Will Lowe
@conjugateprior.org
Señor Research Scientist, NPC at the Hertie School in Berlin 🇩🇪 via Princeton, Mannheim, Edinburgh and a bunch of other ivory towers that will probably be billiard balls and decorative boxes by the end of the decade.
Rome Statute appreciator.
Rome Statute appreciator.
It seems to me you just can't catch everything on entry; inference from paper content is probably too hard & possibly unfair. So most thinking should be on dissuading fraudulent submission, e.g. setting the costs high, and keeping the evidence unequivocal, even if not so many papers are checked.
November 27, 2025 at 4:56 PM
It seems to me you just can't catch everything on entry; inference from paper content is probably too hard & possibly unfair. So most thinking should be on dissuading fraudulent submission, e.g. setting the costs high, and keeping the evidence unequivocal, even if not so many papers are checked.
From that point, a good repeated game could be made by: tying submission to ORCID id or institution (stable identity) putting some volunteer budget on a random audit of "author verified everything" papers (stochastic checking), naming & temporarily banning submissions from fibbers (cost imposition).
November 27, 2025 at 4:42 PM
From that point, a good repeated game could be made by: tying submission to ORCID id or institution (stable identity) putting some volunteer budget on a random audit of "author verified everything" papers (stochastic checking), naming & temporarily banning submissions from fibbers (cost imposition).
We do not *grab* our coffees here, sir
a young boy wearing a suit and tie is holding a cup of coffee .
Alt: a young boy wearing a suit and tie is holding a cup of coffee, pinkie out
media.tenor.com
November 27, 2025 at 11:26 AM
We do not *grab* our coffees here, sir
I presume this is a common enough quantity for the behavioural folk that such questionnaires already exist. Ultimately I suppose it'd be a high granularity and temporally indexed version of the social trust batteries that get fielded by WVS and others all the time.
November 27, 2025 at 9:53 AM
I presume this is a common enough quantity for the behavioural folk that such questionnaires already exist. Ultimately I suppose it'd be a high granularity and temporally indexed version of the social trust batteries that get fielded by WVS and others all the time.
Thinking idly and aloud, a short questionnaire directly asking about the probability of various things happening to them – you get <medical cost> refunded next month, the <council service> happens next week, the <professional> really does call back tomorrow – might calibrate a personal parameter.
November 27, 2025 at 9:50 AM
Thinking idly and aloud, a short questionnaire directly asking about the probability of various things happening to them – you get <medical cost> refunded next month, the <council service> happens next week, the <professional> really does call back tomorrow – might calibrate a personal parameter.
otoh I'm going to start calling all hypothetical discussions of imaginary ML in a clinical context "medical frymblal"
November 27, 2025 at 9:07 AM
otoh I'm going to start calling all hypothetical discussions of imaginary ML in a clinical context "medical frymblal"
Had we but world enough and time,
This coyness, lady, were no crime. (Sozou, 1998)
This coyness, lady, were no crime. (Sozou, 1998)
On hyperbolic discounting and uncertain hazard rates | Proceedings of the Royal Society of London. Series B: Biological Sciences
The value of a future reward should be discounted where there is a risk that the reward
will not be realized. If the risk manifests itself at a known, constant hazard rate,
a risk–neutral recipient sh...
royalsocietypublishing.org
November 27, 2025 at 8:57 AM
Had we but world enough and time,
This coyness, lady, were no crime. (Sozou, 1998)
This coyness, lady, were no crime. (Sozou, 1998)
My rule is: never post hole (unless it's sf 😜)
November 26, 2025 at 8:11 PM
My rule is: never post hole (unless it's sf 😜)
posts in the streets and tweets in the sheets, as it were...
November 26, 2025 at 5:32 PM
posts in the streets and tweets in the sheets, as it were...
Indeed. Though that was always going to depend on whose weapons the Europeans rearmed with. I imagine the assumption (that wouldn't have survived a quick peek down the game tree either) was that it would all be American kit, with all the ownership rights and guarantees of a book bought on Kindle.
November 26, 2025 at 2:36 PM
Indeed. Though that was always going to depend on whose weapons the Europeans rearmed with. I imagine the assumption (that wouldn't have survived a quick peek down the game tree either) was that it would all be American kit, with all the ownership rights and guarantees of a book bought on Kindle.
Reassuring, as that was very much my reasoning. For a little while I did play with formulating what 'fairness' in a grade curving function might amount to and characterizing the function class, but the regrettable, though salutary necessity of doing the job I am actually paid for intervened.
November 26, 2025 at 2:14 PM
Reassuring, as that was very much my reasoning. For a little while I did play with formulating what 'fairness' in a grade curving function might amount to and characterizing the function class, but the regrettable, though salutary necessity of doing the job I am actually paid for intervened.
It was Mungojerrie – or Rumpelteazer – and most of the time they left it at that.
November 26, 2025 at 11:04 AM
It was Mungojerrie – or Rumpelteazer – and most of the time they left it at that.
My modern solution is a little R package that takes a 0-100 mark and reverse-engineers shared difficulty & slope parameters in an imaginary IRT model such that new grades fulfill the institution's three quantile demands. When raw marks allow it's a smooth reasonably fairness-preserving grade curver.
November 26, 2025 at 10:46 AM
My modern solution is a little R package that takes a 0-100 mark and reverse-engineers shared difficulty & slope parameters in an imaginary IRT model such that new grades fulfill the institution's three quantile demands. When raw marks allow it's a smooth reasonably fairness-preserving grade curver.
"...in Times of Crisis" is the go to panel title around here, regardless of topic. One of the panelists must claim that something is "a wake-up call" and that something else is "not a panacea". An audience member will ask something about the "global south" and everyone will politely ignore them.
November 26, 2025 at 10:30 AM
"...in Times of Crisis" is the go to panel title around here, regardless of topic. One of the panelists must claim that something is "a wake-up call" and that something else is "not a panacea". An audience member will ask something about the "global south" and everyone will politely ignore them.
It's somehow appropriate that actual black holes, which famously swallow everything nearby and from which no light or information escapes, have the very much the same effect as talking about "black holes" does in public finance discussions.
November 26, 2025 at 9:23 AM
It's somehow appropriate that actual black holes, which famously swallow everything nearby and from which no light or information escapes, have the very much the same effect as talking about "black holes" does in public finance discussions.
November 26, 2025 at 9:08 AM
This one, because not only is it self-describing but it's the respectable aftermath of the immediately previous holiday that you can't quite remember the details of, and that's ok.
(Outside of Scotland, ymmv)
(Outside of Scotland, ymmv)
November 25, 2025 at 10:17 PM
This one, because not only is it self-describing but it's the respectable aftermath of the immediately previous holiday that you can't quite remember the details of, and that's ok.
(Outside of Scotland, ymmv)
(Outside of Scotland, ymmv)
Without wanting to defend foolish parochialism (or the sophisticated kind, for that matter) this passage does really invite a Ryle 'category error' response.
November 25, 2025 at 10:07 PM
Without wanting to defend foolish parochialism (or the sophisticated kind, for that matter) this passage does really invite a Ryle 'category error' response.
Sometimes Harvey enters and participants see him from the left. Sometimes Harvey enters and participants see him from the right.
Attenuation, heterogenous treatment effect, or plain old SUTVA violation?
Attenuation, heterogenous treatment effect, or plain old SUTVA violation?
November 25, 2025 at 2:41 PM
Sometimes Harvey enters and participants see him from the left. Sometimes Harvey enters and participants see him from the right.
Attenuation, heterogenous treatment effect, or plain old SUTVA violation?
Attenuation, heterogenous treatment effect, or plain old SUTVA violation?
and still others will turn out to be simply out of scope for psychological theory altogether, like phlogiston as an explanation of colour.
So bring it on. We'll see what's standing when the dust settles.
3/3
So bring it on. We'll see what's standing when the dust settles.
3/3
November 25, 2025 at 9:33 AM
and still others will turn out to be simply out of scope for psychological theory altogether, like phlogiston as an explanation of colour.
So bring it on. We'll see what's standing when the dust settles.
3/3
So bring it on. We'll see what's standing when the dust settles.
3/3