d1sxeyes

7 ms·

91.
by d1sxeyes 6mo ago
That “with enough context” is doing a lot of work here. If you take a great engineer, drop them in front of an unfamiliar codebase, it’ll take them more than an hour to do most non-trivial tasks.
92.
by d1sxeyes 6mo ago
I think we're approaching this from different angles. 1 in 30 is better than the observed rate, but worse than the estimated rate . FWIW, the 1/12 is also actually off, the long-term mortality rate for Apollo astronauts is hi
93.
by d1sxeyes 6mo ago
But the idea we're exploring is that the coin is fair (i.e. the 1/1000 estimate is correct, and the Apollo missions were unlucky).
94.
by d1sxeyes 7mo ago
But you doing better is independent of the risk involved. The chances of you getting 3/4 heads or better is around 31%, so theres ~69% chance you’ll do worse next time round. Doesn’t change the fact that each coin toss is still 50&#x
95.
by d1sxeyes 7mo ago
If I toss a coin four times and it comes up heads three and tails once, it doesn’t mean that there’s a 75% chance that this coin lands heads up. Be careful about conflating risk factor and mortality rate.
96.
by d1sxeyes 7mo ago
I’ve seen this a few times, it mostly happens when a subagent returns. It seems that Claude sometimes doesn’t understand that the message coming back from the subagent is not from the user.
97.
by d1sxeyes 7mo ago
One thing you can try⸺admittedly it's not quite correct⸺is replacing them with a two-em dash. I've never seen an AI use one, and it looks pretty funky.
98.
by d1sxeyes 7mo ago
That’s different. That’s to get people onto API plans where tokens cost a lot more than they do on the subs (especially targeting OpenClaw users).
99.
by d1sxeyes 7mo ago
I honestly think he believes he’s doing good for mankind in a “fun and quirky way”, I don’t think he’s particularly trying to flip everyone off.
100.
by d1sxeyes 7mo ago
I think so, I think they proposed ice creams only for those who haven’t been in Nidd since the 2021 Paris incident, but that got voted down.
101.
by d1sxeyes 7mo ago
Not completely 'no good reason'—they needed to test the ability to send heavy payloads, it's great marketing for SpaceX (who intend to make money by having people pay them to put things in space for them) and brand awareness
102.
by d1sxeyes 7mo ago
No, as part of the Cameron rules of 2016, concession means concession, regardless of anything else (including whether or not it’s a good idea).
103.
by d1sxeyes 7mo ago
The Milgram experiment also couldn't be repeated today as it was completely unethical. It caused huge psychological distress to participants to the point that some participants had seizures.
104.
by d1sxeyes 7mo ago
Yeah, I actually have both an alias for `rm` and a custom seatbelt sandbox which means the agent can only delete stuff within the directory it’s working in, so wasn’t an issue, was just fun to watch it say “hm, that doesn’t seem to work. Lo
105.
by d1sxeyes 7mo ago
That's just tax fraud though. There's also the CFC rule, which means that within the EU, if you control a foreign corporation, your country of tax residence can tax undistributed profits. Often tax offices don't bother and
106.
by d1sxeyes 7mo ago
I did this, Claude detected it and decided to run /bin/rm directly.
107.
by d1sxeyes 7mo ago
Almost every new system of governance has been better than what came before.
108.
by d1sxeyes 7mo ago
Can’t tell because of the paywall but I assume this is talking about playing an instrument rather than listening to a record.
109.
by d1sxeyes 7mo ago
Try /btw
110.
by d1sxeyes 8mo ago
Having been involved in this sort of exercise before, I can explain. What they will have done is asked a human who should be knowledgeable approximately how much time they spend on the activity (e.g.: how long do you normally spend copy and
111.
by d1sxeyes 8mo ago
A (small) majority of states require employers to grant time off to vote and a (large) minority require that time to be paid. Although as others have noted, it is often the case that the window for voting exceeds a single shift (dependent o
112.
by d1sxeyes 8mo ago
Even if the ID is nominally free, if I have to take a day off and pay for bus/train tickets to wait in line at some office, it’s not really free.
113.
by d1sxeyes 8mo ago
True, but you need to run the optimisation function to calculate at what X does the increased revenue at lower margin from Pro Max subscribers outweigh the reduced revenue at higher margin for API consumers, and adjust your pricing to appro
114.
by d1sxeyes 8mo ago
> If you assume API demand is unlimited Doing a lot of heavy lifting here. Not everyone on a subscription plan would convert to a 200USD/mo API consumer.
115.
by d1sxeyes 8mo ago
But opportunity cost is not actual cost. “If everyone just kept paying but used our service less we would be more profitable” is true, but not in any meaningful way. Are Anthropic currently unable to sell subscriptions because they don’t ha
116.
by d1sxeyes 8mo ago
OP double negated - cursive is the norm for Russians of all ages. Russian cursive is actually not that bad to read for the most part. Russian “print” is super awkward because all the characters are very angular. There are some differences b
117.
by d1sxeyes 8mo ago
Agree. Also, sometimes I intentionally want the agent to do something differently to how it would naturally solve the problem. For example, there might be a specific design decision that the agent should adhere to. Obviously, this will le
118.
by d1sxeyes 8mo ago
Ah, true! You just restore the one that matches its hash. Elegant.
119.
by d1sxeyes 8mo ago
At least three copies, so you can recover based on consensus.
120.
by d1sxeyes 8mo ago
But what tools do we have to stop this happening? I agree, we can (and should) all refuse to participate in licence laundering, but there will always be folks less principled. I don’t know what happens next, honestly.

More ›