Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
SilasX
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
22 ms
·
331.
▲
by
SilasX
2y ago
I guarantee you that large-cap, highly scrutinized public companies comply with much harder regulations and internal controls than this.
332.
▲
by
SilasX
2y ago
But now you know of the violation! Don't you want to remind Colin of the rule now? Or maybe the rule isn't actually that important, for the reasons in my last comment, and you were just looking for a reason to make your criticism
333.
▲
by
SilasX
2y ago
Since you took notice of the discussion, I thought I'd point out that in the past, you'd given me flak for copying and reposting a previous comment of mine: https://news.ycombinator.com/item?id=16586370 Now, Colin
334.
▲
by
SilasX
2y ago
To adapt the old saying, 3 months of reverse engineering can save you 3 hours in the storage room.
335.
▲
by
SilasX
2y ago
Ah okay -- the second one had a map with the same shape as the border, which I think made me mentally lump the map's border with the box's borders and invalidate the claim that it's looking at the box's borders as part o
336.
▲
by
SilasX
2y ago
Uh, when you draw the map that requires three colors, I couldn't figure out how to submit it to at least be told it's not good enough. Or if it automatically accepts for a 3-min [1], then it was too hard to make sure that the boun
337.
▲
by
SilasX
2y ago
That ... doesn't apply when the topic at hand is the validity of oft-repeated conventional wisdom. In that case, the real anti-pattern to pretend that the conventional wisdom actually means a specific (sane) interpretation that lives
338.
▲
by
SilasX
2y ago
What do you do after you beat it?
339.
▲
by
SilasX
2y ago
>That's literally step one in OPs advice. No, it's your generous, retroactive re-writing of the OP's advice. If you genuinely don't understand that you communicated something that wasn't in the original, then you
340.
▲
by
SilasX
2y ago
For that matter, has OP even met the HN accepted wisdom? "No one knows what they're doing, everyone's faking it, it's fine if you are too" -- so don't take it as a red flag when your fumbling around keeps blowi
341.
▲
by
SilasX
2y ago
Not directly related to the central thesis, but it's annoying how Marcus promotes the narrative of OpenAI thriving on theft of IP ... and then, every time he wants to quote someone, he posts a screenshot instead of linking the source.
342.
▲
by
SilasX
2y ago
Surprised I had to scroll down this far and not see it mentioned in any of the other threads. It was likewise a case of a company being bought out in an all stock offer where the stock turned out to be worthless because of an accounting sca
343.
▲
by
SilasX
2y ago
I've long thought that we need to create a new (short) word that exists solely so you can communicate that you got the message, but not indicate any position on it. Something like AROM (acknowledging receipt of message). Of course, goo
344.
▲
by
SilasX
2y ago
Love the typo, and it's fitting here. I'm going to use it for any time the implicit behavior risks burning (immolating) you, as the sibling comments note applies std::optional. Implicit conversion that immolates: imolicit conversi
345.
▲
by
SilasX
2y ago
Yes! There was even a Supreme Court case where a justice called out its feudal character. See my earlier comment: https://news.ycombinator.com/item?id=13477220
346.
▲
by
SilasX
3y ago
How did it help?
347.
▲
by
SilasX
3y ago
Yeah, if we really wanted to improve the career match, we should reserve a big chunk (say, a year) of teenager's lives to do a whirlwind tour of some representative sample of jobs, so they'd get some kind of direct experience of w
348.
▲
by
SilasX
3y ago
>At one minute per container in the 150hrs it took the ship to arrive you could have delivered 140*60 = 8400 containers to the other side. If you're considering a case of 3000 miles, that's not a canal anymore and can use the f
349.
▲
by
SilasX
3y ago
Yeah I remember having this realization after I ran the numbers. A canal with a massive ship beats rail easily on throughput. So much for my idea to have ships plug into a maglev railway and fire their containers to the other side. https:&
350.
▲
by
SilasX
3y ago
You'd get that same benefit, and reduce numerous, non-obvious wastes of fossil fuels with a carbon tax.
351.
▲
by
SilasX
3y ago
Not everyone thinks like that.
352.
▲
by
SilasX
3y ago
Just BTC mining? Not all uses of energy? It makes more sense to do a fossil fuel tax, which penalizes all activity that doesn't justify its environmental cost, as opposed to playing whack-a-mole with whatever uses are least popular rig
353.
▲
by
SilasX
3y ago
But most people don't even get to that point because the vague title obscures anything of significance.
354.
▲
by
SilasX
3y ago
I always get compliments on how I can make sense of technical topics. It's not a superpower. It's just awareness of the need to provide context like you've described.
355.
▲
by
SilasX
3y ago
Correct, but, to bring it back to the original point, there's a difference between "sloppy code" and "sloppy code that cascades into unintended acceleration". The fact that it didn't actually cascade isn'
356.
▲
by
SilasX
3y ago
Because they're claiming it's a computer, even though that term implies that it's a mechanism by which this all happens automatically . We already have a term for the machine where you have to reset it yourself after every c
357.
▲
by
SilasX
3y ago
I made a comment in the neighboring subthread that I think answers this: https://news.ycombinator.com/item?id=40080169
358.
▲
by
SilasX
3y ago
>Can you clarify what "orthogonal primitives" are missing? Sure. From my earlier comment[1] distinguishing the core primitives -- labeled a) through d) -- necessary to call something a computer: >>if you want to get the c
359.
▲
by
SilasX
3y ago
>But a loop can be constructed with just these gates. No, it can't be. You could unroll a constant length loop into a network of these domino calculators (which is how you get Turing-completeness in homomorphic encryption from a f
360.
▲
by
SilasX
3y ago
Thank you! It feels so good to know I wasn't alone in thinking this! Here's an earlier comment where I fleshed out the point (connected to a possibly dubious, parallel argument about LLMs). Start from "If my point isn't
More ›