Campbell McLauchlan
@universal-soup.bsky.social
170 followers 110 following 26 posts
Postdoctoral research associate in the quantum theory group at the University of Sydney
Posts Media Videos Starter Packs
Reposted by Campbell McLauchlan
ccanonne.github.io
Official petition to the 🇦🇺 govt: make part-time PhD students' stipends tax-exempt!
www.aph.gov.au/e-petitions/...

Stipends are low but at least tax-exempt. Unless they go part time, then it becomes taxable. This disproportionately affects PhD students w/ care duties, health issues, young children..
e-petitions
e-petitions
www.aph.gov.au
Reposted by Campbell McLauchlan
eczoo.bsky.social
Peter Shor receives a small gift on the 30th anniversary of his QEC paper at #QEC2025.
universal-soup.bsky.social
It's been an amazing week at QEC25 seeing the incredible work people have done in the field over the past couple of years!
Abe Jacob did a fantastic job presenting our new work on trivariate tricycle codes, quantum LDPC codes that combine several nice fault-tolerant properties.
universal-soup.bsky.social
That's a wrap on the #Coogee25 Sydney Quantum Information Workshop! Thanks to all the brilliant speakers who came to talk about their research!
Reposted by Campbell McLauchlan
bartlettquantum.bsky.social
Now with a cool logo! #Coogee2025
Reposted by Campbell McLauchlan
universal-soup.bsky.social
Slaughterhouse 5 was incredibly influential for me when I first read it. Still one of the only novels I've re-read. It's an amazing book!
universal-soup.bsky.social
A useless but awesome fact from reddit: the range of a trebuchet does not depend on the planet on which you fire it. E.g. on the moon, the initial launch speed is smaller but the gravity pulling the stone to the ground is weaker, and the effects cancel out. (from this blog post: shorturl.at/8AqV8)
universal-soup.bsky.social
This is really cool though!
universal-soup.bsky.social
Couldn't there be highly-energetic, stable, complex systems in a universe without significant energy dissipation? With slow energy dissipation there would be slow energy exchange between systems which could maybe lead to stability?
universal-soup.bsky.social
With this year's big successes in quantum error-correction, is a Nobel prize somewhere down the track? (Or maybe one for quantum computing more broadly?) If so, who would be on the list of possible recipients?
universal-soup.bsky.social
I hear rumours about Brisbane 2032...
universal-soup.bsky.social
You're thinking of "poll vultures" which are news outlets that, around election time, report constantly on the polls to maintain uninterrupted coverage
universal-soup.bsky.social

3.9bl in 1986 to 93.4bl in 1991.

Although I think it's hard to agree with everything in the book, it is a fascinating description of a tumultuous and very relevant period of history.
universal-soup.bsky.social
None of them really wanted to be heading up a coup.
- Interestingly, you can see continuity between the Russian nationalist reaction to Ukraine's declaration of independence and Putin's claims at the start of the invasion.
- A simple demo of reform failing: the number of rubles printed grew from
universal-soup.bsky.social
This is an incredibly well-researched account of how the Soviet Union fell apart. The main argument is that the collapse was not inevitable and it was helped along tremendously by Gorbachev's reforms.
Some interesting tidbits:
- The August 91 coup plotters were a nervous shambles from the start.
universal-soup.bsky.social
Can't wait for your reaction to tomorrow's connections! (love from Australia)
universal-soup.bsky.social
Additionally, while comparing the X3Z3 code numerically to the CSS Floquet code and two honeycomb code variants, we found some interesting features about the performance of those previously existing codes under biased noise models.
universal-soup.bsky.social
The bad (but interesting) news: you seemingly can't do as well as bias-tailored surface codes (or color codes). We relate this to symmetries and back this up with no-go theorems for Floquet codes suggesting the X3Z3 code might be optimal.
universal-soup.bsky.social
The good news: just as in the surface code, you find better thresholds and sub-threshold scaling using the X3Z3 code compared to other Floquet codes. You can also find bias-preserving measurement circuits.
universal-soup.bsky.social
Okay obligatory paper meme: Recently Setiawan, at Riverlane, and I released a preprint on how to tailor Floquet codes to noise that is biased towards, say, Z errors. This "X3Z3 Floquet code" could improve performance on architectures using, e.g. the heavy-hex lattice.
www.arxiv.org/abs/2411.04974
universal-soup.bsky.social
0 should always be at the top! Change my mind. Also while we're at it, world maps should be oriented South-up.
universal-soup.bsky.social
The Philosophize this podcast put me onto this essay. Love it! The podcast and the essay.