sabraham

17 ms·

1.
by sabraham 11y ago
http://curl.trillworks.com/
2.
by sabraham 11y ago
We use Riemann at Two Sigma to monitor/alert/heal our Mesos cluster [1], precisely because of above reasons to reject. >- You must pick up Clojure to understand and configure Riemann (we're not a Clojure shop, so this is a
3.
by sabraham 11y ago
Mesos uses dominant resource fairness: https://www.cs.berkeley.edu/~alig/papers/drf.pdf
4.
by sabraham 11y ago
To load the framework (application) with the logic with the concern of deciding what resources it wants. The scheduler shouldn't care about what you get, just fairness. Two-tiered scheduling achieves this: Mesos decides how many resou
5.
by sabraham 12y ago
Great story, but total myth! http://www.theguardian.com/politics/blog/2013/oct/02/david-c...
6.
Statistical Analysis of NYC Christmas Trees(tumblebolt.com)
2 points by sabraham 12y ago|0 comments
7.
Writeup of NYC Traffic Visualization(tumblebolt.com)
18 points by sabraham 12y ago|1 comments
8.
by sabraham 14y ago
This sounds like the story Jessica Livingston mentioned at Startup School 2012 < http://news.ycombinator.com/item?id=4699862 >, if I remember correctly -- business in Texas got a term sheet, moved to California, investors pulled ou
9.
Protect Kids, Get F*cked(medium.com)
136 points by sabraham 14y ago|83 comments
10.
by sabraham 14y ago
http://xkcd.com/1168/
11.
by sabraham 14y ago
Poor example-- the Famine should be known as the Starvation. There was plenty of food to ration to the Irish, even on Irish land; the British instead shipped it home, or fed it to cattle and then shipped that home, and let the Irish starve
12.
by sabraham 14y ago
Typically newspapers omit the comma; my guess is to save ink/space, but if anyone knows for sure, I'm all ears.
13.
by sabraham 14y ago
Sorry to be late to the party-- xaa, could you explain a little more what you mean by "Ultimately I believe the breakthroughs will come fastest if we can "close the feedback loop" by automating a lot of bench biology, and then have computer
14.
by sabraham 14y ago
Definitely jokes. I wasn't totally sure until I saw this: http://www.groupon.com/pages/kidz-club-games
15.
by sabraham 15y ago
awesome, thanks
16.
by sabraham 15y ago
have a link for the event?
17.
by sabraham 15y ago
Thanks for the insights! I appreciate it.
18.
by sabraham 15y ago
The other classes sent an email when registration opened; I think it's safe to assume PGMs will be similar.
19.
by sabraham 15y ago
Ah, so I used a 2-degree heuristic to come to that conclusion--I haven't had any first hand experience with her, nor do I have contact with her former students. A few stats professors independently recommended her to me as a supervisor, her
20.
by sabraham 15y ago
PGMs don't get the same sexy treatment that ML and AI seem to get in pop science articles, so it may be worth stating explicitly that they're very much used in ML and are intellectually fascinating in their own right. A graphical model can
21.
Stanford Class: Probabilistic Graphical Models(pgm-class.org)
148 points by sabraham 15y ago|25 comments
22.
by sabraham 15y ago
It's around 40%.
23.
Aneesh Chopra: Market Opportunities in Healthcare Innovation(youtube.com)
1 points by sabraham 15y ago|0 comments
24.
by sabraham 15y ago
except in ultra-rare occasions they do not happen in first world countries You've scuttled your own argument here.
25.
by sabraham 15y ago
As to being bigger and stronger, perhaps we should look to the nation of Japan and the feats its military was able to achieve with men roughly the size of north american women. Women are perfectly capable of defending themselves, not that
26.
Fractal Music With Functional Programming(scribd.com)
2 points by sabraham 15y ago|0 comments
27.
by sabraham 15y ago
That said, I'm American, went to Oxford and am now at Cambridge, and I can assure you that most Humanities graduates are shockingly innumerate--I seriously doubt most Arts students given the Math SAT would be able to do more than shake a st
28.
by sabraham 15y ago
The article said Private Voyeur removed the post. It's not in an easily accessible cache--a quick Google and archive.org search doesn't turn up anything--so naming the site, albeit unnecessary, probably didn't jeopardize anonymity.