Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
aub3bhat
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
36 ms
·
301.
▲
by
aub3bhat
10y ago
Oh yeah and eventually Diaspora will overtake Facebook. Oh wait... If the VC model did not work it would have simply gone out of business. The reality of business is that doing things at larger scale requires capital. Sure you might be able
302.
▲
by
aub3bhat
10y ago
Because your competitor can raise those million dollars, and eventually your 20k profit per month becomes 20k loss per month. You can build all your imaginary arguments about magical businesses that make 20k profit per month while being sel
303.
▲
by
aub3bhat
10y ago
I just finished watching the Nosedive (Season 3, episode 1) from Black Mirror on Netflix, it shows a similar system. Unlike other Black Mirror episodes its quite well balanced. I would highly highly recommend watching it.
304.
▲
by
aub3bhat
10y ago
Hahaha are you joking? Roads are indispensable and essential for last mile connectivity. No matter if trains and subways become 10 times more popular they will always exist in addition to the road infrastructure. Thus autonomous electric ca
305.
▲
by
aub3bhat
10y ago
train networks and subways are significantly more expensive to build, operate and maintain, even electrified ones. Roads + autonomous electric cars are hard to beat in terms of efficiency, utility and ease of operation.
306.
▲
by
aub3bhat
10y ago
I agree its very gracious of Lessig and completely agree that individuals deserve privacy. But lets be honest, even in this case the forgiveness originates in fact that all individuals involved are on the same side in this election. Further
307.
▲
by
aub3bhat
10y ago
I work on 12th floor in the same building:D
308.
▲
by
aub3bhat
10y ago
In a stark contrast, I think is Google's 111 8th Ave Port authority NYC building, its one of the best workplaces. All advantages of having a huge building (Bigger area than new Apple spaceship campus), while still being in an extremel
309.
▲
by
aub3bhat
10y ago
>> I said this one WASN'T. HAHAHAHHAHHAHAHAHAHHAHAHH, ROFL Come on dude, I am sorry but you are wrong. Not just wrong but it seems you fundamentally misunderstand how world works. See Peter Higgs is a Genius. The moment he publis
310.
▲
by
aub3bhat
10y ago
Sure, but it wont let me edit to remove that line, now.
311.
▲
by
aub3bhat
10y ago
Its not me but rather its you who has the wrong idea that all papers with 50 citations are good. Or rather the more citation == better research. Frankly almost 95% of papers are crap and better off not having been written, had it not been f
312.
▲
by
aub3bhat
10y ago
Hahhahah if you think having a Nobel award winning professor is not a great investment, you frankly don't understand how world works. There are Universities in some part of the world that will gladly pay more than what Edinburgh U. cou
313.
▲
by
aub3bhat
10y ago
There is valid criticism, and then there's "this guy did not do anything after making a Nobel worthy discovery, should have been kicked out". Also "haven't been cited particularly much" is utterly bullshit. Unl
314.
▲
by
aub3bhat
10y ago
This is a truly idiotic thing to say. "none of which were particularly impressive" First who are you to judge quality of his work, that guy has a Nobel. Imagine if they did follow your utterly idiotic suggestion and did kick him o
315.
▲
by
aub3bhat
10y ago
>whether compliance personnel should receive stock incentives. I don't this is about underpaying them but rather about legal complexities of inherent conflict of interest between compliance and stock based compensation. E.g. You don
316.
▲
by
aub3bhat
10y ago
It seems like a really interesting approach but it needs to be validated on a larger set of Atari games. Considering even a Linear learner outperforms DQN on several games such as Double Dunk. It's possible that the DQN underperforms b
317.
▲
by
aub3bhat
10y ago
Can someone please explain why this "Google is evil" meme, is spreading without any concrete evidence? E.g."in that they are designed to nudge users to share more and think less." What's the basis behind the claim &
318.
▲
by
aub3bhat
10y ago
I am pretty sure that Gmail database is 1000x times larger than all non-imaging healthcare data for entire population of the world. There are instances where blockchain type technology is useful such as transactions between multiple parties
319.
▲
by
aub3bhat
10y ago
This is a great post, recently I have been trying to re-learn and understand linux (specifically) ubuntu using monitoring tools. In my opinion htop and Facebook osquery are the two best available tools for understanding how an operating sy
320.
▲
by
aub3bhat
10y ago
I completely agree with you. Public health "consensus" is driven often by cost considerations and in this case it might as well be damned. Unlike things like "vaccine refusal", here there is no enormous cost to the socie
321.
▲
by
aub3bhat
10y ago
It's not "garbage" it called science & mathematics, those terms have "meaning", and have lead to progress in hard long-standing problems, which have in turn lead to billions of dollars and millions of man-hours
322.
▲
by
aub3bhat
10y ago
Its not a "small part", its a basic litmus test. The four terms are completely different from each other, and are not names of methods. Unsupervised learning: Learning without a set of labels. Big Data: Collecting / using lar
323.
▲
by
aub3bhat
10y ago
Those two are still orders of magnitude closer to each other when compared to difference between unsupervised learning with deep learning.
324.
▲
by
aub3bhat
10y ago
Frankly as a grad student (The kind that the author apparently considers "dim witted"), the entire article is meaningless babbling without any underlying theme. I wonder if the author truly understands "Machine Learning"
325.
▲
by
aub3bhat
10y ago
Well a portion of $1B is already accounted for in the balance sheet and budget.
326.
▲
by
aub3bhat
10y ago
Well its not 5000 new people I assume ~70% would be restructuring (E.g. as the article states from Bing and Cortana). Even then at 1500 new employees, its a huge undertaking and vote of confidence into importance of AI.
327.
▲
by
aub3bhat
10y ago
Following that line of reasoning lets just assume that Apple has solved Quantum Computers, Teleportation and Room temperature superconductors why stop at AI/ML
328.
▲
by
aub3bhat
10y ago
LOL Dude its "popular press" for a reason, otherwise they would be called "technical journals". And frankly how can you be so clueless about Apple's lack of efforts in AI/ML if you claim to be member of the com
329.
▲
by
aub3bhat
10y ago
Oh wow they do, 0.09$ per hour thats cool! Hopefully it will take another week for others to transfer their models from g2 instances, should reduce price pressure!
330.
▲
by
aub3bhat
10y ago
There are two issues, Google offers a black box service where your Tensor Flow model gets optimized. While AWS offers a generic instance which is far more flexible and can be used with multiple frameworks. Second AWS offers spot pricing on
More ›