Tony 😷 💻 Greenberg @BayesicTony@twit.social
- https://twitter.com/BayesicTony
a geneticist and a Bayesian; freelance scientist
workspace: Arch Linux/dwm/neovim
program in: C++/R/shell/perl
hobbies: BJJ/languages/anime/cooking
Introducing Club Twit! Come on in, the water's fine! https://twit.tv/shows/this-week-in-tech/episodes/819
A very good explanation on how to interpret at-home COVID tests, featuring the Reverend Bayes:
Here is a good explanation why pre-IVF embryo selection is at best useless:
https://liorpachter.wordpress.com/2021/04/12/the-amoral-nonsense-of-orchids-embryo-selection/
The disturbing possibility, though, is that it will sell the rich a "certificate of genetic superiority" for their kids. May not even matter that there is no real basis for it. If these kids get opportunities others don't, will the procedure be considered to have "worked?"
This really pisses me off. These people are peddling eugenics-flavored snake oil. Don't waste your money on this crap.
A really cool hands-on epidemic simulator to learn about herd immunity:
https://infectiousmatter.com/herdimmunity/
Just finished listening to last Sunday's TWiT. The panel talked a bit about personal genomics. @leo mentioned that your decision to share your data affects your children's privacy. Due to how human populations evolved, it's actually much broader than that. It is sometimes possible to de-anonymize a whole genealogy data base by submitting some well-chosen fake profiles (e.g., https://elifesciences.org/articles/51810)
The difference between apples and oranges is statistically significant
Buffer overruns, license violations, and bad code: FreeBSD 13’s close call | Ars Technica - https://arstechnica.com/gadgets/2021/03/buffer-overruns-license-violations-and-bad-code-freebsd-13s-close-call/
Despite not having any kernel developers on-staff, Ars was able to verify at least some of Donenfeld's claims directly, quickly, and without external assistance. For instance, finding a validation function which simply returned true—and printf statements buried deep in cryptographic loops—required nothing more complicated than grep.
Today is Darwin Day. Good luck, and may the odds be ever in your favor.
Genetic ID software is apparently 170,000 lines of MATLAB code. I have so many questions, chief among them: MATLAB? Really? You are just messing with us now!
https://www.theregister.com/2021/02/04/dna_testing_software/
Recently migrated my website to a static site using HUGO. Here's a bit about the experience:
https://bayesicresearch.org/2021/01/21/site-update/
Any tips/questions welcome.
- https://twitter.com/BayesicTony
a geneticist and a Bayesian; freelance scientist
workspace: Arch Linux/dwm/neovim
program in: C++/R/shell/perl
hobbies: BJJ/languages/anime/cooking
