A Lighter Moment About Randomness...

elections, standards, hacks, friends, research

(This is funny, but you have to let me build it up with context...)

In elections, especially in election audits, we often have to produce random numbers that an observer can prove to themselves came from a true source of randomness. Obviously, this nixes computer-generated random numbers as one can't prove that they came from a true source of randomness (and often computer-generated random numbers are from pseudo-random number generation algorithms).

Arel Cordero, David Wagner and David Dill published a great paper a while back proposing the use of sets of 10-sided dice to produce decimal random numbers. Great paper.

In addition to dice, there are other random processes, such as using a tumbler with ten ping pong balls numbered from 0-9. The RAND corporation also publishes a book of random numbers, which you can buy: "A Million Random Digits with 100,000 Normal Deviates".

So, where's the funny in all of this? Check out the reviews on Amazon for the RAND book. They're drop-dead hilarious. My favorite is the first one:

A truly amazing genre-breaking work of art unlike any that has ever been or ever will. I was captivated from the moment I opened the cover until the extremely suspenseful moment I turned the last page. With that said, I was a little disappointed that 71602 was knocked off by 92937 just as the plot was unfolding, but the arrival of 96240 really got my blood pumping and I just couldn't put the book down from that moment on.

I am so glad that Amazon.com is offering the "Search Inside This Book" option for this book so that it can be enjoyed by countless other avid readers who otherwise may not have come across it. I wait, impatiently, for the audio CD version of this fine book.

Although this one is pretty funny too:

There are 10 kinds of people in this world: Those who understand binary digits and those who don't. The author of this book clearly falls into the latter category.

I will give the author some credit though, for capturing the 'riverrun', Joycean stream-of-conciousness in a new form. As other reviewers have commented, once you finish the book you want to start again at the beginning.

and one last one:

All erudite readers know that it reads much better in the original Klingon.

Carl Malamud, "(Re-)Defining the Public Domain"

system, elections, copyright, open source, patents, berkeley, photos, research, policy, usability, legal, podcasts, education

Picture of Carl Malamud speaking at UC Berkeley Carl Malamud gave today's Distinguished Lecture at the UC Berkeley School of Information entitled, "(Re-)Defining the Public Domain". Audio of the event is here and some pictures are here.

It was a fascinating talk -- almost a performance of sorts -- where Malamud described efforts to increase the public domain along the three branches of our government. His talk incorporated something like 31 individual videos and, unfortunately, the audio provided here can do only so much justice to that content.

Here is the abstract of Carl's talk:

While much of the focus on intellectual property goes to battles over copyright or patents, we should not forget that a large proportion of such material is not property at all. The public domain is available for all to use. Of particular interest for the public domain is the U.S. government, all of whose work is available without restriction for all of us to use.

In this lecture, Carl Malamud explains the principles of the public domain with particular emphasis on the works of government. He will discuss how government often backs away from the clear principle of no property interests in order to maintain control, and how citizens can change that attitude through concrete actions. Malamud will use his own experience in forcing changes in government policy with numerous agencies to illustrate these general principles.

UPDATE [2007-10-22T09:25:04]: The audio linked to above has been updated to include the audio from each of the videos that Carl showed (thanks, Carl!). Also, you can watch a broadcast of the video from Carl's slides hosted by the Internet Archive here.

Mapping codes to levels for factors imported from SPSS data in R

hacks, open source, research, education

R, the R-project, is a great, free, statistical software package that can do many many things... however, some things can be a bit of a pain to figure out.

For example, if your collaborators have their data in SPSS and they give you a .sav file, you might be interested in knowing the mapping from codes (or values) to level names (semantic information) for categorical variables (called "factors" in R). R treats the codes as unimportant, but you might want to know their original values so that you can better interoperate with your SPSS-using collaborators.

Fortunately, the read.spss function creates a structure called label.table that contains this mapping.

Here's an example of how to access this mapping... say you have an SPSS .sav file called data.sav and the categorical variable in question is INCOME.

> #Load the foreign library, read in sav file
> library(foreign)
> dat <- read.spss("data.sav")
> 
> #Attach the object to access its variables
> attach(dat)
> 
> #Save the mapping for `INCOME` to `map`
> map <- attr(dat,"label.table")$INCOME
> 
> #Access the codes with as.integer()
> as.integer(map)
 [1] 97 96 95 94 93 92 91 90 87 86 85 84 83 82 81 80 17 16 15 14 
     13 12 11 10 9 8  7  6  5  4  3  2  1
> 
> #Access the labels with names()
> names(map)
 [1] "RF - 75-150K" "RF - gt 75K"  "RF - 35-75K"  "RF - gt 35K"  
     "RF - 15-35K"  "RF - lt 35K"  "RF - lt 15K"  "Refused"
     "DK - 75-150K" "DK - gt 75K"  "DK - 35-75K"  "DK - gt 35K"  
     "DK - 15-35K"  "DK - lt 35K"  "DK - lt 15K"  "Don't know"
     "Over 150k"    "Exactly 150K" "100-150k"     "Exactly 100K" 
     "75-100K"      "Exactly 75K"  "50-75K"       "Exactly 50K"
     "35-50K"       "Exactly 35K"  "25-35K"       "Exactly 25K"
     "15-25K"       "Exactly 15K"  "10-15K"       "Exactly 10K"  
     "Under 10K"   

Kapor: Disruptive Innovations I Have Known and Loved

open source, berkeley, friends, policy, podcasts

Link: http://groups.sims.berkeley.edu/podcast/audio/Mitch_Kapor_UCiSchool_03Oct2007.mp3

Picture of Mitch Kapor in Hawaiian shirt Mitch Kapor gave the second UC Berkeley School of Information Distinguished Lecture of Fall 2007 today entitled, "Disruptive Innovations I Have Known and Loved - Part 1: The Personal Computer". Here is the audio (mp3) and here is the abstract:

A comparative look at the origins, development, and impact of major information technology platforms of the past three decades from the perspective of a leading entrepreneur and software designer who has played a major role in each of them.

Mitch will be givng two more lectures -- part II and part II in this series -- on November 14 and 28.

Two upcoming hip-hop shows...

music, feeds, berkeley

Dead Prez is playing Mezzanine tomorrow night... and The Federation is playing for free at Rasputin's in Berkeley at 5pm on Tuesday.

(I keep a list of upcoming shows that I find of interest on my "Killer Shows..." page... there's an RSS feed for it here.)

Contact / Help. (cc) 2021 by Joseph Hall. blog software / hosting.
Design & icons by N.Design Studio. Skin by Tender Feelings / Evo Factory.
And a few words about the structure of the eye . Everyone " retina ". Especially often we hear it buy clomid online in the phrase " retinal detachment ." So what is the retina ? This - the front edge of the brain, the most distant from the brain part of the visual analyzer. The retina receives light first , processes and transforms light energy into irritation - a signal that encodes all the information about what the eye sees . The retina is very complex and in their structure and function . Its structure resembles the structure of the cerebral cortex. The shell of the retina is very thin - about 0.14 mm.