qup

41 ms·

691.
by qup 3y ago
We have a big enough sample size now to draw conclusions.
692.
by qup 3y ago
They like south america? I moved to Costa Rica for a time. I just wanted to check out a new place. I moved back home eventually, I was homesick.
693.
by qup 3y ago
As a kid I found it impossible. But yes, of course it's better. Joystick to strafe is way better than the yellow buttons to strafe. (I always made the one thumb stick the aiming, plus invert Y--not sure if everyone else did that)
694.
by qup 3y ago
... Which GoldenEye also had
695.
by qup 3y ago
VB6 allowed you to use bmp files as guis. I built an mp3 player that looked like an artist's pallette, completely curved. I'm sure it was some monstrosity of a hack. I didn't understand it at the time.
696.
by qup 3y ago
> today/tomorrow/someday. For me: now, soon, eventually And things might be in the "now" list for a long time. It's the hardest list to work on, for me.
697.
by qup 3y ago
I fall asleep halfway through an energy drink with surprising regularity. In my youth, I would pound a mountain dew before bed.
698.
by qup 3y ago
If we stick a visual input into the brain, it's almost certainly going to use it in some capacity, even if it doesn't become "normal" vision. Maybe re-introduce some neuroplasticity...
699.
by qup 3y ago
The photons have emanated in a sphere from point of origin with a radius of 13.5 billion light years. The radius of the universe is irrelevant.
700.
by qup 3y ago
it aged quickly during the stressful period of Y2K
701.
by qup 3y ago
move the hyperloop to panama, scale it up. rails are strong, they can hold 24000 shipping containers. we'll accelerate the steel ships with magnets to incredible speeds and global efficiency will skyrocket.
702.
by qup 3y ago
or even to just research what are non lethal doses, and set it free
703.
by qup 3y ago
sure, but perhaps they share our viewpoint. also i think "torture" is a real stretch. torture is a state we can have triggered by various methods; we can just remove that state as a possibility for digital intelligence (probably?)
704.
by qup 3y ago
I was imagining a step past what you're talking about, when the outputs are just always correct, and the bots code better than we do.
705.
by qup 3y ago
Start here, friend: https://www.railstutorial.org/ It looks like maybe it's paid now, I don't know. Hartl's tutorial was my introduction to ruby, rails, and TDD all in one go. It's a great tutorial, he&#
706.
by qup 3y ago
It'll be amazing if anyone can request any basic program they want. Totally amazing if they can request any complex program. I cannot really envision a more empowering thing for the common person. It should really upset the balance of
707.
by qup 3y ago
we'll put a portable pool on the rails. float em across. it only increases the complexity a little.
708.
by qup 3y ago
guy asked for prompts, i delivered prompts
709.
by qup 3y ago
Our ethics largely revolve around harm and permanence. It's rarely unethical to do something that is easily reversed or that causes no harm. That's pretty likely to be the case with any AI...we can checkpoint their memory, or what
710.
by qup 3y ago
only if he's multi accounting
711.
by qup 3y ago
I've worked more in Python, but rails was way more intuitive for me than Django. I knew just how I should do everything, and if I needed to alter some functionality, I knew just where to look. I loved its consistency. I like ruby quite
712.
by qup 3y ago
did you try watching the demo video?
713.
by qup 3y ago
No, he means the new OpenAI feature that is really, really poorly named.
714.
by qup 3y ago
https://github.com/paul-gauthier/aider/blob/main/aider/promp... (Not my project, just use it a lot)
715.
by qup 3y ago
maybe the poop knife could work as a poop knife. the requirements aren't as rigid as normal knife work.
716.
by qup 3y ago
That's weird. I have three copies of mine. Admittedly, I got the last one close to 20 years ago, but it was an easy process then (I walked into a building, requested it, they mailed it). My parents gave me the other two.
717.
by qup 3y ago
Which type?
718.
by qup 3y ago
The folks who built this stuff are still alive. Historian would be a better fit, and we can archive the actual first-hand accounts if we don't waste the chance.
719.
by qup 3y ago
if i'm not mistaken, the model has to be trained for a specific context window
720.
by qup 3y ago
I know this is deep down a bad comment thread, but I thought I'd chime in here. I have been writing function names and test names, and then telling gpt to fill in the test, which is usually does how I want (maybe with errors, but it te

More ›