Because my life wasn’t hard enough, I’ve decided to start writing a hydrodynamics solver in pure bash :ablobfoxhyper:

Upgraded RAM and added a fan, time for a beer and dead space :ablobfoxbongo:

Really need to set up ssh access to my home desktop but what a pain..

It took 16 months, but my paper was finally accepted 😭😪

worked in drawing my new vintage cast irons today and made some bacon tonight and by god was it good

~ 3x serial speed up after some code refactoring :blobcatrainbow:

Brandon :archlinux: :biflag: boosted

#JWST has a new press release with a beautiful comparison image between it and Spitzer showing just how much more we are getting from this new space telescope! You can see the interstellar gas with so much detail.
Full details at

Shameless self promotion! In my data visualization struggles, I often find myself searching "matplotlib colors" far too many time -- so I wrote a tool to display them in the terminal! Colorbars too :)

Brandon :archlinux: :biflag: boosted

Want to boost your programming skills? Dip your toes into LibreOffice development, and learn how to write a good commit message with your changes:

Finally get to spend some time learning for parallelism and really excited to implement this in my project

The mindset of open source software has been a game changer for my field, helping the collaborative nature of science, making for more robust code, and ultimately better science

Show thread

With so many different methods, approximations, initial conditions, etc, used, its so important for us to be able to truly compare and verify our codes.. but only in the last decade or so has it become at all common to replace scientific software as open source (speaking to my field)

Show thread

But only in a few cases are these software, producing (taxpayer funded) scientific results, open source. Most of the time we have to trust the authors and the few validations they have provided to verify that their software does what it says. And many times we have seen cases of "oh, up until now there was a bug in our code here. you still can't see the code. moving on..." What good is this attitude?

Show thread

(This is clearly an outdated slide, we're nearly to an exaflop machine)

Show thread

From a software standpoint, this requires, mostly, solving coupled, nonlinear partial differential equations for the fluid, radiation, gravity, and more. There's a lot of other bits that go in, and a lot of uncertainties in the various fundamental physics involved. Lots of people have software to do this, and lots of people have ideas about how things work.

Show thread

How do we know this? Telescopes tell us that supernovae exist in multitude, but cannot see into the core of the star to see what is happening under the hood. Enter software. I, among others, work on and use simulations to understand how these explosions originate, work, and how we can understand ones that we see in nature. It requires a vast amount of physics -- magneto-hydrodynamics, general relativity, neutrino radiation, and more, so only recently have we been able to do real 3D simulations

Show thread

The devil, again, lies in the details. The answer, as we currently understand it, is a combination of turbulent, convective fluid motions (hot plasma pushing against the stalled shock) and neutrinos heating the matter below the shock, ultimately (in some cases) leading to the successful revival of the shock, and the complete disruption of the outer layers of the star. What is left is a neutron star, or perhaps a black hole.

Show thread
Show older

Fosstodon is an English speaking Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.