rlt

21 ms·

301.
by rlt 3y ago
What are the things you think can only be done on a smartphone, not a laptop, that are so important for kids to learn early?
302.
by rlt 3y ago
Of course “tech” isn’t a monolith, and neither is “crypto”. Many people into Bitcoin or crypto don’t appreciate being lumped into the very long tail of scams that are an inherent byproduct of permissionless money. It would be a bit like pre
303.
by rlt 3y ago
My vote is for LSD.
304.
by rlt 3y ago
> No decent dentist would ever work in Mississippi. This sounds like some elitist BS.
305.
by rlt 3y ago
> undisciplined spending on programs to address San Francisco’s increasingly severe homelessness crisis. Maybe throwing money at the problem is counter productive? Nah.
306.
by rlt 3y ago
> But in the 20th century, TCE was used for many purposes, including making decaffeinated coffee, dry cleaning, carpet cleaning, and as an inhaled surgical anesthetic for children and women in labor. The things we did with (now known) ha
307.
by rlt 3y ago
From a product perspective it’s certainly better. But his antics drove away a significant chunk of the user base I care about.
308.
by rlt 3y ago
> I am talking about creative work. The kind of work where people had to go and first identify the issue themselves (not to mention come up with a unique solution in the first place), and then tell others how to work with it. I’m sure fi
309.
by rlt 3y ago
It’s not a zero sum game. Yes, some people will get screwed in the short term, as they have in every past technological revolution, but all of humanity will benefit in the long run (well, if it doesn’t destroy us)
310.
by rlt 3y ago
In my experience it’s generally accurate when summarizing content provided in the prompt context. Where it can run into trouble is “recalling” (if you can call it that) content that it was trained on.
311.
by rlt 3y ago
> you actually had to manually put in the work and the hours to try and piece together multiple key points from 10 different sources to create one unique perspective It’s supposed to be a bad thing that OpenAI is making the world more ef
312.
by rlt 3y ago
The real fun will begin when people start laying jailbreak landmines across the web and your AutoGPT or whatever gets taken over.
313.
by rlt 3y ago
Fortunately for SpaceX they’ve already proven they can build (reusable) rockets, satellites, manned spacecraft, etc. It’s the great Elon-hater paradox: his companies successes have nothing to do with him, and his companies will fail because
314.
by rlt 3y ago
Yes, space is huge, however offsetting that a bit is satellites in LEO orbit at something like 20,000 MPH, so reaction time for collision avoidance is lower than you might expect, and impact velocity will result in a very large debris field
315.
by rlt 4y ago
Elon too, regardless of how much of an inconvenient truth this is to people who don't like Elon.
316.
by rlt 4y ago
Test scores?
317.
by rlt 4y ago
Or perhaps it’s unwise to try to deny/suppress all differences between the sexes. Yes there are some masculine behaviors one might consider “toxic”, but masculinity itself isn’t inherently toxic or something that should be eliminated.
318.
by rlt 4y ago
Well yes, they eventually did... https://www.theverge.com/2021/4/14/22383957/fbi-san-bernadin...
319.
by rlt 4y ago
Only about a decade after malicious USB ports started showing up at DEFCON, etc.
320.
by rlt 4y ago
On top of that, you should fire bad employees because they’re bad, not because they do drugs and may or may not be bad.
321.
by rlt 4y ago
Lighten up. I don’t think that was voluntary https://sports.yahoo.com/elon-musk-said-whole-spacex-1217322... I’m sure he won’t be smoking weed on podcasts any more.
322.
by rlt 4y ago
Sure, but for now it seems like a nice succinct way of communicating the difference from the legacy rocket companies.
323.
by rlt 4y ago
I wonder if there’s an incentive for a large group of companies to fund open source models, sort of like Linux.
324.
by rlt 4y ago
> you can build a product with a very limited amount of people and resources If you can do it with limited resources, then so can anyone else. (This is an oversimplification, of course)
325.
by rlt 4y ago
Classic “innovators dilemma”.
326.
by rlt 4y ago
> Also, there is no "working in AI", a few thousand people are doing real AI at most. The rest of us are calling an API. I would call that “applied AI” and there’s no shame in figuring out novel ways to apply a new technology.
327.
by rlt 4y ago
You weakened the argument by portraying it as more insulting against a narrower claim than it was. But yes, also snarky.
328.
by rlt 4y ago
> Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith. https://news.ycombinator.com/newsguidelines.html > questioning the
329.
by rlt 4y ago
> Actually Musk was not the founder of Tesla Such a tired trope. For all intents and purposes Musk is absolutely a founder of Tesla. > It is basically a terrible car maker know for its poor quality And someone should tell all of the s
330.
by rlt 4y ago
Well $10M “seed” rounds are still a thing I guess.

More ›