landl0rd

18 ms·

271.
by landl0rd 1y ago
The former is controlled by the most notable enemy of America. The latter is shitware that the government was stupid to buy. They are both threats to national security but in different ways and the government being stupid about TeleMessage
272.
by landl0rd 1y ago
Strictly speaking most have been arrested on immigration-related charges but been targeted for prosecution due to criticizing israel. This is bad and a large step in the wrong direction but it's markedly less bad than the "hate sp
273.
by landl0rd 1y ago
Examples like the raid on the quaker meeting are real and contemptible. I'm guessing it's because it's been weaponized against right-wingers for longer not because it's only been used against them. I hope people on the l
274.
by landl0rd 1y ago
Police have broad discretion over what they enforce and when. In America Castle Rock v. Gonzales affirmed this strongly (although it also had other questionable implications). I believe things are similar in Britain. Over here we've se
275.
by landl0rd 1y ago
Yes actually. I'm not opposed to any government above a highly-local level. The point of government is basically to fix externalities and run defense/diplomacy. For rivers we do in fact need co-ordinated governance to prevent peop
276.
by landl0rd 1y ago
You're attributing to avg consumer a much higher degree of both tech-savviness and market power than is realistic.
277.
by landl0rd 1y ago
Because objectively most of the population doesn’t depend on accessibility. Most people can complete a visual captcha.
278.
by landl0rd 1y ago
This fixes indians in boilerrooms and nigerian spam emailers but specifically not ticket scalpers. The profit is too large. Also because users don't actually control the number of HTTP requests they make. Think of sites that load indiv
279.
by landl0rd 1y ago
Second this question; would be really cool if we had solid support for some of the qvina2/qvina-w/gpuvina implementations too along with stuff like haddock for affinity-based protein-protein. P.S. putting this here since I can&#x2
280.
by landl0rd 1y ago
Objectively a lack of bot-resistance can make the service unusable for everyone. Good examples include twitter, where interesting stuff gets flooded by spam DMs and indian payout farmers, and the mentioned ticketing example, where objective
281.
by landl0rd 1y ago
This looks really nice, but what's the plugin/extension story? Part of the reason PyMol is popular is bc it's easy to write plugins in Python, which ngl is much closer to what a lot of researchers are comfortable with. This l
282.
by landl0rd 1y ago
I'd rather see bot resistance (important for everyone) and privacy (important for everyone) take precedence over accessibility (important for a small minority) and have laws change to reflect that.
283.
by landl0rd 1y ago
Can you give me examples of situations like this? I am perfectly fine with something like the clean air and water acts (in principle if not implementation). Obviously people in an upstream town aren't allowed to dump pollution in the r
284.
by landl0rd 1y ago
Yeah I really don’t care about this case much. Actually a good example of less important stuff. It’s practical things like nuclear physics (buddy majoring has had it refuse questions), biochem, ochem, energetics & arms, etc. that I disl
285.
by landl0rd 1y ago
Well-put thank you. I think the end state that makes this not a slippery slope is that there are certain practicalities that make some responsibilities impractical beyond a certain scale. Where it is correct in its maximalist form though is
286.
by landl0rd 1y ago
No, it's fine for governments to fix externalities. I just believe they should do it at low levels. I don't support Alberta meddling in local politics. I don't support the US federal or state governments doing this. If cities
287.
by landl0rd 1y ago
It goes pretty granular. I don't really like secession in concept, in this case it's a manifestation that governments like to steal more power than they absolutely must hold at that level instead of leaving it at lower ones. This
288.
by landl0rd 1y ago
Admittedly they are benefiting mostly from American refining tech in this sense. They would have a tough time negotiating advantageous trade terms on their own and not many refineries can handle it, meaning they are mostly dependent on pipe
289.
by landl0rd 1y ago
Given that existing pipelines are already overcommitted or at capacity and carry not much more than a tenth, and given that America is mostly the only option for refining the heavy sour tarry crap they pull out of the athabasca sands, this
290.
by landl0rd 1y ago
This doesn’t need to be so focused on the current set of verboten info though. Just make practice making it not say some set of random less important stuff.
291.
by landl0rd 1y ago
1. Those people don’t need frontier models. The slop is slop in part because it’s garbage usually generated by cheap models. 2. It doesn’t matter. Most people at some level have a deontological view of what is right and wrong. I believe it’
292.
by landl0rd 1y ago
Let me put it this way: there are very few things I can think of that models should absolutely refuse, because there are very few pieces of information that are net harmful in all cases and at all times. I sort of run by blackstone's p
293.
by landl0rd 1y ago
This is pretty horrifying. I sometimes try using AI for ochem work. I have had every single "frontier model" mistakenly believe that some random amine was a controlled substance. This could get people jailed or killed in SWAT raid
294.
by landl0rd 1y ago
Not the name of the user, a separate name. I'm aware of this. That is also not true, it can access old conversations, this is a known feature. I have been able to have it access back to the beginning of my using the site.
295.
by landl0rd 1y ago
Yeah we really should stop focusing on model alignment. The idea that it's more important that your AI will fucking report you to the police if it thinks you're being naughty than that it actually works for more stuff is stupid.
296.
by landl0rd 1y ago
It's not incorrect, you drop water and glycogen quickly starting a diet. This isn't a "repeatable" gain unless you put it back on. Still I wish they were less prone to barfing ten pages of disclaimers and "safety&qu
297.
by landl0rd 1y ago
I am not sure how I'd provide proof. But I'd encourage you to test it. It's always possible it's a bug. You can check with something like telling it that your real name is something very identifiable/not a typical n
298.
by landl0rd 1y ago
No I'm talking about specific information not related to that. You're right that it has access to that sort of rough information.
299.
by landl0rd 1y ago
By specifically testing it. I even made an extra account to get a clean state. You can check its memory interface and find nothing, delete all chats. It will still remember it. If you delete that and start a new thread, it may even mention
300.
by landl0rd 1y ago
One important note is that chatgpt has a memory you cannot see, besides chat history and besides memory. You cannot purge or manage this memory. I don't yet know how long it lasts. I don't know if it's some form of cached rec

More ›