whirlwind013.bsky.social
@whirlwind013.bsky.social
7 followers 39 following 87 posts
Posts Media Videos Starter Packs
One of those subjects, like 'belt-fed or magazine section LMG?' and 'which calibre of rifle should we use?', that the fact we are still discussing it is in itself a black mark against the profession of arms...
Really enjoyed re-reading this. Strong arguments for replacing/refurbishing 105mm, if that kind of independent light role persists.
Conversely, that suggests a relationship between the simplicity and clarity with which one can articulate the problem and the likelihood that a silver bullet for that problem will exist.
Reposted
We keep being told "there are no silver bullets". Nonsense. There are plenty of silver bullets. If they're not working as well as you hoped, perhaps you're not shooting at a werewolf.
on.ft.com/42JHpyB
The silver bullet fallacy
[FREE TO READ] The idea that a lot of problems are difficult to fully solve doesn’t mean we should stop trying
on.ft.com
at £41/article and £221/issue, it isn't clear that your review will be able to achieve the purchase on the popular Churchill discourse it undoubtedly deserves! could you do a 1-para tl;dr for those of us without academic access, please?
In London the post used to be delivered up to 12 times/day, and collected 10 times/day. In smaller towns there could be 4. My mum says this is one reason why her parents and grandparents were only 'somewhat' impressed with modern comms tech.
that's great, thanks for explaining
Don't those figures imply RFM gain?
Reposted
Very soon, the blocker to using AI to accelerate science is not going to be the ability of AI (expect to see this soon), but rather the systems of science, as creaky as they are.

The scientific process is already breaking under a flood of human-created knowledge. How do we incorporate AI usefully?
Reposted
This seems like a pretty big finding on AI generalization: If you train an AI model on enough video, it seems to gain the ability to reason about images in ways it was never trained to do, including solving mazes & puzzles.

The bigger the model, the better it does at these out-of-distribution tasks
re: Anne Frank and lying: a constraint is not a compromise.
TBF, Caricom's demands did not take a 'balancing the books' approach, so there is no 'discount' to be made? The 'consideration' would then be part of the 'historical record' elements, both in the apology and in the setting up of the museums, I imagine.
Reposted
In our new paper, we discovered "The AI Double Standard": People judge all AIs for the harm done by one AI, more strongly than they judge humans.

First impressions will shape the future of human-AI interaction—for better or worse. Accepted at #CSCW2025. See you in Norway! dl.acm.org/doi/10.1145/...
do we know which album was being referred to as such in 1987?
Reposted
After reading it, this does seem like a big deal

Industry experts outlined important, real-world, hard tasks for AI to do. Other experts were asked to do the tasks themselves (avg time: 7 hours) & yet others graded human & AI output

Models approached parity with humans & AI is getting better fast.
is there a fuller treatment of this idea anywhere? from the article, the dates don't work (if withdrawal from Empire caused loss of national wealth and inequality, then poverty and inequality should have increased much more, much earlier), but would be interested in a more rigorous article.
"Does Boris Johnson not realise that his betrayal of the promise that Brexit is the main cause of the fall of the Tory party and the rise of Reform?" - apologies, struggling to parse the opening sentence; should it be 'of Brexit' rather than 'that Brexit'?
can someone in this thread please explain why the class action is against Anthropic rather than Libgen?