ulyssessword
No bio...
User ID: 308
ulyssessword
9hr ago
Capsules are not tablets.
He does not have a prescription for tablets, therefore he can not receive tablets. He has a prescription for capsules, and nobody can give him capsules.
ulyssessword
1d ago
Dang, I wish we had some sort of organization that would contextualize comments like that so that less-informed people (like me) could come to a better understanding of the world. Lol, like that would ever happen, right?
More seriously, inaccurate or incomplete news articles have always annoyed me. I like being informed, but sometimes article length limits, different focuses, or pure lack of skill means that news pieces don't have the information I want to know. It's not perfect, but I'll deal with what I have.
Strategically incomplete articles that use their own lack of rigour to say there's nothing to see are a pet peeve of mine. I know that it's impossible to prove that the reporter knew about that peace deal, but loudly focusing on a point of ignorance is hardly any better.
(Also: nobody is going to correct it. A healthy news ecosystem would have a way to get the true information to me, regardless of whether it came from a correction there or a counterargument from a rival.)
ulyssessword
2d ago
Compute is dirt cheap, and dropping by the month. Doubling your compute costs means you're about three months behind the curve on economic efficiency, and (using your assumptions, which are quite generous to me) still at the frontier of capabilities.
ulyssessword
3d ago
ulyssessword
5d ago
Yeah, fentanyl is tiny. A "lethal" amount is 2 mg, so a single pound could theoretically depopulate a small city. I don't know the exact usage rates but it can't be that fast.
ulyssessword
7d ago
From your first comment:
I struggle to understand how anything but higher promiscuity could explain the difference between gay and straight people contracting HIV.
This comment thread is how something other than higher promiscuity could explain the difference between gay and straight people contracting HIV. The different transmission rates of different sex acts are enough on their own with no difference in promiscuity.
Sorry, I rephrased it to make it less accusatory.
Thanks, that clarifies it. I was mostly focusing on the image, which is why I mentioned his bad comparison in a parenthetical aside instead of the main part of the comment.
ulyssessword
7d ago
Assuming weekly sex with people from a pool of how many partners?
A big enough pool that it looks exponential instead of sigmoidal. Once it's spread to >50% of the group, you can't exactly double the prevalence.
As I said, swapping partners annually from that pool is frequent enough for the dynamic to play out.
Maybe your source's sensibilities are a bit more delicate than mine, but I would not be shocked by someone having sex with their partner 52 times in one year, then with a different partner 52 times in the next year, and so on.
Ah ok... uh, so why did you completely ignore the attached image?
What did I ignore? I saw the rate for anal sex, I saw the rate for vaginal sex. Was I supposed to discuss blood transfusions and needles? Construct a model that is more detailed than "heterosexual" and "homosexual"? Increase the precision from "1%" to "1.11%"?
I mentioned his deceptive claim because it jumped out at me. I hadn't noticed the problems with his second sentence when I posted that.
ulyssessword
7d ago
If you assume weekly sex and other simplifications, then a 1% rate of transmission doubles the infected population every two years. A 0.1% rate doubles it every 20 years. That's moderate promiscuity IMO, particularly since it still mostly works if they change partners annually (as opposed to weekly).
What am I missing?? It says "estimated median risk of HIV transmission per exposure". Where did you get anything about pregnancy?
The first sentence of the tweet (emphasis added):
Even gay sex per se is less risky than people think - comparable to female pregnancy risk with a condom
ulyssessword
7d ago
I see "only" a 10x difference
That's a shockingly huge difference that could explain the entire disparity and then some.
Masking for COVID is still poorly studied (or at least poorly publicized), but the range I saw was between 1.05x - 6x difference compared to unmasked. I have no problem believing that a 90% effective intervention could stop an epidemic in its tracks.
(As an aside, I'm pretty sure that tweet is comparing the 1.11% risk of HIV per exposure to the 2% risk of pregnancy per year.)
ulyssessword
7d ago
If you think HIV does spread readily from heterosexual sex, then why hasn't it?
Last time I ran the numbers (several years ago. It might be time for an update), gay men got HIV at >80x the rate of the rest of the population. This was reported like "20x the general population" or something, which neatly hides the fact that people-who-aren't-gay-men get HIV at less than half the rate of the general population.
I don't think your mechanism of action can dismiss such stark differences.
ulyssessword
8d ago
I just heard of a challenge on the radio: 2000 pushups from yesterday to the end of the month (for suicide or something). I just might participate too.
ulyssessword
9d ago
The reason this is being done so crudely is because every less-crude attempt made in the past was stopped.
My thoughts throughout this Presidency (all three weeks of it) has been a mix of:
-
Damn, Trump is reckless, unprofessional, and vain.
-
How the fuck does he have so much ammo?
There's a plane crash? Air Traffic Controllers were hired under a racist system. Foreign aid? Transgender operas in Colombia. Funding basic science? >60% "administrative overhead" tacked on. Threaten Canada with tariffs? Suddenly our border security is a valid issue. Random whatever? $20M in subscriptions to the Associated Press, and another $1.6M to the NYT.
It feels like a weird mirror to the Texas Sharpshooter fallacy: He gives every indication of shooting blindly, but there has actually been a bullseye where he hits all along. That could be luck or good spin, but the most compelling story is that everywhere is that bad.
(Related joke: There has been a shooting at a peaceful protest! A child molester, a sexual assaulter, and a convicted felon illegally carrying a gun are the only people injured.)
I still don't think he's doing a good job, but damn does he have a strong narrative.
ulyssessword
9d ago
The closest recent example I can think of...
Granted, I can't think of a closer one either.
There's a huge gap between pseudonymous shitposts and a government agent directing their subordinates. The raw content of their statements is similar enough (not actually that similar, but I can't be picky), but context is everything.
ulyssessword
10d ago
IIRC, it just alters their UI so none of your content shows up. You can just try blocking someone and seeing what the site looks like afterwards.
ulyssessword
10d ago
It matches the denotation perfectly: they had a platform (government-funded contracts/publications/activities), and now they don't. The connotation got all screwed up because of cancellation campaigns, so I share your hesitation.
...despite having himself gone to war on X against people making similar tweets.
Was there a link here that didn't copy over?
When I asked about DOGE's supposed hypocrisy last week, I got a deeply unsatisfying response that relied on hypotheticals and vibes. I suspect this is more of the same, but I'd like to be proven wrong.
Has Elon Musk fired or pushed for the firing of any left-winger for offences similar to that DOGE employee? While we're at it I'll repeat the question from last week: has any left-winger in the American public sector been fired for similar conduct?
ulyssessword
11d ago
The narrative of the 90s were that gay kids were just normal people that happened to like the same gender. Sometime in the 2010s this got morphed into gays being a protected class, and now
This reminds me of when The Onion was still funny (2001):
“After centuries of oppression as an ’invisible’ segment of society, gays, emboldened by the 1969 Stonewall uprising, took to the streets in the early ’70s with an ’in-your-face’ attitude. Confronting the worst prejudices of a world that didn’t accept them, they fought back against these prejudices with exaggeration and parody, reclaiming their enemies’ worst stereotypes about them and turning them into symbols of gay pride,” Thorne said. “Thirty years later, gays have won far greater acceptance in the world at large, but they keep doing this stuff anyway.”
ulyssessword
12d ago
It says "annual savings" for that $30B figure on Doge Tracker.
ulyssessword
12d ago
That's speculation. Can you point to any left-winger who was as an actual matter of fact fired for similar conduct? It doesn't even have to be at/by DOGE, anywhere in the American public sector is close enough for me.
Without that evidence, I think you're just projecting your ideals onto them and accusing them of hypocrisy when they don't live up to standards they never claimed.
How long is it going to be before video games start including fully dynamic soundtracks? Divinity Original Sin 2 did it a little bit (swapping out instruments based on the character focus), but there's room for a lot more than that, particularly if it's in realtime.
(See also Vaudeville for AI-based dialogue)
Also, how long before they get past one terabyte of data?
ulyssessword
13d ago
The point is that they're making an exception...
An exception from what? Can you point to any left-winger who was fired for similar conduct? (preferably enough of them to make a pattern, but that's way to much for me to realistically ask for)
ulyssessword
15d ago
Give LLMs zero credibility under the rules, and most of the situations can be handled smoothly.
-
Can you research using AI, and present your findings in a comment? Of course! You can research with anything, and the other people can push back on mistakes or low-quality sources. (You can also skip this step entirely and post without researching anything).
-
Can you research using AI, and present its findings in a comment? No, no more than you can dig up a random blog and copy/paste it in support of your argument.
-
Can you talk about something Claude said? Kind of. You can talk about something your uncle Bob said, but you shouldn't expect us to put any weight on the fact that he said it. Similarly, the LLM's statements are not notable. Go ahead and use it as a jumping-off point, though.
-
Can you use an LLM as a copyeditor? Go ahead and use whatever writing strategy you want.
-
Can you use an LLM as a coauthor? No, you have to write your own comments.
Maybe add a text host next to the image hosting we already have? It could give a place to dump them when appropriate.
ulyssessword
15d ago
Maybe fifty-third or so, depending on how the rest of his term plays out.
ulyssessword
17d ago
There are probably hidden links between the items. Consider "The government shouldn't subsidize college" vs. "The government shouldn't subsidize college for white people" vs. "The government should subsidize college". That's 61% in agreement vs. 0%, right?
Reading downthread, social issues vs. government size don't have as clear of a link as equality under the law, but it's easy enough to come up with a few that wouldn't make it into normal conversation. Maybe: Small conservatism is "don't take people's money", large progressivism is "give people money", large conservatism is "give businesses peoples' money" and small progressivism is "abandon the core functions of government".
Now I'm imagining an annual recalculation that adjusts the mound position to maintain the optimal stats. +2.4% one year, -0.3% the next...
Maybe do the same thing in football: Who said it has to be ten yards to a first down? Why not 9.7?
More options
Context Copy link