Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
drxzcl
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
91.
▲
by
drxzcl
9y ago
That's a fair point. To be honest, I'm not convinced we'll ever find a way to do software development well.
92.
▲
by
drxzcl
9y ago
Can we stop with this "programming is hard" business? In a CS program, the programming classes are going to be the easiest ones you'll do. The hard stuff is the math, theoretical computer science and algorithms.
93.
▲
by
drxzcl
9y ago
I have extensive experience writing professional (i.e. Not consumer oriented) desktop applications in both Python and C++, and you're right on the money. Java may not be sexy or purty, but it works and you can actually hire people to d
94.
▲
by
drxzcl
9y ago
I have a PhD in physics. My adviser and I used say (as a joke, but with a kernel of truth) that there's no need to go beyond chapter one of a given Landau & Lifshitz book to perform research on the subject. In fact, if you were to
95.
▲
by
drxzcl
9y ago
You'd be right, if all value production was in utilitarian, physical stuff. Increasingly however we're spending a smaller percentage of our income on stuff, and more on services and immaterial goods. To produce a two ton truck, yo
96.
▲
by
drxzcl
9y ago
Most western social democracies already have an overlapping network of benefit systems that are essentially a patchwork UBI in disguise. The system is designed not to go without food and shelter, but the exact mechanism varies depending on
97.
▲
by
drxzcl
9y ago
Productivity is usually value/worker or value/hour worked. There's no indication, as far as I'm aware, that value (inflation adjusted money) is a conserved quantity. In fact, most of the productivity increases are in ind
98.
▲
by
drxzcl
9y ago
What do you base this on? Is there some theoretical limit to productivity?
99.
▲
by
drxzcl
10y ago
This is also very true! Usually we hire AI/stats folks and do a heck of a lot of training to get them up to speed on the development side of things. You can do it the other way around, but math is a lot harder to pick up outside of for
100.
▲
by
drxzcl
10y ago
Not OP, but I think many companies aren't qualified to judge who is and who isn't a qualified candidate, at least for the first hires. This turns the whole thing into a "market for lemons". I've helped a few organis
101.
▲
by
drxzcl
10y ago
"Don't ask for too much money"
102.
▲
by
drxzcl
10y ago
So basically the pan is driving you like you're some cheap organic PID controller? No thanks!
103.
▲
by
drxzcl
10y ago
The upside of teaching decision trees first is that it allows me to spot the people that dropped out of the online course after week one right away ;) Seriously, I see an amazing amount of people that deploy basic vanilla decision trees (no
104.
▲
by
drxzcl
10y ago
Are we all going to ignore the fact that the author doesn't know what "prosaic" means?
105.
▲
by
drxzcl
10y ago
Also, it's the tip of the spear for our soft power efforts. We get more than our money's worth just in agricultural regulations alone.
106.
▲
by
drxzcl
10y ago
It's was split off at version 2.6.3(i) into an "international version" that was exported by printing the source into several large books and scanning them in, and then further developed outside of the US. I believe most curre
107.
▲
by
drxzcl
10y ago
I write a lot of code like that. Usually I start off with a huge comment block, detailing what the function does, outlining the algorithm used and explaining the notation. Only then I include both the link to the paper and and the full cita
108.
▲
by
drxzcl
10y ago
SKDH is as much of a physics insider as they come. So, if anything, she has something to lose.
109.
▲
by
drxzcl
11y ago
Interesting theory. What do you case this on? The lack of FPU?
110.
▲
by
drxzcl
11y ago
Hah, I've done this for years! I'm glad it has a name now. I call mine "halfbakery.txt", but it's not very catchy.
111.
▲
by
drxzcl
11y ago
I spend about that on a digital NYT subscription. It's not as good as a "netflix for news", but it fills out the gaps in internet coverage pretty nicely. I think Blendle (or something like it) is a necessary step on the way t
112.
▲
by
drxzcl
11y ago
It sounds like it would be straight out of a "Silicon Valley" scene.
113.
▲
by
drxzcl
11y ago
Harassment is not acceptable, regardless of whatever cynical-sounding excuses you come up with.
114.
▲
by
drxzcl
11y ago
Did they really have to drag the Dutch east India company into this by using that logo? It's going to make things... complicated around here.
115.
▲
by
drxzcl
11y ago
I know this attitude all too well. I've replied to a few questions on gamedev.stackexchange with pointers to the usual articles on game state synchronization and lag compensation, and I get lukewarm reactions of "too complicated&q
116.
▲
by
drxzcl
11y ago
It's not a "formula", it's entertainment. There is no attempt to quantify any of the variables, or even aim for basic dimensional consistency. It's, to use a technical term, horse shit.
117.
▲
by
drxzcl
11y ago
Just in case you are tired of old fashioned fixed function CPUs, check out this awesome realtime full HD FPGA renderer! http://hamsterworks.co.nz/mediawiki/index.php/Mandelbrot_NG_...
118.
▲
by
drxzcl
11y ago
I was thinking the exact same thing reading the article. It would have to be an actual class, preferably taught by the same person (to control for personal charisma etc) that does not directly translate into behavioral modification but does
119.
▲
by
drxzcl
11y ago
Did you read the article? They quite clearly make the point that there is no clear, direct relationship between offense rate and severity of punishment (17 vs 18 year olds), and they speculate this is because much of the behavior in questio
120.
▲
by
drxzcl
11y ago
I had a good chuckle at that. It instantly reminded me of the Itanium sales forecasts[1]. It's like people are afraid to say "I don't know" (especially when money is involved) and just end up telling people what they wan
More ›