germsandnumbers.bsky.social
Computational Epidemiologist and Associate Prof at WSU. Maker of artisinal, small-batch simulation models for the discerning infectious disease consumer. WoW and zombie epidemiology. Adopter of dogs. Not taking this terribly seriously. He/him.
184 posts
3,807 followers
938 following
Regular Contributor
Active Commenter
comment in response to
post
Happened during COVID too. My stance on the origins of COVID was "My expertise really kicks in about two weeks after that initial spillover event, so here's some experts I follow..." and meanwhile HackerNews is all "Let me, a full stack UI developer, tell you about furin cleavage sites."
comment in response to
post
This is what I hate. I've got a PhD in Infectious Disease Epidemiology. I've written highly cited papers on seasonal influenza. I'm still gonna leave this to the people who are experts in *this specific thing*.
comment in response to
post
Let me know if you figure out a solution to that one.
comment in response to
post
If you discount all the bullet wounds, the patient is healthy.
comment in response to
post
Give your co-authors a "Comments by X date or I assume you're okay with it" deadline.
comment in response to
post
Ah yes, the American Dream of absolutely zero upward mobility.
comment in response to
post
Supremely
comment in response to
post
Yeah. I made myself do this back in the day, and it helped me "get it".
See also building a preferential attachment network one node at a time with two 10-sided dice.
comment in response to
post
Physically simulating something like the PageRank algorithm?
comment in response to
post
comment in response to
post
If you *are* going to phone them in don't *also* have your biostatistical support on the grant for like, a week of FTE a year.
Because that opens you up to the much *worse* criticism that you have stumbled upon a hard problem, are underequipped to deal with it, and worse, don't realize it.
comment in response to
post
If you have a complex study, that's fine. But don't present your power calculations as a two-sample test of a difference in means if you've got a four-arm study with survival outcomes.
We'll notice.
comment in response to
post
You married up.
comment in response to
post
I really like yEd for graph diagrams generally. But my use for DAGs is for illustrative examples, I tend not to use them in the context of needing R or the like to actually be doing graph operations on them.
comment in response to
post
Tech Bros to Factory Workers and Miners: "Learn to code."
Tech Bros to Knowledge Workers: "Become Factory Workers and Miners"
comment in response to
post
Homeboy thinks assistants and IT admins are useless.
comment in response to
post
GBD cannot fail, it can only be failed.
comment in response to
post
Good luck Ellie - hoping for a successful recovery.
comment in response to
post
I suspect he means universities that don't dominate NIH funding. The University of Idaho has (or had, I don't know the status) one.
comment in response to
post
Della learning what pastry cream tastes like
comment in response to
post
This. There's no level of "Good Boy/Girl Points" you can accumulate that will make them meet you halfway.
comment in response to
post
@catanita.bsky.social on conflict and #AMR ... A sobering talk finishing with a call to action.