Tom Young

I was moderately worried about the future of AGI. youtu.be/VcVfceTsD0A did nothing to help having spent all of 5 minutes thinking about it: the AI alignment problem seems too difficult. how do you build a system with empathy?

Open sourced an hour ago and there’s already some spicy PRs 🌶

@inigo_iribarren @jla_gardner what did you do... 😉

@TYoungResearch I'm afraid that now I can't even see the button to select the molecule 🥴

Writing for web with (almost) no javascript is awesome. Molecular visualisation in rust 🦀: t-young31.github.io/mol-wasm/ (from a .xyz file), UX needs some work to say the least.. code: github.com/t-young31/mol-…

@inigo_iribarren have maybe fixed the windows line endings. any better now?

@TYoungResearch it works on my linux laptop (fireforx) f***** windows TT_TT

@jla_gardner truly the best kind of bug report. feel free to hit merge!

@TYoungResearch done, and with a PR to boot 😊

@jla_gardner shocked that there's not anything more catastrophic! would you mind adding them as issues on GitHub? thanks 😄

@TYoungResearch Two quick bugs I can see are that 1️⃣ the whole web page scrolls (at least in Safari) at the same time as the structure zooms, and that 2️⃣ pressing escape makes the structure go invisible.

@inigo_iribarren it could be weird windows line endings maybe?

@TYoungResearch Ohhh ok ok... That looks really cool anyway ☺️ sure it's an easy fix

@inigo_iribarren yeah 😞

@TYoungResearch IDK, I tried firefox too and same thing, I can select the file but I don't see anything. Should it appear automatically?

@inigo_iribarren sad times. was working on Chrome and Safari for me. wonder if it’s a wasm compatibility thing or (more likely) my fault

@TYoungResearch yes! I can select the file now. but nothing happens after

@inigo_iribarren 😢 can you scroll down to the upload box?

@TYoungResearch it might me my computer but ... :S

watching a JWST doc and _damn_ is it impressive. shocking that it worked. go humans

@lewischewis I’ve found Cython to be super intuitive for making python bindings to C++ code: cython.readthedocs.io/en/latest/src/… working example: github.com/duartegroup/au…

Does anyone know of a simple beginner-level guide to compiling a C++ function for importing by python? Preferably one that can take a numpy array, edit it, and return it? TY

I *will* have introduced bugs so please do create an issue (at github.com/duartegroup/au…) when you find them – thanks

Just as we hit 100k downloads (😲) autodE v1.3.0 is out 🚀 anaconda.org/conda-forge/au… Lots of improvements and a few bugfixes – thanks to all involved! Highlights include: a new TS search method, better bond definitions, frequency scaling and NWChem defaulting to autodE optimisers

@NatalieFey_NLS hope you're feeling okay 😟

Oh bugger!

@alexjbinnie congrats @alexjbinnie 🎉

After three and a half years, it is finally done! 🎉

@ruben_laplaza @kjelljorner @janhjensen adjacency determination ≡ defining what is bond(?) definitely environment specific radii would be awesome!

@TYoungResearch @kjelljorner @janhjensen Isn't that more of a covalent radii / adjacency determination issue? I always thought environment-specific radii would be a banger (for this and other purposes)

Without wanting to start a 🔥 war: what defines a bond? - Distance? Electron density? Dissociation energy?

@corminboeuf_lab many thanks for the reference. I don't suppose there is a tldr; for what aligns best with a chemists perception of a bond?

@TYoungResearch Perhaps a useful read on the topic (amongst many other insighful perspective/papers). pubs.rsc.org/en/content/art…

@ruben_laplaza @kjelljorner @janhjensen That's exactly the approach taken in autodE (twitter.com/TYoungResearch…) too, but isn't ideal for edge cases like a very long N-N "bonds"..!

@TYoungResearch @kjelljorner If that is the world you want, then the safest option is satisfying valence rules! This is akin to the approach taken by xyz2mol by @janhjensen in order to build graphs from 3D coordinates.