Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
jhgb
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
80 ms
·
841.
▲
by
jhgb
5y ago
Maybe that's just me being an EU citizen but I fail to see the supposed threat. Is that a US thing to see a legal threat in everything?
842.
▲
by
jhgb
5y ago
In that case the response should have been just that, presumably. Nothing prevents you from making requests, they just might not be answered.
843.
▲
by
jhgb
5y ago
> An open "we're researching responses" would've been fine. That seems to be highly unscientific/prone to skewing the sample. It probably would have rendered the results useless.
844.
▲
by
jhgb
5y ago
> that is an edge case which Google would need to figure out How do you figure out legality of having a document from the document? Unless it's child porn, where it's implicit, this simply isn't computable.
845.
▲
by
jhgb
5y ago
> I think it isn't a huge deal if an English or French or Social Studies teacher isn't great in math I think it is a big deal if any teacher isn't great in math, aside from perhaps PE instructors. Considering how much ou
846.
▲
by
jhgb
5y ago
> crystal radio coherer I'm not sure there was such a thing. Coherers predated crystal detectors and from what I recall worked quite differently.
847.
▲
by
jhgb
5y ago
polote: "I got banned!" wayoutthere: "If you want tinder to work for you consistently as a man, get your personal hygiene in order and hit the gym." Did you just blame GP's getting banned on him not having personal
848.
▲
by
jhgb
5y ago
> as soon as they detect non standard behavior they shadowban, people keep seeing profile and keep paying That's called "fraud" in most countries. If you believe that fraud is "not 100% wrong" then I don't k
849.
▲
by
jhgb
5y ago
What if I put sharp-ish rocks in my shoes?
850.
▲
by
jhgb
5y ago
You can require whatever you want. Doesn't mean it will happen the way you want to.
851.
▲
by
jhgb
5y ago
How about microwaves? If the soil is moist, perhaps microwaving the area could sterilize it thermally. Slower rate of heating should prevent explosions.
852.
▲
by
jhgb
5y ago
> while fairly moist, are not particularly flammable These two things seem to be connected. In my experience infants get more flammable as they get drier.
853.
▲
by
jhgb
5y ago
I was confused by the quadrillion figure for a while, then I remembered that our quadrillions are bigger. (Does this qualify as an exception to the rule that everything is bigger in Texas?)
854.
▲
by
jhgb
5y ago
Well, I didn't dispute any of these two. I'm just not sure how one can claim that a breeder's spent fuel "has relatively short half life" when the actinides from a breeder (such as Pu239) have medium half life and a
855.
▲
by
jhgb
5y ago
I think the interesting question is for what alternative purpose you could use that water for in Texas or Arizona. While processing a wafer might require quite a bit of water, it is also the case that the economic value of that finished waf
856.
▲
by
jhgb
5y ago
As I said, the chances of anything like this being adopted on a national level are virtually zero. Good luck to any such directive coming from Brussels. Hell, I wonder if this couldn't trigger yet another constitutional amendment like
857.
▲
by
jhgb
5y ago
Then you're reconfiguring a breeder into a burner. You don't have both at the same time. But I have to say that the part where you concentrate a substantial mass of an unholy mix of medium-lived actinides to be burned into a fairl
858.
▲
by
jhgb
5y ago
But you can't end up simultaneously with less and more actinides at the same time. That alone is mathematical impossibility.
859.
▲
by
jhgb
5y ago
You're talking about fast-neutron reactors, not about breeder reactors. Fast-neutron reactors can split certain actinides that thermal reactors won't split, but a fast-neutron reactor doesn't have to be a breeder and a breede
860.
▲
by
jhgb
5y ago
I'm Czech. I'm really not aware of any "efforts to ban encryption", and given our history with the StB ( https://en.wikipedia.org/wiki/StB ), local population really wouldn't react well to th
861.
▲
by
jhgb
5y ago
It's not if it can't be distinguished of the same blur of something completely different.
862.
▲
by
jhgb
5y ago
Oh, I forgot lazy Lisps. But those would probably not be considered "major". > Alternatively, make letrec wrap the values in thunks. Aaaand of course I forgot that as well. My bad. So basically full lexical scoping and the pres
863.
▲
by
jhgb
5y ago
Well I'm not aware of any such efforts in my country. I thought the comment above about "the current administration" was about Belgium.
864.
▲
by
jhgb
5y ago
Of course you can. The problem is you almost certainly won't end up with the original image unless the ML tool has been trained with this original image (and even then it might not necessarily appear as an output). There's so many
865.
▲
by
jhgb
5y ago
All that a breeder does is that it consumes U238 in addition to U235 (in much higher quantities than a non-breeder, of course). I don't quite see how this changes the amount of actual waste per unit of energy generated. Which means tha
866.
▲
by
jhgb
5y ago
That seems a little bit extreme unless you mean lossless reversibility, which for a process like this will never be available, as per information theory. Obviously the best thing you get by this process is a smudge of colors which will neve
867.
▲
by
jhgb
5y ago
> Determining that that X is fundamentally like Y and therefore Y rules apply. This seemingly doesn't require decision making at all. Wouldn't such determination require decision making in itself?
868.
▲
by
jhgb
5y ago
Uh, there's plenty of spent fuel components that aren't fissile. It's not a perpetual motion machine. So to say that it's not "actual waste" is rather unhelpful.
869.
▲
by
jhgb
5y ago
Time for all companies to stop providing online services in Belgium, then.
870.
▲
by
jhgb
5y ago
> Been in software dev industry for over 20 years. Never encountered a single project employing the language. Are you saying you've never seen AutoCAD or Emacs in your life? While possible, I find it somewhat difficult to believe.
More ›