vidarh

34 ms·

811.
by vidarh 8mo ago
A compatibilist view, to me, is usually immoral, because it seems to maintain the pretence of agency while admitting it's an illusion, and so persist in accepting treating people as if they have agency. People who at least genuinely be
812.
by vidarh 8mo ago
It doesn't matter if the code is different if the spec is formal enough to validate the software against it. I have no idea about codespeak - I was responding to the comments above, not about codespeak.
813.
by vidarh 8mo ago
It's possible the specifics are different but that the overall idea still could work for humans. It seems worth at least exploring.
814.
by vidarh 8mo ago
And in this case it sounds like the pathway to determining if this has an effect in humans as well might be relatively short given there is a pool of patients receiving vagus nerve stimulation for other things that might provide data.
815.
by vidarh 8mo ago
The point would presumably be to formalise it, then verify that the formal version matches what you actually meant . At which point you can't/shouldn't regenerate it, but you can request changes (which you'd need to ver
816.
by vidarh 8mo ago
This is a compatibilist view. However, we can tell that most people don't adhere to a compatibilist view of free will because it tends to make people very upset if you suggest they have no "genuine" free will or agency, and
817.
by vidarh 8mo ago
Even if it's 50% accurate the benefit of picking both is marginal. If it's 50% accurate and you pick both boxes, 50% of the time you get nothing, 50% of the time you get 0.1% more. So unless you know that the predictor is worse
818.
by vidarh 8mo ago
I would say that if we presume memories of everyone involved have been modified, that is an equally strong predictor that we are in a simulation.
819.
by vidarh 8mo ago
Too late to edit my original comment now, but my intent was actually to make the argument for picking one , hence my comment on not caring much about the $1000. If the predictor is indeed flawless, or almost flawless, if I were to be the t
820.
by vidarh 8mo ago
That's great. It's still over-engineered given processing this data in-process is more than fast enough at a scale far greater than theirs.
821.
by vidarh 8mo ago
Both Claude and Codex handle Ruby just fine.
822.
by vidarh 8mo ago
> Ruby always drives me nuts how things are just in-scope and we don’t know why or where they came from. irb(main):005:0> Foo.new.method(:bar).source_location => ["tmp/test.rb", 5]
823.
by vidarh 8mo ago
I was about to say they have a self-hosting guide, but I see they use third party services that seem absolutely pointless for such a tiny dataset. For comparison, I have a project that happily analyzes 150 million tokens worth of Claude ses
824.
by vidarh 8mo ago
Read: is accessible from wherever you are. The implications being that you're not already carrying your laptop around with you.
825.
by vidarh 8mo ago
Assuming I have no way of testing the predictor, my decision would be to pick both boxes on the basis that $1000 is not a lot of money to me, but $1000000 is, and I wouldn't worry about the odds, because without knowing the nature of t
826.
by vidarh 8mo ago
I agree with you that it doesn't require that you are in a simulation, but a flawless predictor would be a strong indication that a simulation is possible , and that should raise our assumed probability that we're in a simulation
827.
by vidarh 8mo ago
Well, there can be an advantage in a bot too, if it can actually resolve some subset of problems faster and allows for more timely escalation of what is left. The problem is of course that is far too often not how they are used.
828.
by vidarh 8mo ago
The frontier models all handle Ruby just fine. So does th cheap Chinese models like Mini, Qwen, Deepseek.
829.
by vidarh 8mo ago
Per the article I linked to, '93. Which is presumably why I didn't notice until my son started playing with it, as I'd stopped playing with mine by then.
830.
by vidarh 8mo ago
Interesting - never come across Pantasy/GoBricks, or Lumibricks but then it's a few years since my son decided he was too old for LEGO, and I see Pantasy is just a few years old, and Funwhole/Lumibricks just a few more. Great
831.
by vidarh 8mo ago
More than just bricks fitting into each other at a superficial level, it matters how firmly they fit together, and it's one of the areas where LEGO is generally superior to the similar types of bricks. A detail I didn't realise un
832.
by vidarh 8mo ago
The difference is that I know what my "symptoms" are, but I don't know whether the combination of them are sufficient to justify an ADHD diagnosis. I don't like to assume because there's a tendency for a lot people
833.
by vidarh 8mo ago
It's just not true LLMs are limited to "static text". Data is data. Sensory input is still just data, and multimodal models has been a thing for a while. Ongoing learning and more extensive short term memory is a challenge,
834.
by vidarh 8mo ago
That's just nonsense. That they produce slop does not negate that I and many others get plenty of value out of them in their current form , while we get zero value out of fusion so far - cold or otherwise.
835.
by vidarh 8mo ago
Thankfully it's not as bad as that. The 50% that goes red means we re-execute those steps, potentially several times, to see if they succeed, before we even bother manually looking at it. But the overall principle holds: First yoou mul
836.
by vidarh 8mo ago
It's sufficient to think that there is a chance that they will not be, however, for there to be a non-zero value to fund other approaches. And even if you think the chance is zero, unless you also think there is a zero chance they wi
837.
by vidarh 8mo ago
Cold fusion reactors haven't produced usable intermediate results. LLMs have.
838.
by vidarh 8mo ago
Exactly. At the moment it's close enough to be a wash for some cases, or tilts seriously one direction or other for others. I expect improved harnesses means more and more we'll just be able to re-run a couple of times, and fall b
839.
by vidarh 8mo ago
I still won't use what? I use Opus now, and I will use Opus then too, but as I clearly stated: My default model has now dropped to Sonnet, because Sonnet can now do most of my tasks, and we already use Kimi, Deepseek, and Qwen. They&
840.
by vidarh 8mo ago
I have data going back many years, but this recent effort is a few months old at this point. It's however notably an iteration that has reduced the amount of time I spend on collating and reviewing data, by automating away most of my

More ›