Profile

Scrapbook photo 1
Scrapbook photo 2
Scrapbook photo 3
Scrapbook photo 4
Sai (saizai)
8,660 followers|1,990,620 views
AboutPostsPhotosYouTube

Stream

Sai (saizai)

Shared publicly  - 
 
+Alex Fink: "I feel like you're trying to lay an egg through my face."

/out of context, not that it'd help
7
Ed S's profile photo
Ed S
 
Too bizarre.
Add a comment...

Sai (saizai)

Shared publicly  - 
 
"NIST requests comments on an initial public draft report on NISTIR 8053, De-identification of personally Identifiable Information. This document describes terminology, process and procedures for the removal of personally identifiable information (PII) from a variety of electronic document types. 
 
Background:
This draft results from a NIST-initiated review of techniques that have been developed for the removal of personally identifiable information from digital documents. De-identification techniques are widely used to removal of personal information from data sets to protect the privacy of the individual data subjects. In recent years many concerns have been raised that de-identification techniques are themselves not sufficient to protect personal privacy, because information remains in the data set that makes it possible to re-identify data subjects. 
 
We are soliciting public comment for this initial draft to obtain feedback from experts in industry, academia and government that are familiar with de-identification techniques and their limitations. 
 
Comments will be reviewed and posted on the CSRC website. We expect to publish a final report based on this round of feedback. The publication will serve as a basis for future work in de-identification and privacy in general. 
 
Note to Reviewers:
NIST requests comments especially on the following: 
 
    • Is the terminology that is provided consistent with current usage?
    • Since this document is about de-identification techniques, to what extent should it discuss differential privacy?
    • To what extent should this document be broadened to include a discussion of statistical disclosure limitation techniques?
    • Should the glossary be expanded? If so, please suggest words, definitions, and appropriate citations?

Please send comments to draft-nistir-deidentify@nist.gov by May 15, 2015."

http://csrc.nist.gov/publications/drafts/nistir-8053/nistir_8053_draft.pdf
4
1
Amber Yust's profile photoSatyr Icon's profile photoSai (saizai)'s profile photo
7 comments
 
If you don't do it right, yes it is. I did some work on that with the CSS History Hack, as have others (eg iSecLab had a great full identification attack using social networks). Sweeney has some great papers on it with SSNs and other suffiicient-entropy identifiers (zip, dob, etc).

The point of proper anonymization is that it reduces the entropy such that there isn't enough to reverse, nor to add to other datasets.

See eg
https://www.eff.org/deeplinks/2009/09/what-information-personally-identifiable &
https://www.eff.org/deeplinks/2010/01/primer-information-theory-and-privacy
to understand how to fundamentally view these issues.
Add a comment...
 
Some preliminary playing with the Social Security Death Master File:  # unique last names appears to follow Zipf's law:
1076044 n>1
386954 n>9
65659 n>99
9638 n>999
1009 n>9999
49 n>99999

Still to do: SSN derivation from date/place of birth, lifespan correlation w/ name, place of birth, & date of birth, etc. But at least the DB is now only ~6.6GB (half data, half index) for 86M rows and not too horrible to query with not a lot of resources…
1
Ed S's profile photoSai (saizai)'s profile photo
62 comments
 
+Ed S I don't think we'll get solid answers without the CDC natality/mortality data to baseline against. It's in process…
Add a comment...

Sai (saizai)

Shared publicly  - 
7
3
Sai (saizai)'s profile photoShaun Griffith's profile photoDianne Hackborn's profile photoIan Petersen's profile photo
3 comments
 
+Kyle Parks Rap News is pretty consistently solid, both in quality of satire and lyrics.
Add a comment...

Sai (saizai)

Shared publicly  - 
 
For any DBAs in the audience… please help me avoid risking another 33M-row rollback. >.>
I have the following table, which stores the Social Security Death Master File database (~86M rows): CREATE TABLE `death_master_files` ( `id` int(11) NOT NULL AUTO_INCREMENT, `ssn_an` smallint...
3
Danial Hallock (Kysimir)'s profile photoIan Petersen's profile photoEd S's profile photoSai (saizai)'s profile photo
13 comments
 
+Ed S Registered deaths of people with SSNs assigned, which is not everyone.
Add a comment...
Have them in circles
8,660 people
PIROTECNIA ROGELIO RAMOS's profile photo
Carrie Welborn's profile photo
Natthawat Wongrat's profile photo
Ramon Holguin's profile photo
Zie Weaver (Zearen)'s profile photo
Jan van Steenbergen's profile photo
James Hull's profile photo
Brandon Green's profile photo
Pax Gethen's profile photo

Sai (saizai)
owner

Computer security  - 
 
 
"NIST requests comments on an initial public draft report on NISTIR 8053, De-identification of personally Identifiable Information. This document describes terminology, process and procedures for the removal of personally identifiable information (PII) from a variety of electronic document types. 
 
Background:
This draft results from a NIST-initiated review of techniques that have been developed for the removal of personally identifiable information from digital documents. De-identification techniques are widely used to removal of personal information from data sets to protect the privacy of the individual data subjects. In recent years many concerns have been raised that de-identification techniques are themselves not sufficient to protect personal privacy, because information remains in the data set that makes it possible to re-identify data subjects. 
 
We are soliciting public comment for this initial draft to obtain feedback from experts in industry, academia and government that are familiar with de-identification techniques and their limitations. 
 
Comments will be reviewed and posted on the CSRC website. We expect to publish a final report based on this round of feedback. The publication will serve as a basis for future work in de-identification and privacy in general. 
 
Note to Reviewers:
NIST requests comments especially on the following: 
 
    • Is the terminology that is provided consistent with current usage?
    • Since this document is about de-identification techniques, to what extent should it discuss differential privacy?
    • To what extent should this document be broadened to include a discussion of statistical disclosure limitation techniques?
    • Should the glossary be expanded? If so, please suggest words, definitions, and appropriate citations?

Please send comments to draft-nistir-deidentify@nist.gov by May 15, 2015."

http://csrc.nist.gov/publications/drafts/nistir-8053/nistir_8053_draft.pdf
7 comments on original post
2
Add a comment...

Sai (saizai)

Shared publicly  - 
 
Reading the documentation on some old, large datasets (1968 onwards) is amusing. E.g.:

"Machine used: IBM/3081/K
Language used: PL/1
File organization: One file, multiple reels
Recording mode: lBM/EBCDIC 8-bit"

Also, in 1978 they changed "Negro" to "Black" and "Legitimacy" to "Marital status"…
6
Sakari Maaranen's profile photoAndres Soolo's profile photoSai (saizai)'s profile photo
3 comments
 
+Sakari Maaranen Whoops. Fixed.
Add a comment...

Sai (saizai)

Shared publicly  - 
 
 
The discussion on a recent post of mine turned to sex work, its morals, its politics, and its law. A theme that came up repeatedly -- and rightly so -- is the importance of seeing any work from the perspective of the people who do it. +Joshua O'Madadhain found this extraordinary article in that category: written by a porn star, about his work and his life and how they intertwine.

I've known a number of people who work in this field, and I would say that Habib's description really rings true to everything I've ever heard anyone else say -- but few have written quite so powerfully about it.

(The link is entirely SFW, incidentally, unless the mere mention of the word "porn" frightens you.)
Why Do You Hate Porn Stars? What Is It About Us that You Don't Like?
104 comments on original post
6
Scott Wyant's profile photo
 
WoW! That's a powerful story. A bit long, but worth every second to read.
Add a comment...

Sai (saizai)
owner

Neuroscience  - 
 
I'm just going to leave this absolutely gobsmacking abstract here. It's always fascinating when some piece of neurological architecture has been clearly press-ganged into supporting some computationally related but conceptually dissimilar mechanism. 

In this case, it appears that the same architecture used for processing physical discomfort is also involved in processing violations of conceptual boundaries. From the abstract:

The meaning-maintenance model posits that any violation of expectations leads to an affective experience that motivates compensatory affirmation. We explore whether the neural mechanism that responds to meaning threats can be inhibited by acetaminophen, in the same way that acetaminophen inhibits physical pain or the distress caused by social rejection. In two studies, participants received either acetaminophen or a placebo and were provided with either an unsettling experience or a control experience. In Study 1, participants wrote about either their death or a control topic. In Study 2, participants watched either a surrealist film clip or a control film clip. In both studies, participants in the meaning-threat condition who had taken a placebo showed typical compensatory affirmations by becoming more punitive toward lawbreakers, whereas those who had taken acetaminophen, and those in the control conditions, did not.
14 comments on original post
2
1
Sai (saizai)'s profile photoRafael Espericueta's profile photoTrond Arild Tjøstheim's profile photo
3 comments
 
+Sai "In both studies, participants in
the meaning-threat condition who had taken a placebo showed typical compensatory affirmations by becoming more
punitive toward lawbreakers, whereas those who had taken acetaminophen, and those in the control conditions, did
not."   Well, one  simple hypothesis one could test is that conservatives would become slightly less conservative after a dose of acetaminophen. Right-wing authoritarian personalities tend to be more punitive toward lawbreakers.
Add a comment...

Sai (saizai)

Shared publicly  - 
 
... you may wish to consider the following information in re: your salary. You won't be surprised to learn that Edmundson is a "city" in St. Louis County, about a ten-minute drive from Ferguson (might take a lot longer, of...
30
4
Shival Wolf's profile photoMartin Espinoza's profile photoMichael Chui's profile photoPaul Bennett's profile photo
6 comments
 
At least with a traffic ticket there is often a vaguely visible and preventable offense.  Many of these burgs have offenses like "manner of walking on the street" which are "discretionary", meaning they occur when an officer thinks they do and that sets up the whole program of rising fines,  
btw, the state of Missouri is trying to eliminate some of these little municipalities but that bill is being circumvented by their voting to change their official status from "village" to "fourth class city" (an overestimate) (I live in a place that did just so vote.)
Add a comment...
People
Have them in circles
8,660 people
PIROTECNIA ROGELIO RAMOS's profile photo
Carrie Welborn's profile photo
Natthawat Wongrat's profile photo
Ramon Holguin's profile photo
Zie Weaver (Zearen)'s profile photo
Jan van Steenbergen's profile photo
James Hull's profile photo
Brandon Green's profile photo
Pax Gethen's profile photo
Work
Occupation
Syncretist
Story
Tagline
Syncretist
Introduction
Hi! My full name is Sai; my handle is saizai.

Right now, I'm primarily working on Make Your Laws (G+), a practical liquid democracy project. I also have CogSai (G+), a YouTube channel about cognitive science.

- I was twice suspended :-/

MYL disclaimer: things I post to my own stream are my own musings, not official positions on behalf of Make Your Laws. If I don't mention MYL, they're not even that, they're just my own personal political beliefs — something I do not mix with the non-partisan, systemic PoV I take with MYL. If you're reporting on MYL, please respect the difference; I do not want to be a public figure, and my personal political beliefs are not relevant.

If I know you outside G+, or if you have something interesting for us to talk about that hasn't come up in comments, please send me a private message. Unfortunately, I just get too many adds to track them. If I don't add you back, that's only a comment about my lack of mental bandwidth, not about you; I still enjoy good crunchy conversation, though.

About G+:

G+ games:
Mafia/Werewolf signup; games #1 #2 



I'm interested in lots and lots of things; this is only a small part of them. See my website for more: http://s.ai

[Last updated: ~2013]
Bragging rights
I'm mononymous; my full legal name is just "Sai".
Basic Information
Relationship
In an open relationship
Other names
saizai