iiinova — where your digital self keeps evolving

Philipp Bayer @philippbayer.bsky.social · 09/10/2026

They announced the literature Nobel and for once i have read a book of theirs before

020

Philipp Bayer @philippbayer.bsky.social · 26/09/2026

Booksale! Booksale!! Another success

000

Philipp Bayer @philippbayer.bsky.social · 24/09/2026

Was wondering how much UWA's executive layer had grown in the last 10 years and had an LLM try to piece together the numbers - from 4 pro-chancellors to 9 PVCs in 10 years. Sporting! At the same time, rankings went up, so UWA will be Nr. 1 uni in the world at about 18 PVCs....

010

Philipp Bayer @philippbayer.bsky.social · 23/09/2026

New Opus is out and immediately refuses to work on my repos, time to get whitelisted. Seems like security theatre

Warning from positron session sating that opus 5.5 can't work on this repo

010

Philipp Bayer @philippbayer.bsky.social · 20/09/2026

This is such a joy to read; after spending all week drowning in AI slop from left and right, just wonderful to read short, honest, funny, beautiful sentenced again

Cover: a journey round my skull, frigyes karinthy

010

Philipp Bayer @philippbayer.bsky.social · 28/07/2026

Weekend book haul at Bella Books while my son had class! All in on the dad-core

Six books including Orientalism and Smartest Guys in the Room

000

Philipp Bayer @philippbayer.bsky.social · 20/07/2026

join me we're doing decelerationism

Toad sat and did nothing. Frog sat with him

041

Philipp Bayer @philippbayer.bsky.social · 14/07/2026

kinda neat how you can see me switching employers/fields in my Github contributions

010

Philipp Bayer @philippbayer.bsky.social · 04/07/2026

Everybody come look at the hole

000

Philipp Bayer @philippbayer.bsky.social · 23/06/2026

Finally a name for my longevity startup

130

Philipp Bayer @philippbayer.bsky.social · 21/06/2026

Six months at least 30 minutes studying per day to get to level sixteen in Wanikani kanji learning - it goes up to more than 60 levels, haha

010

Philipp Bayer @philippbayer.bsky.social · 03/06/2026

Proud of the wanikani progress I'm making. About 40 minutes per day since the first of January - about 400 kanji and about 1000 words studied to a decent level. Only 1600 kanji to go to read on a high-school level.... (Burned means the word has been recalled often enough to be ingrained)

progress from wanikani table - 1447 entries on Enlightened level, 240 on Burned

020

Philipp Bayer @philippbayer.bsky.social · 25/05/2026

Very timely as I found the hidden Pratchett corner in our local library just yesterday :) so we start from some beginnings

080

Philipp Bayer @philippbayer.bsky.social · 22/05/2026

Tidal keeps on recommending me clearly AI-generated music :( that is not Hiromi Uehara, this is slop

000

Philipp Bayer @philippbayer.bsky.social · 14/05/2026

A fun side-effect of the AI-slop internet is that once in a while you find a paragraph so ridiculous, no human being could've written this

Start Small To Learn the Ropes

When first getting started in real estate, look for affordable properties to gain experience before scaling up. For example, Leonardo DiCaprio purchased his first home in the late 90s in L.A.’s Silver Lake for $769,500, according to UpNest. Arnold Schwarzenegger began by buying a small apartment building in the ’70s that gave him stable rental income early on. Beginning with lower-priced homes or multi-units builds knowledge so you can confidently advance to bigger deals.

081

Philipp Bayer @philippbayer.bsky.social · 13/05/2026

Big fan of the person who had the idea to list the missing system-level packages for easy copy-pasting when an R package installation fails. Will shout that person a beer, forever

sing PKG_LIBS=-lMagick++-6.Q16 -lMagickWand-6.Q16 -lMagickCore-6.Q16
--------------------------- [ANTICONF] --------------------------------
Configuration failed to find the Magick++ library. Try installing:
- deb: libmagick++-dev (Debian, Ubuntu)
- rpm: ImageMagick-c++-devel (Fedora, CentOS, RHEL)
- brew: imagemagick or imagemagick@6 (MacOS)
If Magick++ is already installed, check that 'pkg-config' is in your
PATH and PKG_CONFIG_PATH contains a Magick++.pc file. If pkg-config
is unavailable you can set INCLUDE_DIR and LIB_DIR manually via:
R CMD INSTALL --configure-vars='INCLUDE_DIR=... LIB_DIR=...'

040

Philipp Bayer @philippbayer.bsky.social · 25/04/2026

Found a super-chonky seahare today on the beach - look at the size of it!! (Also pictured, a shoe)

020

Philipp Bayer @philippbayer.bsky.social · 11/04/2026

I took the kids to the Beverley Air Show, and among all the loop-di-loops stunt plane flights, they had a helicopter drop a car from the sky. It was as awesome as that sounds like

010

Philipp Bayer @philippbayer.bsky.social · 03/04/2026

Great find at Elizabeth's in Fremantle, but yeah they're the priciest used book shop in Perth

Three books: guns of august, business essentials, cracking quantum physics

000

Philipp Bayer @philippbayer.bsky.social · 07/03/2026

Haven't fun isn't hard /// When you have a library card

010

Philipp Bayer @philippbayer.bsky.social · 04/03/2026

looks like TIDAL is getting more and more AI slop that TIDAL then auto-assigns to known artists :( it sounds like a hotel lobby

Screenshot from Tidal showing an album named SWEET SPRING

010

Philipp Bayer @philippbayer.bsky.social · 14/02/2026

The AI-generated code review is criticising the AI-generated pull request

130

Philipp Bayer @philippbayer.bsky.social · 13/02/2026

when the README on github looks like this you KNOW the pipeline hasn't been updated in >2 years

010

Philipp Bayer @philippbayer.bsky.social · 12/02/2026

the most annoying (of many) things of this AI/copilot integration in Positron is when the AI integration is faster than the file-system integration, so you get an AI-hallucinated file path (always wrong) far faster than the real file on your filesystem

140

Philipp Bayer @philippbayer.bsky.social · 10/02/2026

They know when I'm getting too relaxed. That's why they send these at 6:33pm

110

Philipp Bayer @philippbayer.bsky.social · 04/02/2026

Finally arriving at the good vocabulary

010

Philipp Bayer @philippbayer.bsky.social · 03/02/2026

I finally caught an Excel-doing-dates in the wild!! Opened an SRA metadata sheet in Excel where the experiments were named 6-1, 6-2, 6-3, 8-1, 10-4 etc. and Excel did its thing

Excel putting dates into library names like 6-1

150

Philipp Bayer @philippbayer.bsky.social · 03/02/2026

was banging my head against a bug (VEP classifying my in-exon variants as 'downstream') that Claude Sonnet 4.5 fixed in about 5 minutes. I can understand the scepticism on LLMs being jammed into every corner of life, but there's merit in these tools

he VEP command looks correct, but the issue is that VEP is annotating based on ALL transcripts/features in the region, and "downstream_gene_variant" is being picked as the consequence. Since you didn't specify --pick or --pick_allele, VEP returns consequences for multiple transcripts, and your code is just taking the first one.

Let me check how you're extracting the consequence from the VEP output:

Found it! You're extracting only the second field from the CSQ column with map_chr(2). This is getting the consequence from the first transcript annotation, which might be a pseudogene or non-canonical transcript.

Let me check what a VEP CSQ field actually contains:

Perfect! I can see the issue now. The VEP CSQ field contains multiple transcript annotations separated by commas. The first annotation is for TSC2 (a different gene!) and shows "downstream_gene_variant". But if you look at the second annotation (after the first comma), you'll see it's for PKD1 ENST00000262304 and shows "3_prime_UTR_variant" for exon 46 (which is correct - exon 46 contains the 3' UTR).

Your code is only extracting the consequence from the first transcript annotation (TSC2), not the PKD1 transcript!

The fix is to parse all transcript annotations and filter for the PKD1 canonical transcript (ENST00000262304). Let me create the fix:

PKD1_mutation_rate.qmd+24-1
Now let's test the fix by re-running the code to see if we get proper consequences for exons 40-46:

Perfect! The fix worked! Now you're seeing proper consequences for exons 40-46:

Missense variants (the most common)

141

Philipp Bayer @philippbayer.bsky.social · 31/01/2026

Went digging through the epstein files a bit and among the horrors it's always so frustrating how plain stupid these messages are. You found alzheimers in plants in Monsanto data?!? Sure, man. Sure

Weird text message chain saying they found alzheimers in plants in Monsanto data

130

Philipp Bayer @philippbayer.bsky.social · 30/01/2026

Good luck to anyone applying for a DECRA!!!!

screenshot from the research office email announcing that DECRAs are open

130

Philipp Bayer @philippbayer.bsky.social · 29/01/2026

this is a beautiful album and you're all missing out

030

Philipp Bayer @philippbayer.bsky.social · 29/01/2026

Wooo level 8 on wanikani (I've always refused to learn kanji and focused on vocabulary, but wow does it help to retain new Japaanese words when you know the underlying kanji meaning - 'well of course mortality means shibou. just look at the kanji's readings!')

000

Philipp Bayer @philippbayer.bsky.social · 24/01/2026

In a world of darkness, a bright light shines

6569

Philipp Bayer @philippbayer.bsky.social · 19/01/2026

It has finally happened: qs was removed from CRAN. I only realised because I tried to build my fancy-pants targets pipeline from start (the solution is to use qs2, but it seems like targets does not yet support qs2?) #rstats

152

Philipp Bayer @philippbayer.bsky.social · 10/01/2026

I was studying Kanji and was wondering whether every single Kanji has the reading kou/コウ?? So I made a figure. Turns out, yes, kou/コウ is the most common reading in the Joyo kanji! but not by much Pictured are how often the top 15 most common onyomi readings appear in the Joyo kanji.