Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
Fede_V
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
74 ms
·
271.
▲
by
Fede_V
11y ago
It's kind of ironic that Marc Andreessen dismissed the New Yorker several times in the past as an 'elitist east coast literary magazine' yet this profile is absolutely fantastic, and very generous as well. I'd be curious
272.
▲
by
Fede_V
11y ago
Having young directors that have spent multiple years in a company is much better than having a professional CEO caste all from the same HBS/Accenture track.
273.
▲
by
Fede_V
11y ago
Very helpful reply, thanks. If it's not too much to ask, do you have any examples (that you are free to share) of taking some Haskell code, then optimizing it bit by bit?
274.
▲
by
Fede_V
11y ago
Very early on, people experimented heavily with non-gradient based optimization (genetic algorithms and the like) to optimize neural networks, but the results were rather poor. For very large parameter spaces, gradient based optimization is
275.
▲
Gender and Academia Redux: Researcher Bias Is a Big Deal
(slatestarcodex.com)
1 points
by
Fede_V
12y ago
|
0 comments
276.
▲
by
Fede_V
12y ago
Hey, what you did is awesome. It is a small step, but it is decidedly in the right direction - and often us Italians are too cynical and resigned to a shitty system to push back against it as much as we should. Kudos!
277.
▲
by
Fede_V
12y ago
The article itself is not terrible, but I have a strong aversion for horrible titles like that. Why not just say something like: "Start ups greatly benefit from experienced mentors"? I get the same kind of eyeroll response when p
278.
▲
Living science: Triaging Shakespeare
(elifesciences.org)
1 points
by
Fede_V
12y ago
|
0 comments
279.
▲
by
Fede_V
12y ago
I genuinely have no special insight into what Linus is like as a human being. I just think that his abrasiveness (even if it only manifests itself in a few circumstances) is a negative that people tolerate because he is such an exceptional
280.
▲
by
Fede_V
12y ago
Linus is a technical genius with a very abrasive personality. People tolerate the abrasive personality because he is an excellent programmer and steward for the kernel. What's intolerable is when untalented people cargo-cult Jobs or L
281.
▲
by
Fede_V
12y ago
That office plan looks fantastic. I was worried this would be another 'thinkpiece' about replacing the traditional open office plan with something which is more or less the same thing, but somehow has the best of both worlds and
282.
▲
by
Fede_V
12y ago
Wow, that seems wildly disproportionate to say the least.
283.
▲
by
Fede_V
12y ago
You realize that is exactly the model of VC? :) You bet on a lot of very long shots, but the few mega successes pay off sufficiently to make it all worth it.
284.
▲
by
Fede_V
12y ago
Given that the dataset of YC companies is starting to be somewhat sizable, I'd be incredibly curious to see if people tried to run a regression between lots of plausible variables that affect success (founder demographic variables, sub
285.
▲
'Bright Spot' on Ceres Has Dimmer Companion
(jpl.nasa.gov)
14 points
by
Fede_V
12y ago
|
0 comments
286.
▲
by
Fede_V
12y ago
I actually think that part of the reason why Comcast has had such a horrific time getting their proposed merger approved is because they suffer from such a horrific reputation. Obviously, shaming only goes so far... but realistically speaki
287.
▲
by
Fede_V
12y ago
One of the most effective ways we can shut down companies that make this kind of crapware is public shaming. Publicly naming companies and executives associated with this horrendous breach of privacy and security should hopefully serve as a
288.
▲
by
Fede_V
12y ago
Uhh, are you aware of what the Obama administration was doing to James Risen, one of the greatest investigative journalists from the New York Times? They have no problems going after very, very big name journalists with huge organizations b
289.
▲
by
Fede_V
12y ago
The killer feature of Nim is its amazing syntax. If you know Python, Nim will instantly feel very, very familiar. Rust looks incredibly useful to do systems level programming, and guaranteed pointer safety without GC costs is amazing, but
290.
▲
by
Fede_V
12y ago
A better analogy would be if an software hacker paid for several starving EE students to come build an amazing Arduino project for their child's science fair, then bragged to everyone about how smart their child is. Alternatively, imag
291.
▲
by
Fede_V
12y ago
That's because those people have no business being there. Those mountains are for genuine climbers that can earn their ascent there, not idiots with too much money and no respect for a mountain, who have sherpas basically push them to
292.
▲
Can The Human Brain Project be saved, and should it be?
(chronicle.com)
108 points
by
Fede_V
12y ago
|
81 comments
293.
▲
by
Fede_V
12y ago
What struck me about this article is how much dignity each of those people had. They are all working incredibly hard to better themselves, in situations far more difficult than I've ever had to deal with, while still remaining proud.
294.
▲
by
Fede_V
12y ago
It highly depends. If it's just large linear algebra operations, then it won't matter, since all the work will be done by BLAS. If it's lots of small operations, I think pypy can inline them and you might see a significant s
295.
▲
by
Fede_V
12y ago
I'm really looking forward to the numpy specific announcements. Numpy is THE basic building for every single scientific library - if pypy can get a high performance numpy, that will go a long way towards allowing scientific users to u
296.
▲
by
Fede_V
12y ago
This is great detective work, and hopefully Yahoo steps in and fixes this. Online advertising has become and more vicious - on all my computers, I run adblock/no script, etc, but on my phone, browsing sites like retractionwatch or thea
297.
▲
by
Fede_V
12y ago
Very interesting. Did the differential geometry module in SymPy not really satisfy your needs? http://docs.sympy.org/0.7.2/modules/geometry.html
298.
▲
by
Fede_V
12y ago
I would be incredibly interested to know how the model was crossvalidated. It is utterly trivial to cherry pick features post-facto to correctly predict winners, however people who are unfamiliar with machine learning might not know this.
299.
▲
Bristol University Seeks 'Associate Dean of Eureka Moments'
(jobs.timeshighereducation.co.uk)
1 points
by
Fede_V
12y ago
|
0 comments
300.
▲
by
Fede_V
12y ago
This is a really neat project OP, and I hope you can make enough money to sustain development. I really, really wish academia did a better job of sponsoring people to maintain high quality software libraries, so you didn't have to &#x
More ›