Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
moab
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
22 ms
·
211.
▲
by
moab
12y ago
Why on earth would you have to brute force all numbers? You can do this optimally: just hash the numbers you see (or get requests for).
212.
▲
by
moab
12y ago
Sources? I've seen first hand the limits of the 'data' we had at school. Papers constantly citing in their experiments: "the largest real world graphs we are aware of are on the order of 1B vertices" (a twitter grap
213.
▲
by
moab
12y ago
He wasn't saying google code was bad, just that it's likely the next service to be cut by our friends over at Google. Given that v8's moving, this is pretty reasonable prediction.
214.
▲
Why Is American Teaching So Bad?
(nybooks.com)
29 points
by
moab
12y ago
|
75 comments
215.
▲
by
moab
12y ago
Yup - thanks for phrasing that better :) Price reduction has played out its tune in tech before - at cost pricing was how Amazon won (and continues to have a chokehold over e-commerce). At Amazon, however, employees from the warehouse stock
216.
▲
by
moab
12y ago
Probably because taxi companies are a dime-a-dozen across various cities and states, and the capital required for an engineering investment to pull something like this off just isn't there at any given shop. It's a technology prob
217.
▲
by
moab
12y ago
Your point reminds me of that recent New Yorker article on learning [1]. In particular, the author points out how particularly niche industries (competitive sports, theatre, orchestras) have seen significant improvements in median performan
218.
▲
by
moab
12y ago
I think what'll come out of this entire experiment is something very, very similar to the current taxi industry - simply with better apps and dependability, and I will not complain at all. A little part of me suspects that this will no
219.
▲
by
moab
12y ago
I feel like you're playing the devil's advocate here. Despite the recent witch-hunt surrounded the company lately (which I won't try to defend), you can't deny the fact that they've engaged in some fairly unscrupulo
220.
▲
by
moab
12y ago
Yeah, I know about hbgary. They definitely have a tainted public perception - I just question the extent to which people seem to vilify them. Regarding ethics - absolutely. It's also harder to notice when you're being paid quite a
221.
▲
by
moab
12y ago
The hbgary stuff that trhway mentioned was what I was thinking of. It's sketchy, and the higher ups definitely don't seem to be above this sort of underhanded nonsense.
222.
▲
by
moab
12y ago
OP is hinting at the fact that Palantir supposedly does some fairly unscrupulous work for governments. Whether this is true or not isn't exactly 'known', but it's been widely hinted and spread around the valley. Take it
223.
▲
by
moab
12y ago
You're probably right. I guess I'm looking at a visible divide in skill and knowledge occuring in my friends right now and trying to find some sort of casual factor. While casting people along the 'startup/big-company&#x
224.
▲
by
moab
12y ago
This distinction is already visible today in our field. Looking at my batch of CS grads, there's clear divide between people who took on riskier positions or joined startups and people who continued to do what they were already good at
225.
▲
by
moab
12y ago
What? I hope you're being ironic, because if you're not you seriously don't understand the point of bot-based strategy games: https://www.battlecode.org/info/ Having one person being able to pay for more
226.
▲
by
moab
12y ago
The context is the paper. Please try not to read a title and post an inane response.
227.
▲
by
moab
12y ago
Sure, I feel that way about a lot of work related things. Honestly, at the end of the day as solipsistic as it might sound, you have to play the game where you're happy, and that's 'winning'. What game this is differs fr
228.
▲
by
moab
12y ago
A lot of the post feels like hyperbole to me - here's a successful and intelligent professor, telling people to simply do what they want to do, and not worry about 'competition' and finite games. He clarifies his views a litt
229.
▲
by
moab
12y ago
I would strongly recommend the following book which speaks on this topic at length: http://en.wikipedia.org/wiki/The_Varieties_of_Religious_Expe...
230.
▲
by
moab
12y ago
This glosses over so much detail that really helps you understand MinHash and LSH. I feel like matt has really solid intentions, but this is one of the algorithms where actually reading the math and understanding it helps you write a good i
231.
▲
by
moab
12y ago
Someone's already said it, but starting with something like meetup.com just to give you that initial confidence boost and practice socializing would be perfect. Once you've gotten used to seeing, talking, and shooting the shit wit
232.
▲
by
moab
12y ago
Inheriting off-topicness, while that stereotype does apply to most mathematicians, a large number of the CS theory people I've worked/TA'd with were actually pretty good programmers, and a few had even taught multiple years o