lazyslurm is a tui for managing slurm jobs on HPC clusters. It's kind of like lazygit or lazydocker. I've been chipping away at it for the last few months. Its built in rust w/ ratatui.

https://github.com/hill/lazyslurm

- https://banksia.bio: I suspect there is a market for private consumer whole genome sequencing services. Think "Mullvad vpn" of sequencing - I shouldn't have to know the identity of the person I am sequencing, and they can be identified with a client number not tied to their PII.

- lazyslurm: A TUI tool for managing/viewing slurm / HPC setups. Similar to lazygit or lazydocker (https://github.com/hill/lazyslurm)

https://banksia.bio - I think private personal genome sequencing is an underserved market. I'm trying to start a "Mullvad VPN"-esque service where I as a provider do not need your personal information to perform the sequencing. Very much a WIP at the moment :)

---

https://github.com/hill/lazyslurm - a terminal ui for SLURM jobs (like lazygit but for slurm!)

---

https://github.com/hill/pg_tamagotchi - a tamagotchi that lives in your postgres instance (mainly to learn about pg extensions)

Unexpected Application Error! Unexpected token '<', "<!DOCTYPE "... is not valid JSON SyntaxError: Unexpected token '<', "<!DOCTYPE "... is not valid JSON

thanks kaoussi, where did you see this error? just loading the page?

Saw it too on first load of banksia. Loaded again and it went away. Chrome, Mac.

thankyou both, i've just fixed it :)

Agreed that it's an underserved market (personal genome sequencing).

Do you own or have access to a sequencing platform / machine? Or do you just buy and prepare kits, and then send the prepared sample out for sequencing?

Wonderful idea

thanks!

I'm building a lazygit / lazydocker like TUI for slurm jobs on HPC.

It's called lazyslurm - https://github.com/hill/lazyslurm

Would love feedback! <3

Where does the slurm come from?

Lazyslurm: https://github.com/hill/lazyslurm - a terminal ui for managing and visualizing slurm jobs. Heavily inspired by lazygit and lazydocker.

Very early into this - would love feedback!

Richard Rhodes' introduction to Serber's "The Los Alamos Primer" suggests 24:

> Despite this leavening of older men (Oppenheimer was thirty-eight), the group's average age was only twenty-four.

Recipe Cleaning - A little tool I've been working on to extract the actual recipe from SEO-bloated recipe content: https://recipe.cleaning/<YOUR_RECIPE_URL_HERE>

It uses the microdata and json-ld documents embedded in the page to find the content you actually want to see. No need for AI/LLM extraction.

xkcd for the xkcd: https://xkcd.com/1393/

The first XKCD was published closer to the release of the original Doom than to the present day.

That's not even close: the first XKCD was published closer to Intel's release of the 386SX than today.

18 years before XKCD, a 386-based PC capable of running Doom would have set you back $6,000–$10,000.

Imagine a game released today that has a 64-core CPU and a 4090 as minimum requirements, and that's roughly what Doom's system requirements would have looked like in 1987.

That was the entry price of legit Compaq Deskpro 386 on first release.

Within six months hobbyist computer enthusiasts were paying half that (or less, IIRC) for PC clones direct from asia.

Source: That's about what I paid for a Doom capable PC at the time.

A pair of second-generation T800 transputer chips and a 4-chip array board to fix them to cost way more at about the same time.

Source: https://twitter.com/sincethestudy/status/1761099508853944383

At my previous employer, our (very small < 10 dev) team was on a rotation with 2 devs per week. All devs were required to be on call (which I think was the correct decision - makes us ship better code because we don't want to be supporting its failure!)

Our head of engineering gave us 1 day time off in lieu (TOIL) per week of on-call support. I was pretty happy with that as comp.