iiinova — where your digital self keeps evolving

Brandt Gaches @brandt-gaches.space · 29/09/2026

I would guess they're getting submissions now with people reducing ALL of the section/subsection headings to have no padding at all and maximally fill out the text...

020

Brandt Gaches @brandt-gaches.space · 29/09/2026

It doesn't help the TeX template that is used likes to add asymmetric white spaces all over the place. The difference in the white space above and below section headings makes my eyes twitch...

240

Brandt Gaches @brandt-gaches.space · 28/09/2026

I'm happy to announce that I will be starting as an associate editor for Astronomy & Astrophysics (@aanda-journal.bsky.social )! I will primarily be covering the fields of the interstellar medium and circumstellar medium. I'm excited to start this new role in science!! www.aanda.org/about-aa/edi...

aanda.org

Astronomy & Astrophysics (A&A)

Astronomy & Astrophysics (A&A) is an international journal which publishes papers on all aspects of astronomy and astrophysics

020

Brandt Gaches @brandt-gaches.space · 25/09/2026

Sleepytime is probably the best usage of classical music in a kids show I've ever seen.

010

Brandt Gaches @brandt-gaches.space · 25/09/2026

The package is still in a bit of development, but everything looks like it is working. Rewriting a bunch of these things from ugly Python 2 "get it working" monolithic behemoth scripts into modern code was definitely a "learn as you go" moment - especially getting proper documentation in it..

000

Brandt Gaches @brandt-gaches.space · 25/09/2026

I decided to finally modernize old heritage Python scripts I have into one package for protostar clusters: OcotilloPMF! Ocotillo allows the user to both sample masses from a protostellar mass function and generate synthetic clusters spatially. github.com/AstroBrandt/...

github.com

GitHub - AstroBrandt/OcotilloPMF: Python package for sampling steady-state clusters of protostars.

Python package for sampling steady-state clusters of protostars. - AstroBrandt/OcotilloPMF

110

Brandt Gaches @brandt-gaches.space · 18/09/2026

*Job hire announcement*: I'm hiring a postdoc in my group! The postdoc will work on a project to use molecular dynamics to build the first atomistic view of the chemistry of icy grains following heavy-ion impact. Ad: www.uni-due.de/karriere/ste... Deadline: Nov 15, 2026 Start: Mar 1, 2027

uni-due.de

Stellenausschreibung

033

Brandt Gaches @brandt-gaches.space · 04/09/2026

I imagine that first one was by someone typing into Claude Code "Make me a shitty reskin of Flappy Bird, themed for American patriotism with an Eagle and Roman-esque pillars", and they just deployed whatever came out.

010

Brandt Gaches @brandt-gaches.space · 04/09/2026

Also, the universities that kowtow to the Trump administration always inevitably see it come back to haunt them anyway. Placating the administration does nothing but signal that they can keep pushing and pushing and doing whatever they want.

000

Brandt Gaches @brandt-gaches.space · 04/09/2026

Maybe this is the writing on the wall that university- or state-hosted GitLab or Bitbucket instances may be the best for academic codebases and data repositories. I am guessing GitHub will rely more and more on AI scraping to make decisions, removing any communication and support more and more.

000

Brandt Gaches @brandt-gaches.space · 04/09/2026

Two weeks now since my @github.com account was randomly suspended without any information (repositories were just astrophysics research code bases). Still no communication or anything from them. The plan was to host my database on GitHub for long term stability - not sure if that is ethical now...

110

Brandt Gaches @brandt-gaches.space · 02/09/2026

And the title actually says pretty clearly what you'll find in the paper!

010

Brandt Gaches @brandt-gaches.space · 02/09/2026

Finally finished the Dungeon Crawler Carl collection! Took some time to get the English hard cover first book here. And then found out that there are two different English editions sold here: one from US and one from UK that are slightly different sizes and different cover styles/textures...

Book set of the Dungeon Crawler Carl series on a shelf

010

Brandt Gaches @brandt-gaches.space · 26/08/2026

Still one of the best movies of my childhood. Certainly helped that it seemed like Tim Curry just viewed himself as another muppet.

static.klipy.com

Tim Curry as Long John Silver, Professional Pirate

ALT: Tim Curry as Long John Silver, Professional Pirate

031

Brandt Gaches @brandt-gaches.space · 25/08/2026

It would seem that instead of using open-source software and databases, it is more long-term stable to use local gitlab instances or similar (I swapped the database for now to a mirror on the NRW Git). Overall, super disappointed in the lack of any support or communication!

001

Brandt Gaches @brandt-gaches.space · 25/08/2026

This experience really has just made me question whether @github.com can be used at all anymore reliably for open-source software or databases long-term. Without any communication or effective support, suspending accounts based on - most likely - automated detection scripts makes it unstable. 3/

101

Brandt Gaches @brandt-gaches.space · 25/08/2026

the audacity to send an email today asking for me to share my feedback on the recently closed ticket. Again - no communication or reason as to why. This also impacted my public cross-section database, which was down until I was able to set up a mirror using a GitLab instance. 2/

101

Brandt Gaches @brandt-gaches.space · 25/08/2026

Last Friday, I tried to open a @github.com page I made for a conference of a 3D volume rendering, only to find out my account was randomly and without notice suspended. (I suspect a bot auto-flagged the plotly file due to the timing?) After filing tickets to try to get communication, they had 1/

100

Brandt Gaches @brandt-gaches.space · 14/08/2026

The data vis explorer may take a bit of time to load, since the site is ~94 MB. It loads a number of 64^3 (downscaled from the model calculations) so that different molecules can be toggled between to see where they exist in the PDR front! It can also toggle the C- and H- fronts on and off, too :)

000

Brandt Gaches @brandt-gaches.space · 14/08/2026

#Astronomy /Astrochemistry paper time! We built the first 3D Orion Bar-like photo-dissociation region model, resolving the relevant transitions, including full H2 level populations: arxiv.org/abs/2608.04116 I also made a fun data vis explorer for it: astrobrandt.github.io/3DPDROrionBa... #astrosci

arxiv.org

Resolving dense photo-dissociation regions: the structure of photochemical fronts in three-dimensional gas distributions

For decades, the Orion Bar has been the prototypical photo-dissociation region. Viewed nearly edge-on, it offers a unique window into the stratified chemical structure of the atomic-to-molecular trans...

131

Brandt Gaches @brandt-gaches.space · 01/08/2026

Heading to COSPAR next week in Florence, and the weather does...not excite me. Nearly 40C every day! Hopefully the venue and my hotel have working AC...

000

Brandt Gaches @brandt-gaches.space · 17/07/2026

The worst part about changing the morning alarm sound to Sam Reich singing the "Count the Rice" song from this past week's Game Changer on @dropout.tv is having to wait until the morning for the reaction. The best part is the "Really?!" after the realization hits first thing in the morning.

010

Reposted by Brandt Gaches

National Radio Astronomy Observatory - NSF NRAO @thenrao.bsky.social · 01/07/2026

News You May Have Missed ⤵️ 3I/ATLAS Contains 30X More Semi-Heavy Water Than Comets In Our Solar System ALMA reveals interstellar comet’s home star system formed under far colder conditions than our own #RadioAstronomy #Astronomy #3iatlas #ALMA

public.nrao.edu

3I/ATLAS Contains 30X More Semi-Heavy Water Than Comets In Our Solar System - National Radio Astronomy Observatory

The latest news from the NSF's National Radio Astronomy Observatory.

02310

Brandt Gaches @brandt-gaches.space · 01/07/2026

Dropout's Game changer this season has been the best type of unhinged!

010

Brandt Gaches @brandt-gaches.space · 01/07/2026

senior faculty and researchers. The protections against discrimination against disability are a great start, but also feature noticeable and easy loopholes and a lack of oversight. I also hope that during this month, my fellow astronomers with disabilities will tell some of their stories!

000

Brandt Gaches @brandt-gaches.space · 01/07/2026

There is a significant and stark gap in tangible support systems for astronomers with disabilities, both socially and financially. It took until I had a more stable situation to start to feel comfortable talking about my experiences: I have (and still continue) to receive negative comments from

100

Brandt Gaches @brandt-gaches.space · 01/07/2026

So, I hope during this disability month, especially since it is during part of the conference season, my fellow #astronomers will take some time and ponder about how their recent conferences would be experienced by those with a wide variety of disabilities.

100

Brandt Gaches @brandt-gaches.space · 01/07/2026

yet to go to a conference where there were planned alternative socials that were accessible to those with physical handicaps. I've given talks at conferences where they had stages with only stairs. After most conferences, I need several days to recover and get back to normal pain levels.

100

Brandt Gaches @brandt-gaches.space · 01/07/2026

looking at posters and sitting back down in a different room where the talks are held. It has meant constantly having to choose between doing the social excursion - which often involves significant walking (even conferences that have nature hikes as the official social!) - or staying back. I have

100

Brandt Gaches @brandt-gaches.space · 01/07/2026

Far too many conferences I have been to have seemingly not had any thought or regard for those with disabilities. As someone with a physical disability, I see this in the stark lack of seating outside the rooms where talks are given. At coffee breaks, I may have to choose between being social or

100

Brandt Gaches @brandt-gaches.space · 01/07/2026

July is #disability Pride Month worldwide. As I progress more into my academic career, I notice time and time again that many institutions usually will only do what is minimally legally required for accessibility, and minimally regard disability in terms of diversity. 1/

The disability pride flag. A charcoal grey flag bisected diagonally from the top left corner to the lower right right corner by five parallel stripes in red, pale gold, pale grey, light blue, and green

153

Brandt Gaches @brandt-gaches.space · 01/07/2026

Also, the significant irony of him saying his critics merely lack imagination, despite the fact that claiming aliens with no actual evidence because you don't understand something is about the least scientifically imaginative thing you can do.

050

Brandt Gaches @brandt-gaches.space · 01/07/2026

Well I guess he finally found someone who would listen to his pseudo-science. Don't worry, I'm totally sure he'll suddenly take a serious science-first approach to things...

150

Brandt Gaches @brandt-gaches.space · 30/06/2026

I haven't watched a great many games, but the flopping and acting in this game is almost beyond insane at this point! One player tapped another's back and he threw himself on the ground! Another had his foot grazed, and after the next step did nearly a flip! Starting to be almost a parody...

000

Brandt Gaches @brandt-gaches.space · 29/06/2026

And the potentially winning goal reversed after review! This game may go on forever...

100

Brandt Gaches @brandt-gaches.space · 29/06/2026

Have to hand it to Paraguay: as much as Germany is pushing heavily offensively, they've held up a pretty good defense!

100

Brandt Gaches @brandt-gaches.space · 24/06/2026

BlueSky #coffee people: a recent fire that impacted my flat caused too much smoke damage to my Timemore C2 grinder! So, I have to look for a replacement. Does anyone have any good recommendations for solid manual (or electric) grinders for filter strength? (There is, of course, just the C3...)

000

Brandt Gaches @brandt-gaches.space · 16/06/2026

We had a car fire outside our house, impacting our flat this morning. I learned through this that a Mercedes hearse burns hot enough to melt/incinerate an aluminum bike, but not melt steel. The picture of our garage next to the car shows this - two ash piles from bikes in the metal garage!

Photo showing a garage destroyed by a fire, with two piles corresponding to bikes that were totally destroyed, and a metal frame of a cart.

000

Brandt Gaches @brandt-gaches.space · 12/06/2026

I think this is a great decision. Tech companies have largely tried to excuse bad behaviors from their systems by saying, effectively, "it's just the algorithms." They have wanted all the benefits of content generation and curation, but none of the responsibilities born by similar entitites!

020

Brandt Gaches @brandt-gaches.space · 09/06/2026

They will unironically name it something dumb like "Maxwell's Demon cooling system"

030

Brandt Gaches @brandt-gaches.space · 31/05/2026

By coincidence, 1 parmin is about the distance light would travel during the requested time to return a referee report (light-anxiety-time), but sadly, this time is far from a constant, so it could only be used as a fuzzy unit at best...

000

Brandt Gaches @brandt-gaches.space · 31/05/2026

Astronomy has a number of cursed units and conventions. While on my parental leave time, the lack of good sleep time has brought me to these new cursed units: 1 parmilliarcsec (pmc) = 1 kpc 1 parmin (pn) ~ 3438 AU 1 pardeg (pg) ~ 57 AU 1 parrad (pd) ~ 0.6421 AU Maybe this is why one needs sleep?

100

Reposted by Brandt Gaches

Jonathan O'Callaghan @astrojonny.bsky.social · 28/05/2026

A nice paper by @asborlaff.bsky.social showing that launching telescopes to space doesn't necessarily solve the light pollution issue from satellites. These are satellite trails seen by NASA's SPHEREx space telescope, which appear in 73% of all its images. Eek. arxiv.org/abs/2605.27501

Satellite trails seen by NASA SPHEREx telescope

07240

Brandt Gaches @brandt-gaches.space · 26/05/2026

If you want a combo of free hosting and cheap domain, I would highly recommend GitHub page + Porkbun (porkbun.com) for the domain. I use it for my database site. Porkbun offers an direct way to connect to the GitHub page, security stuff and still quite cheap...

porkbun.com

020

Reposted by Brandt Gaches

Linn Boldt-Christmas @nplinnspace.bsky.social · 07/05/2026

🔭 A cautionary tale for undergrads (and above!) using AI for #astro plots: our BSc student wanted to make a "quick" HR diagram showcasing M-dwarfs & thought some chatbot could make one and... well... ...extra credits are now offered to students who can correctly find all errors on cursedHR.png 🫠 🧪

A pathetic attempt by AI sloppery to make a HR diagram, generated by one of our BSc student, and shared here with their permission. Honestly, I can't even begin to explain all the deviations. Imagine a normal HR diagram but one that appears to you in a fever dream. It gets so much worse the longer you look at it. WHY IS THIS ALLOWED TO EXIST AAAAA

4827580

Brandt Gaches @brandt-gaches.space · 07/05/2026

It makes me quite depressed when I see a paper whose abstract looks interesting and I load up the PDF and the introduction was clearly written by an AI model . Complete with garbage or weird citations and all! Especially when I see that even the methods section seems to be written by an AI model...

010

Brandt Gaches @brandt-gaches.space · 07/05/2026

This week I had another horrifying experience with the SOCIALIST healthcare system in Germany! I needed new crutches for long term use. It took me 6 days to get my appointment (sure 2 were the weekend and 1 was a holiday, LAZY!!). They had the audacity to charge 5€ for them! For SHAME! (!sarcasm!)

010

Brandt Gaches @brandt-gaches.space · 01/05/2026

It took me way longer than I care to admit, but I figured out how to get the journal style of DOI links in bibliographies in #typst. A combination of weird, janky hacks into the csl and regex matching in the template, but it seems to work! (The templates will be public if/when I get the ok!)

Example bibliography section of a Typst template showing links for the journal section of the reference that point to the DOI if it exists

010