I can't abide this Ventures erasure
I can't abide this Ventures erasure
Let's not use adverbs like "effectively" until I actually pass the replication hurdle
Extra demerits to Dataverse for assuming that any RData file must be a tabular data file
How I feel preparing replication archives
I am always shocked to discover that there are people who actually read conclusions. Sadly, some of these people even review my papers
"I write as one who spent the most recent election night watching the New York Knicks, confident that news of the election outcome would reach me well before the winner was sworn in some ten weeks later" is also a nice aspirational model for those of us who are Very Online.
political scientists will do literally anything to avoid learning how to take a partial derivative
I can't imagine these weirdos would actually tell their own teenage kids to do this stuff, even if they have some generalized abstract weirdo concern about the overall dip in risky behavior among teens ... right?
Oh yeah. For one thing, I've heard of reviewers asking to see the prereg plan and editors denying the request, which is perverse in all kinds of ways (including killing my lazy disclosure principle logic)
*almost every
I would actually think the disclosure principle would imply the opposite of the claimed theorem. Assuming prereg is verifiable info, in equilibrium every author should reveal it
BJPolS abstract graphic
#OpenAccess from our latest issue -
A Drop in the Ocean: How Priors Anchor Attitudes Toward the American Carceral State - cup.org/3Np96E5
- Allison P. Anoll & Andrew M. Engelhardt
I'm looking forward to using this functionality to create custom chatbots for my courses. Looks good in *very* preliminary testing. It can answer dumb questions like "What are this week's readings?" but also draw higher-order connections between materials.
www.theverge.com/2023/11/6/23...
"The names of Asquith,Β Bethmann-Hollweg, Berchtold and Poincare are barely remembered, yet on any reasonable accounting they belong among the great criminals of history."
An Armistice Day classic from Crooked Timber.
Absolutely. One of many of our disciplinary hiring/evaluation pathologies. (Others include prioritizing "having the idea" over doing the work, "asking big questions" over answering them well, counting a book as equivalent to many articles... I could go on.)
"But it's impossible to evaluate individual contributions if we only see jointly produced outcomes," say people who use linear regression every day of their working lives
I sympathize, but to Anton's point: once you're doing something "kind of ok", wouldn't you be better off collaborating with someone who can do it excellently?
I've only been to faculty meetings at one institution, and I think we have a fairly healthy perspective here.
But every year I read plenty of letters trying to convince me that a job candidate is a jack of all trades. I don't want a jack of all trades, I want a high-quality specialist!
It will never not drive me crazy that we reward people for doing a dozen different things badly, rather than for doing one thing very well
S-tier: Lynx
A-tier: Netscape Navigator 3.0
Punk In Drublic > ...And Out Come the Wolves
come at me bro
It's actually kind of hilarious to imagine someone who loathes Spiritualized so much that they lurk the Internet waiting for an opportunity to pick a fight about it
yes even I'm enough of a normie to acknowledge that Dinner Party > Pool Party
This is even hotter than my take that "Pool Party" is the second-best Office episode
I think you're just doing a bit
Did I skip timelines into an alternate reality where Master of Puppets simply doesn't exist?
wait I thought your take was that Sad But True is *good*, not that it's Metallica's ***BEST SONG***, holy shit
I thought you were gonna say that one song off Metallica's black album that you're always absurdly stanning
best performance of the best song of 89-95: www.youtube.com/watch?v=Pvwq...
also gonna tag @bradsmithunc.bsky.social here because I'm sure he has (incorrect) opinions
Reproducibility is one of the big reasons I've just ported my own workflow over to Python (though it's hard to get coauthors on board with that). venv and requirements.txt just make package management so much saner.