery stores paying minimum wage and drive them out of business...it is not the state-run store's fault. It's the profit-seeking private sector model's gross incompetence.

The Economist@economist.com · 2w ago

Discontent about rising food prices is inviting all kinds of intervention from politicians. Yet most of these measures are looking in the wrong place

I tell them how to do well in the class. Essentially, work the problems and don't fall behind because it all builds on itself and you will not do well if you try to cram it all in at the last moment. It's process, not content. I reiterate that several times. I'm all out of ideas beyond that.

Dana Menard@shufflersunite.bsky.social · 2w ago

Watching undergrads panic in the uWindsor subreddit because they've got @trantteam.bsky.social for organic chemistry is the gift that keeps giving.

A PhD thesis can be written in as little as 8 weeks! If I give the student all the data, and think the only thing that matters is a written document. A PhD is not about the thesis. It's about the skill development. And no, that can't be done in 18 months.

Times Higher Education (THE)@timeshighered.bsky.social · 2w ago

A science PhD can be completed within 18 months: supervisors should involve doctoral students with conducting, writing up and publishing experiments straight away, says Adrian Furnham #PhDs #research #supervision #academicsky https://ow.ly/5Fxb50ZpZ4O