jordan.matelsky.com
making computers better using brains and making brains better using computers. computational neuro and dumb little web apps
sticking a steam wand directly into my brain and frothing it like a cappuccino at @kordinglab. senior research scientist at JHU/APL
107 posts
174 followers
512 following
Regular Contributor
Active Commenter
comment in response to
post
Ooh, that's a very good point!
comment in response to
post
if that doesn't make you want to declare @thewaroncars.bsky.social ....
comment in response to
post
Kaweco buddy spotted!
comment in response to
post
bsky.app/profile/keen...
comment in response to
post
haha luckily I couldn't figure out how to do that part anyway! So for now, that's exactly how it works. And likely to stay that way now :)
comment in response to
post
there were some of these in baltimore last I checked — and they were dope af.
comment in response to
post
unfortunately also has a python backend... but i think i wrote things to be serverless-compatible, probably needs one student to be willing to devote ~15 min to standing it back up... otherwise agree! once the semester cools down I'll give it a whack if I can still find the registrar details!
comment in response to
post
ostensibly I think it existed at biascorrect.org but I'm not sure who's keeping the lights on there anymore...
comment in response to
post
Mollie Marr led a project I contributed to a few years ago to quantify these things in med school rec letters — some of these biases (gender in particular) are so pronounced that you can detect them with "dumb" heuristics — no clever NLP needed! github.com/gender-bias/...
comment in response to
post
punishment chair? for when you've been bad?
or reward chair, for when you want a cozy warm nook to work in
comment in response to
post
maybe this isn't the point, but... do you think they like it
i feel like it's probably pretty neat for them
comment in response to
post
refresh
refresh
refresh
comment in response to
post
"shibboleth" lends it a certain mystery ... but mysterious it most definitely is not
comment in response to
post
I use 'watch ./foo.sh' for this so that if it dies or exits cleanly it still restarts
comment in response to
post
am I pangea . ?
comment in response to
post
unfortunately also probably true
comment in response to
post
en.wikipedia.org/wiki/A_Canti...
A good read!
comment in response to
post
ha this was exactly my thought! amazing work, @swordsjew.bsky.social . Haunting and exactly the information that needs to be shared right now
comment in response to
post
PRIVATE SYNAPSE DO NOT TOUCH
comment in response to
post
i.e. if your rate limiter is patient discovery then maybe a higher overall screening rate is enough??? seems unlikely
comment in response to
post
my friends and I were wondering about this too. maybe the increased odds of someone asking for a prostate screening has a big enough financial return to them?