TropaeanTortiloquy
No bio...
User ID: 1899
TropaeanTortiloquy
11mo ago
Projecting images onto buildings? That one…I feel like there must be a headline behind that
Probably targeted at NatSoc Florida, who projected hate symbols / "Kanye was right about the Jews" on to buildings.
TropaeanTortiloquy
1yr ago
I recently read most of the decision after this came up in another forum. The most convincing part of the argument I found was:
-
mifepristone can lead to severe complications if used at too high a gestational age, or in the case of ectopic pregnancy
-
if delivered in the mail, there is often no ultrasound, so the gestational age or the presence of ectopic pregnancy is not checked
-
potential users can include people who don't check this themselves, and then sometimes die or experience severe adverse effects
-
while the data shows this is rare, it's undercounted because the FDA removed reporting requirement for non-fatal adverse effects, and also when the women go into emergency services the side effects are misdiagnosed as side effects from miscarriages instead of from mifepristone (because the doctor doesn't know they've taken it).
-
this implies the need to ban sending the drug through the mail, or to impose an ultrasound requirement to check for the dangerous conditions
The judge goes further and tries to ban mifepristone entirely.
There is some mention of strange bureaucratic interpretations:
the American Medical Association explained that “[Mifepristone] poses a severe risk to patients unless the drug is administered as part of a complete treatment plan under the supervision of a physician”). Thus, to satisfy Subpart H, FDA deemed pregnancy a “serious or life-threatening illness[]” and concluded that mifepristone “provide[d] [a] meaningful therapeutic benefit to patients over existing treatments.”
Whether this is true or not I don't know.
There's some heavy shenanigans about the Comstock Act, which prevents sending abortifacients through the mail, which was rendered null and void by Griswold v. Connecticut but not overturned and so still shambles on, offering a zombie-like ability to infect any cases about sending abortifacients through the mail.
TropaeanTortiloquy
1yr ago
I'm in software, and I'm impressed with it. I think it's a good tool. If it goes mainstream, I can see it replacing simple queries that get SEO'd to hell with autogenerated unhelpful articles (e.g. "compare koa and express", "most commonly used Azure services"), and queries for documentation or basic initial code generation. I think it's better than Google.
To summarise the problem: for the sort of questions I ask, it's right 90% of the time. If I don't know anything about a topic, this looks incredible but is useless -- I have no way of knowing what it's got wrong (I can ask follow-ups, but it might get them wrong too). It also has a tendency to blag, saying meaningful things which are relevant but still wrong.
If I know a lot about a topic, or an adjacent topic, this is fine: I can read the paragraphs, identify the likely errors, and refresh myself. If I ask it to make a table of equivalent services in AWS and Azure, it is mostly correct -- and it makes a more helpful table than Azure's docs, as Azure's docs are alphabetical, but ChatGPT puts the most popular ones at the top.
When writing code, it doesn't necessarily do it the best way, and sometimes makes (large) errors. If you know roughly what you're doing, you can prompt it further to iterate, and it's a great assistant. If you don't know what you're doing, you're probably just going to propagate the errors.
As an example: I just asked it to compute the inode percentage in a volume. It wants to use df -i (perfect, and 90% of the way there for a human). To compute the percentage, it observes that the calculation is "used divided by total": fair, but you could just read off the value. But when it comes to writing the script, it calculates "total divided by percentage", which is total nonsense.
If you take the docs together with the code, you can figure out what to do a lot more easily than starting from scratch. But you have to take what it gives you with a pinch of salt.
TropaeanTortiloquy
1yr ago
teeth
A comment on the status of teeth in Britain (at least where I am): I'd say that having teeth, not having any fillings, is better than having unhealthy teeth. But when it comes to teeth whitening, it's a cosmetic procedure that doesn't make your teeth any better at anything you want your teeth to do, so having it done implies you're vain and a bit frivolous, like getting a fake tan. Very white teeth look uncanny valley level to me. I understand they're the norm in America, but they're really not the norm around where I live.
This doesn't apply to being fat, because being fat implies you're too lazy to exercise or control your diet and affects your health negatively. Having aligned teeth might make them slightly better for biting, but having them whiter doesn't help at all.
For two people, one with missing teeth (or other obvious flaws) and one with all their teeth, I'm more positively inclined towards the one not missing teeth. But two people, one with "natural" teeth and one with the perfect American smile, I'm more favourable towards the former.
This isn't true globally. For example, in the UK the Public Lending Right Act 1979 grants authors a small payment each time their book is borrowed from a public library.
Checking Wikipedia for "Public Lending Right" indicates that similar schemes exist in other countries, e.g. Canada, Germany, Israel.
More options
Context Copy link