site banner
Advanced search parameters (with examples): "author:quadnarca", "domain:reddit.com", "over18:true"

Showing 7 of 7 results for

domain:ashallowalcove.substack.com

The more clear-headed I think just don't think that the actions needed to stop the boats, and the fight with the blob that it would require, are worth it.

This requires indigenous young men to go out and shoot the people on the boats. They'll stop coming once they know it's a death sentence.

Europe isn't capable of doing that; its old men, old women, and (to an extent) its young women are all in agreement that indigenous young men should be replaced. They'll do anything to avoid raising their station in life because they believe they'll revolt as soon as it does, which is not an unreasonable thing to fear given that's when regime change generally happens.

ℝ is perhaps the most real character.

At the same time, they're not exactly keen on having tens of thousands of young men who are, at best, drains on the welfare state and, at worse, serious criminals, coming to the country. Especially with the papers carefully documenting every landing.

What's the evidence for this?

Out of curiosity, what was your opinion on similarly extremely online and extremely vain billionaire Bill Ackman buying his way into a tennis tournament, playing doubles with a guy he hired against real pros.

He got murdered and people were upset about it.

I thought it was great. The tennis tour get money, the guys playing got attention, and nobody got hurt.

Buying your way into being cool, whether by playing the impressario to a sports team or arts scene, or you build a submarine or a rocket ship to make yourself into an explorer; sometimes you win tournaments by paying everyone to pretend you're good at things.

Hell, in BJJ we have goofy-ass categories in tournaments, where they have such narrow weightclasses and belt levels and age ranges where guys get "medals" because there are only 3 people competing. (Anthony Bourdain won "silver" in a BJJ tournament, which sounds really cool if you don't realize this)

This is just a new version of that, isn't it?

I think those are called "dogs".

This seems like a great example of vibes based thinking from both ends.

The change is made because it has some slight vibes of being woke (since the column is called gender) so that's good enough to score an easy win. And it's without much effort, which a lot of them seem to be really lazy and uncaring with this work given how they've messed up multiple times this same way with the Enola Gay or that Army Corp biologist page that included fish gender. I'm not even kidding

And a photo of Army Corps biologists was on the list, seemingly because it mentioned they were recording data about fish — including their weight, size, hatchery and gender.

Nobody wants to do an in-depth investigation or look through data because that's boring and the only benefit is that you might have to say "sorry boss I looked at it and I didn't find woke" when you can instead go and say "Boss we removed 200k instances of woke"

And then people online are upset without even knowing the details because it has the vibes of being against the Trump admin despite it most likely not being any data deletion and just a change in header.

https://www.psypost.org/secret-changes-to-major-u-s-health-datasets-raise-alarms/

I had to do a double take when I saw this article because I was on the exact team at the VA that did (part of) this. The reddit discussion is being hysterical about data loss but as the article reflects, the changes were purely to column headers and data element names-- or at least, that was what I heard during meetings. (I didn't actually make any of the changes, it was very much "not my job".) The bigger issue is that it sounds like the VA failed to advertise what happened to outside stakeholders. In case any of them are listening... the data tracks and has always tracked the sex at birth, and has never included the gender identity. The columns were called "gender" for the historical reason that the medical field didn't always view gender as being separate from sex.

In effect the whole change was just CYA thing-- the big bosses were making a stink about culture war stuff, and they spat out the easiest possible fix. So far as I know this had no actual impact on any healthcare measures. I can't rule out the existence of eCQM that include gender identity, and there's a (now-deprecated) FHIR extension for gender identity. but frankly I doubt we ever used it. Our data source didn't even keep track of ethnicity, which gets used as a supplement for basically every QDM measure.

Basically a waste of time, and therefore money. Being optimistic, maybe it'll be less confusing for measure developers, but it's hilarious to me that the conservative administration was basically ceding the point here by differentiating at the schema level that "sex" is different than "gender".