placeholder

Go Performance Tales

The last few months I've had the pleasure of working on a new bit of intake processing at Datadog. It was our first production service written in Go, and I wanted to nail the performance of a few vital consumer, processing, and scheduling idioms that would form the basis for future projects. I wrote…

Click to view the original at jmoiron.net

Hasnain says:

This is a really insightful article. Especially:

"A quick grep aes /proc/cpuinfo showed that the aws c1.xlarge box I was on lacked these. After finding another machine in the same class with them, throughput increased by 50-65% and strhash's prominence was drastically reduced in the profiles."

Posted on 2014-04-17T17:03:33+0000

placeholder

Railroaders' Guide to Healthy Sleep

� A resource from the Division of Sleep Medicine at Harvard Medical School Sponsored by the Federal Railroad Administration, an operating mode of the U.S. Department of Transportation Produced in partnership with WGBH Educational Foundation and the John A. Volpe National Transportation Systems Cente…

Click to view the original at healthysleep.med.harvard.edu

placeholder

http://norvig.com/experiment-design.html

When an experimental study states "The group with treatment X had significantly less disease (p = 1%)", many people interpret this statement as being equivalent to "there is a 99% chance that if I do treatment X it will prevent disease." This essay explains why these statements are not equivalent. F…

Click to view the original at norvig.com

Hasnain says:

"Prof. Michael Wigler has a more pessimistic way of putting it (quoted by Natalie Angier): "Most of the time, when you get an amazing, counterintuitive result, it means you screwed up the experiment.""

Worth reading. Norvig concludes with

"By now you should see that much can go wrong between the simple statement of "this result is significant at p=1%." and the conclusion about what that really means. As Darell Huff said, "it is easy to lie with statistics," but as Frederick Mosteller said, "it is easier to lie without them." By scrutinizing experiments against the checklist provided here, you have a better chance of separating truth from fiction."

Posted on 2014-04-16T21:36:54+0000

placeholder

Hasnain says:

Some cool security/vision work. I'm pretty sure their algorithms for CAPTCHAs beat most humans by a wide margin. Which kind of defeats the point of having a CAPTCHA in the first place.

Posted on 2014-04-16T21:05:24+0000

placeholder

placeholder

http://artyom.me/learning-racket-1

Alternative title: A Haskell Programmer Tries to Learn Racket (a Language from Lisp/Scheme Family), Documenting This New Experience with Quite Unusual Honesty and Diligence, All the While Secretly Plotting to Steal the Good Bits Which Haskell Doesn't Currently Have.

Click to view the original at artyom.me

placeholder

placeholder

How Burrowing Owls Lead To Vomiting Anarchists (Or SF’s Housing Crisis Explained) | TechCrunch

Today, the tech industry is apparently on track to destroy one of the world's most valuable cultural treasures, San Francisco, by pushing out the diverse..

Click to view the original at techcrunch.com

placeholder

placeholder

How to Lie with Data Visualization | Heap Data Blog

How to Lie with Data VisualizationTeam HeapData visualization is one of the most important tools we have to analyze data. But it’s just as easy to mislead as it is to educate using charts and graphs. In this article we’ll take a look at 3 of the most common ways in which visualizations can be mislea...

Click to view the original at data.heapanalytics.com