supern0va

10 ms·

181.
by supern0va 10mo ago
LLMs are honestly rather amazing for product search and comparison. Here's a use case for me last week: I'm re-organizing my bathroom sink/vanity, and I want a few counter top organizers to keep things neat and tidy. I have a
182.
by supern0va 10mo ago
It turns out that running free common internet infrastructure at scale is both hard and expensive, unfortunately. What we really need is a non-profit to run something like imgur.
183.
by supern0va 10mo ago
If you have ample free time and few commitments and/or you read very short pop fiction, I could see how you might believe this. But there's a vast world of very long and dense literature, and also...people have kids and a life tha
184.
by supern0va 10mo ago
I don't think that's the use case they're aiming for. If you want to have a meandering voice recording that lasts minutes, the friction to pull out your phone is less of an issue while you're out walking your dog. But wh
185.
by supern0va 10mo ago
I do, aggressively. Honestly, the messages from friends are the hardest to ignore/defer. That's why I also have a highly tedious DND schedule. That aside, there's also far more friction to pull out a device, turn it on, unloc
186.
by supern0va 10mo ago
I wouldn't think about the total recording time so much as the actual usage you'll get out of it. If based on typical usage, you could get even eighteen months out of it, then you're looking at around $5/month for a devi
187.
by supern0va 10mo ago
...and face the seductive allure of all those delicious notifications trying to get me to task switch to a bunch of other things? No, sir. I'm pretty much on board with anything that reduces the number of times I need to light up that
188.
by supern0va 10mo ago
Oh, to be clear, this isn't a criticism. I think it's super cool that we're moving onto the nitpick/refinement phase of this tech. :)
189.
by supern0va 10mo ago
Honestly, if you had showed this article to me even eighteen months ago, I would have been blown away at how good of a job Claude did. It's remarkable how high our expectations have been steadily creeping.
190.
by supern0va 11mo ago
I think by the time these things tend to happen, projects have outgrown their singular founder and usually have a large developer community (because issues usually take a long time to bubble up after an acquisition). So, it's hard to f
191.
by supern0va 11mo ago
How so? Presumably Jarred got a nice enough payout that if Anthropic failed, he would not need to work. At that point, he's more than welcome to take the fully MIT licensed Bun and fork it to start another company or just continue to w
192.
by supern0va 11mo ago
"We routinely devalue craftsmanship because it doesn't bow down to almighty Business Impact." I actually disagree with this pretty fundamentally. I've never seen hacker culture as defined by "craftsmanship" so
193.
by supern0va 1y ago
>I can for example iterate on the sequence in my initial post and make it novel by writing down more and more disparate concepts and deleting the concepts that are closely associated. This will in the end create a more novel sequence tha
194.
by supern0va 1y ago
With 60 minutes of talk time included, I kind of get the impression this isn't designed so that you can hand it to your kid and let them spend the day talking to Santa. I'm assuming the idea is that they do this in lieu of writing
195.
by supern0va 1y ago
>If you believe computers can think then you must be able to explain why a chain of dominoes is also thinking when I convert an LLM from transistor relay switches into the domino equivalent. Sure, but if you assume that physical reality
196.
by supern0va 1y ago
There's a marked difference between running a Twitter-like application that scales to even a few hundred thousand users, and one that is a global scale application. You may find quickly that, network effects aside, you would find yours
197.
by supern0va 1y ago
>And you can't ask "why" about a decision you don't understand (or at least, not with the expectation that the answer holds any particular causal relationship with the actual reason). To be fair, humans are also very
198.
by supern0va 1y ago
This piece is rather puzzling. The author is essentially claiming that no one can trust their own judgement on AI (or anything?), and that the lack of scientific research means we should be in a "wait and see" pattern. ...and yet,