Profile

Cover photo
Florian Loitsch
Works at Google
Attended Université Nice Sophia Antipolis
Lives in Aarhus
318,214 views
AboutPostsPhotosYouTube

Stream

Florian Loitsch

Shared publicly  - 
 
Just finished reading "Thinking, Fast and Slow" (by Daniel Kahneman). Nice read.

He references an experiment by Sarah Lichtenstein [0] that highlights (in a funny way) how humans are not good at making rational choices.
This particular experiment shows how humans value a sure gain much more than a risky higher benefit. However, (and that's where the interview got funny), they still recognize and insist on the greater value of the risky bet.
Follow the (G+) link and listen to the audio to hear the conversation. It's worth it :)

Here is similar experiment you can try with your friends (from Kahneman's book):
Decision (i): Chose between
A. Sure gain of $240
B. 25% chance to gain $1000 and 75% chance to gain nothing

Decision (ii): Chose between
C. sure loss of $750
D. 75% chance to lose $1000 and 25% chance to lose nothing

Most humans prefer A over B (trading $10 in potential win against a sure bet), and D over C (preferring the hope of not losing anything).

Now, ask the same person the following choice problem:
AD. 25% chance to win $240 and 75% chance to lose $760
BC. 25% chance to win $250 and 75% chance to lose $750

Here, BC is the clear winner. However, they are just the combination of choices B and C that were previously rejected (by most players). The inferior option AD is the combination of the options on (i) and (ii) that most players picked. 
 
[0] http://www.decisionresearch.org/publications/books/construction-preference/listen.html
1
Sorin Mocanu's profile photo
 
It's one of the best professional books I've seen in a while and it's very nicely written. I must also say, it proves that a research carrier can actually be exciting and rewarding.
Add a comment...
2
1
Luper Rouch's profile photoFlorian Loitsch's profile photoPeter von der Ahé's profile photo
Add a comment...

Florian Loitsch

Shared publicly  - 
3
2
Gabor Orosz's profile photoIstván Nagy's profile photo
Add a comment...

Florian Loitsch

Shared publicly  - 
 
 
Denmark needs to be Up All Night to Get Lucky. Clever campaign to raise birth rates in Denmark, by a travel company.
10
3
Kiril Vatev's profile photoRobert Hahn's profile photoDainius Jocas's profile photoGabor Orosz's profile photo
 
I officially deem this not as good as the Singapore birth rate campaign video.
Add a comment...
 
My article about zones was just published.
Hope you like it!
 
Learn how zones can make your asynchronous code easier to debug (better stack traces! zone-local values! print() overriding!) and much more.

This new article from +Florian Loitsch (with help from +Kathy Walrath) tells you how to use zones to corral your asynchronously executed code.

#dartlang
4
1
Erik Corry's profile photo
Add a comment...

Florian Loitsch

Shared publicly  - 
 
It's basically an ad for his book, but the interview is still a nice (and short) read.
 
"They have this unbelievably flat affect that's really palpable when you look in their eyes."

We spoke with neuroscientist Kent Kiehl about what's it like to spend 20 years listening to psychopaths for the sake of research:

http://wrd.cm/1kEvV3I
2
1
Matt Dragon's profile photoPeter von der Ahé's profile photo
 
Interesting how treatment seems to be rather effective in juveniles.
Add a comment...

Florian Loitsch

Shared publicly  - 
 
My Deebot 76 [0] recently decided to die on me. I'm out of warranty, and the robot was slowly losing power (even after changing the batteries it would just not move as easily or suck as strong as it used to). I therefore decided not to get it repaired, but open it myself. I didn't have strong hopes (I suspect the main CPU was fried), but it allowed me to take some nice photos.

This robots have tons of sensors and a few motors. It would be said to just throw them away. If anybody in Aarhus wants them, please let me know.
Otherwise: does anybody know how to correctly dispose of the robot?

[0] http://www.ecovacs.com/features/Deebot-D76.html
2
Yannis Bres's profile photoFlorian Loitsch's profile photoPeter von der Ahé's profile photo
12 comments
Add a comment...
 
 
"...they gave elementary school children either a blue T-shirt or a red T-shirt to wear throughout the school day for six weeks.

"Teachers treated those color groups in the same ways they would use gender. Teachers said, “Good morning, blue and red kids!,” “Let’s line up blue, red, blue, red.” Kids had their names on either a red or blue bulletin board and had either a red or blue name card on their desk. But again, teachers had to treat both groups equally and not allow them to compete with one another. They simply “used” color in the same way many teachers “use gender.”

"After only four weeks, children formed stereotypes about their color groups. They liked their own group better than the other group. Red-shirted children would say, “Those blue-shirt kids are not as smart as the red-shirt kids.” Just like they do with gender, they said that “all blue kids” act one way and “no red kids” act another way (this differed based on which group they were in). They began to segregate themselves, playing with kids from their own color group more than with those from the other group.

"They were also more willing to help kids in their own color groups. Children walked into a classroom in which we had staged two partially completed puzzles. We had surreptitiously draped a red shirt across one puzzle and a blue shirt across the other. When given the option, children were more likely to help out the child they thought was in their group.

"In all of these studies, there was always a very important control group—in addition to the group of students who wore colored T-shirts, there were classes in which the teacher who didn’t talk about the color groups. She didn’t sort by color or use the color grouping to label each child. In other words, it was like being in a class of boys and girls where the teacher doesn’t mention or sort by gender; she simply treated them like individuals. In these classes, children didn’t form stereotypes and biased attitudes about groups. If the adults ignored the groups, even when there were very visible differences, children ignored the groups too. [...]

"... it seems that children pay attention to the groups that adults treat as important. When we repeatedly say, “Look at those girls playing!” or “Who is that boy with the blue hat?,” children assume that being a boy or girl must be a really important feature about that person. In fact, it must the single most important feature of that person. Otherwise, why would we point it out all the time?

"If children see a difference, they look to experts in the world (us grown-ups) to see if the difference is important or not. Don’t forget that they see plenty of differences in people. For example, they see differences in hair color. We come in brown hair, black hair, blond hair, red hair, and gray hair. But no adult ever labels this visible category, saying “Look at that brown hair kid.” “Okay, all the brown-haired kids and black-haired kids over here. All the red- and blond-haired kids over there.” Children ultimately learn to ignore these as meaningful categories, but they still notice they exist. If I ask someone’s hair color, a child can tell me. It just isn’t a meaningful category. They don’t develop attitudes about what it means to have red hair or brown hair (even the occasional blond joke isn’t constant enough for children to notice).

"But with gender, children notice the difference and adults make it meaningful. Children see the category. We made sure of that with our pink or blue shirts. Also, the experts in the world, their parents, always label the category. We put a figurative flashing neon arrow on gender and say “Pay Attention! Important Information Here!” And guess what, they pay attention."
3
1
Matias Meno's profile photoTony Bonavera's profile photo
 
Women like to color code their faces.

I think we color code kids because it makes it easier to tell if they are a boy or girl. As they get older though that kind of differentiation isn't as important anymore because you see physical differences between male and female kids and I think kids already start to actually notice those differences as well if not for observing just how their parents are different because of sex but because of how they are physically different themselves from each other because of their sex. Add to that sexual attraction and it becomes even more apparent.
Add a comment...

Florian Loitsch

Shared publicly  - 
 
I don't know how Google handles warrants in Europe, but this nice video shows how it's done in the US.
 
How Google deals with search warrants.  You should watch this.  It has cows. Also, "Nice Moustache."

Way of a Warrant
1
Add a comment...
 
Remember the "Reproducibility in Computer Science" paper? (I posted a link a few days ago).
Turns out that some of the evaluators didn't really put a lot of effort into building the sources and sometimes were just incompetent:
No readme has been provided. There is a .hs file (google says it is a haskell file but no download for haskell available). Tried to run the above using javascript (sudo apt-get install nodejs) and got the following error: [...]

http://reproducibility.cs.arizona.edu/data/oopsla12_KangR12_build.txt

Edit: Shriram (see shared post) is trying to fix it.
 
For those of you who've been following this reproducibility discussion: we're now making it possible for the community to participate in a review of the artifacts for these papers. Please contribute!
1
1
Paul Brauner's profile photogeorge oloo's profile photoFlorian Loitsch's profile photoRay Cromwell's profile photo
6 comments
 
I guess I should blame the students and not the researchers...
Add a comment...
Work
Occupation
Software Engineer
Employment
  • Google
    2010 - present
  • Wootsoft
    2002 - 2003
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Currently
Aarhus
Previously
Montreal - Wien - Nice - Bad Vöslau
Education
  • Université Nice Sophia Antipolis
    Computer Science, 1998 - 2000
  • École supérieure en sciences informatiques
    Computer Science, 2000 - 2005
  • McGill University
    Computer Science, 2001 - 2002
  • BG & BRG Baden Biondekgasse
    1989 - 1997
Basic Information
Gender
Male