darawk

21 ms·

181.
by darawk 4y ago
That seems to be what the case will turn on: The difference between the government providing "advice" or "helpful recommendations" and implicit/explicit coercion. I don't know enough about the details here to h
182.
by darawk 4y ago
There is no right to hear a message from a particular channel, but there is a right not to have the government intervene to suppress that channel. If the private actor decides of their own accord to censor you, that's perfectly legal.
183.
by darawk 4y ago
They worked together at LinkedIn previously, that's how he knew him. https://en.wikipedia.org/wiki/Adam_Nash_(executive)#Career He doesn't explicitly state this in the piece, but it was pretty easy to infer f
184.
by darawk 4y ago
How does it demonstrate that? Dylan had funding by the time this conversation occurred. He didn't need the authors money. The article doesn't describe how it is that Dylan acquired his initial funding. Also important to note that
185.
by darawk 4y ago
Every concrete prediction Gary has made has been falsified. All of his others are insufficiently precise to be falsified. His GPT-2 examples were thoroughly defeated by GPT-3. Horse riding astronaut is solved. Neural knowledge graphs are a
186.
by darawk 4y ago
It's a lame test, but I don't think most people were claiming that it proves general compositionality. What it does prove is that compositionality is possible with these models, and will likely improve rapidly, as everything else
187.
by darawk 4y ago
"Years of binging on weed and 'some lsd precursor'" are not exactly the same thing as traditional plant based psychedelics. The only LSD precursor you could be referring to is Ergotamine (rye fungus), which is pretty wel
188.
by darawk 4y ago
There is no evidence that psychedelics cause schizophrenia. There is very weak evidence that it may precipitate latent schizophrenia earlier than it otherwise would have manifested, but even that isn't very well established.
189.
by darawk 4y ago
There are a couple of things to keep in mind here: 1. The link between psychedelic use and persisting minor visual disturbances is reasonably well established. 2. The link between psychedelic use and serious, problematic visual disturbances
190.
by darawk 4y ago
I'm sympathetic to the impulse, and I think it's a good one to be skeptical of things like that when they conflict with anecdotal evidence. However, I think you have to look at the context of the anecdata in this case to conclude
191.
by darawk 4y ago
Psychedelic use in SF is not primarily among the homeless population. That population is using very different substances.
192.
by darawk 4y ago
https://www.nature.com/articles/nature.2015.16968
193.
by darawk 4y ago
Yes, this can happen, but it is extremely rare. Advil probably has a higher death rate than psilocybin.
194.
by darawk 4y ago
Great news. Whether or not you believe these substances have significant therapeutic potential (I do), it's basically indisputable that they are socially harmless. Nobody gets addicted to them, there are no known negative health effect
195.
by darawk 4y ago
For anyone else who was new to the phrase "isotropic model": https://github.com/christianversloot/machine-learning-articl...
196.
by darawk 4y ago
It seems odd to define AGI in a way that blind and deaf humans don't qualify for. This standard is certainly a standard for some kind of utility, but it seems very unrelated to the thing normally referred to as AGI.
197.
by darawk 4y ago
That sounds true, but the Mississippi state legislature is 29% black. That's not a majority, but it seems difficult to justify the view that the state government isn't helping these people because of racism in light of this fact.
198.
by darawk 4y ago
> The findings were somewhat lackluster: it found that the psychedelic was only marginally better than traditional treatments at relieving depression. Err, what? What could possibly be lackluster about that? A one time (or few times) tre
199.
by darawk 4y ago
Sure, but it seems pretty silly to ban the police from using data that any private citizen can access.
200.
by darawk 4y ago
No, what was at issue was precisely their creation of a dress code: > The NLRB's 3-2 decision went along party lines, with Republicans dissenting. The Democratic majority said it "found that it was unlawful for Tesla to maintai
201.
by darawk 4y ago
> There is one and only one indication for parenteral cyancobalamin (B12) administration and that is deficiency thereof due to inadequate production of gastric intrinsic factor or a disease of the terminal ileum. Period. You are aware of
202.
by darawk 4y ago
People get B12 injections from nurses all the time. It's extremely common in the US. NAD+ is probably less common, but it's not the nurse's credibility that's recommending the substance. The nurse is just there to inject
203.
by darawk 4y ago
This is right. If people vote for censorship in a democracy, that's a perfectly fine form of governmental heterogeneity. What's happening in China is not that.
204.
by darawk 4y ago
You don't need them, but friction matters. You can stay in touch via the postal service. That does not mean that newer mediums with lower friction and more capabilities do not deliver real value to people. Whether or not social med
205.
by darawk 4y ago
Ya, I wouldn't defend the broader war. I think Afghanistan was reasonable and necessary but probably should have been more surgical, and Iraq was basically indefensible. However, conditional on being committed to fighting the wars, I t
206.
by darawk 4y ago
I didn't say that, nor did I bring up the casualties comparison. The person I was responding to asked what the relative casualty counts were.
207.
by darawk 4y ago
Sure, there's plenty of uncertainty. But there are also far more casualties of 9/11 than the immediate ones. Tons of people got cancer and died, or were seriously disabled, etc. Drone strikes had some similar collateral damage too
208.
by darawk 4y ago
Why would we exclude the largest terrorist attack in American history, the thing that the interventions in question were explicitly designed to prevent a recurrence of? And anyway, if you want to get into that question, then we'd have
209.
by darawk 4y ago
These are actually easy things to look up. 910-2,200 civilians killed by drone strikes[1]. More than that were killed on 9/11 alone [2]. 1. https://en.wikipedia.org/wiki/Civilian_casualties_from_U.S._... 2. https
210.
by darawk 4y ago
Pretty similar to the appeals process for being killed by the terrorist that the CIA chose not to assassinate.

More ›