Amund Tveit
Create an account or sign in to get started
Filter: Blog delicious Google Reader Gmail/Google Talk LinkedIn Picasa Web Albums Twitter Facebook FriendFeed
FriendFeed
posted a message
“wonder how stackless python 2.6 compares to erlang (perf.wise).”
October 3 at 4:48 pm - Link
FriendFeed
posted a message
“Look forward to travel to Mountain View tomorrow :)”
October 1 at 4:08 am - Link
FriendFeed
posted a link
September 30 at 9:14 am - Link
I can't really see how they arrived at the final sentence; that MSFT hopes Fast will improve its websearch and ad business, though... - Vidar Larsen
FriendFeed
posted a message
“seven 00 000 000 000”
seven 00 000 000 000
September 29 at 11:47 am - Link
FriendFeed
posted a message
“primes”
primes
September 29 at 9:27 am - Link
s/are completely factored/can't be factored/gi; - Amund Tveit
Google Reader
shared an item on Google Reader
September 3 at 6:43 am - Link
FriendFeed
posted a link
September 25 at 12:15 am - Link
With case studies that illustrate how Hadoop solves specific problems, this book helps you: * Learn the Hadoop Distributed File System (HDFS), including ways to use its many APIs to transfer data * Write distributed computations with MapReduce, Hadoop's most vital component * Become familiar with Hadoop's data and IO building blocks for compression, data integrity, serialization, and persistence * Learn the common pitfalls and advanced features for writing real-world MapReduce programs * Design, build, and administer a dedicated Hadoop cluster * Use HBase, Hadoop's database for structured and semi-structured data - Amund Tveit
(rough cuts means early alpha) - Amund Tveit
FriendFeed
posted a link
September 25 at 4:31 am - Link
Place a rubber duck on your monitor and describe your problems to it. There's something magical about stating your problems aloud that makes the solution more clear. - Amund Tveit
FriendFeed
posted a link
September 25 at 4:36 am - Link
FriendFeed
posted a link
September 24 at 8:00 am - Link
Happy is a framework for writing map-reduce programs for Hadoop using Jython. It files off the sharp edges on Hadoop and makes writing map-reduce programs a breeze - Amund Tveit
Map-reduce jobs in Happy are defined by sub-classing happy.HappyJob and implementing a map(records, task) and reduce(key, values, task) function. Then you create an instance of the class, set the job parameters (such as inputs and outputs) and call run(). When you call run(), Happy serializes your job instance and copies it and all accompanying libraries out to the Hadoop cluster. Then for each task in the Hadoop job, your job instance is de-serialized and map or reduce is called. - Amund Tveit
Google Reader
shared an item on Google Reader
September 21 at 7:42 pm - Link
FriendFeed
posted a message
“hm, not entirely convinced of the new friendfeed ui, in particular 2 things: 1) distance between what used to be tabs and the search field/button, and 2) increased density of links to click on when adding a new message.”
September 19 at 9:19 am - Link
but then again, I am a trackpoint-person (read: thinkpad) and never used to using "mouseplate" on home laptop. - Amund Tveit
I use twhirl so whatever they do to the UI is invisible to me - Knut via twhirl
can you do friendfeed search using twhirl? - Amund Tveit
I can search from twhirl. Not sure if it is the exact same thing, but I think so. - Knut via twhirl
FriendFeed
posted a message
“poll”
September 8 at 11:59 am - Link
should I blog about: i) pragmatic rule induction, ii) a home refurnishing analogy for software engineering, or iii) my take on computational creativity? - Amund Tveit
i) Pragmatic rule induction sounds interesting. - DeWitt Clinton
I like pragmatic, so I'll go with i) as well. - Thomas Brox Røst
FriendFeed
posted a link
September 4 at 8:31 am - Link
Disco is an open-source implementation of the Map-Reduce framework for distributed computing. As the original framework, Disco supports parallel computations over large data sets on unreliable cluster of computers. - Amund Tveit
The Disco core is written in Erlang, a functional language that is designed for building robust fault-tolerant distributed applications. Users of Disco typically write jobs in Python, which makes it possible to express even complex algorithms or data processing tasks often only in tens of lines of code. This means that you can quickly write scripts to process massive amounts of data. - Amund Tveit
Yes, I saw that on github :) - Vidar Larsen
Vidar: tried it already? - Amund Tveit
amund: nope, I'm still going to try Hadoop with Pig first. But I'm starting to see a sharp increase in Erlang usage internally :) - Vidar Larsen
FriendFeed
posted a message
“Ran the SunSpider javascript benchmarks with 3 browsers, here are the results:”
September 2 at 2:18 pm - Link
browser 1: 5567ms - http://tinyurl.com/63879t - Amund Tveit
browser 2: 28616ms - http://tinyurl.com/5osz7c - Amund Tveit
browser 3: 10641ms - http://tinyurl.com/5goxnq - Amund Tveit
chrome, ie8, ff3 - Amund Tveit
FriendFeed
posted a message
“Today is my 3 year anniversary working for Google :)”
September 1 at 7:30 am - Link
recap from my first day - http://www.vldb2005.org/social... - Amund Tveit
Congrats! - Olivier Tharan
Congratulations! - Robert Konigsberg
Coming up on my two-year next week. Time goes by fast. - Steve Weis
FriendFeed
posted a message
“WALL-E looks like a mechnical E.T.”
September 1 at 12:28 pm - Link
that can't be a coincidence. - Amund Tveit
FriendFeed
posted a link
August 30 at 8:12 pm - Link
Here, we present a computational model that learns structures of many different forms and that discovers which form is best for a given dataset. The model makes probabilistic inferences over a space of graph grammars representing trees, linear orders, multidimensional spaces, rings, dominance hierarchies, cliques, and other forms and successfully discovers the underlying structure of a variety of physical, biological, and social domains. Our approach brings structure learning methods closer to human abilities and may lead to a deeper computational understanding of cognitive development. - Amund Tveit
Google Reader
shared an item on Google Reader
August 30 at 6:45 am - Link
Google Reader
shared an item on Google Reader
August 27 at 8:09 am - Link
Scala has Erlang inspired Actors if you want to stay on the jvm. Wonder if that is enough to realize the potential of future multi-core architectures... I'm currently working with both :) - Vidar Larsen
Erlang and Scala are somewhere on my TO-Look-at-list, but not there yet. I am still a fan of stackless python - http://amundtveit.info/publica... :) - Amund Tveit
Vidar: may I ask what you are using it for? :) - Amund Tveit
@Amund: Sure :) Currently I'm using Erlang in a distributed, robust state aggregation framework; not yet production ready. - Vidar Larsen
@Amund(2), and Scala I'm using as a more convenient implementation language for a RESTful servlet API based on Jersey/JSR311. The syntax is much terser than java for bean-like structures, and the built-in xml syntax eliminates the need for a templating layer. - Vidar Larsen
FriendFeed
posted a link
August 26 at 7:10 am - Link
FriendFeed
posted a link
August 24 at 12:26 pm - via Bookmarklet - Link
interesting (early phase) research field. - Amund Tveit via Bookmarklet
Google Reader
shared two items on Google Reader
August 23 at 12:24 pm - Link
August 23 at 12:24 pm - Link
Tip: Now you can add FriendFeed to your blog with our new customizable FriendFeed widgets!

Other ways to read this feed:Feed reader