| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
|---|---|---|---|---|---|---|
| 1 | 2 | |||||
| 3 | 4 |
5 |
6 |
7 | 8 |
9 |
| 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| 17 | 18 | 19 |
20 | 21 | 22 | 23 |
24 |
25 |
26 |
27 | 28 |
29 |
Editor Login
Convener in chief:
David Lazer
(Methodology, Networked Governance)
Editors:
Stanley Wasserman
(Current Trends, Methodology, Social Networks)
Allan Friedman
(Simulations)
Nathan Eagle
(Technology, Social Computing, Powerlaws, Current Trends)
Ben Waber
(Technology, Social Computing) Thomas Langenberg
(Technology, Social Computing, Social Networks, Current Trends)
Ines Mergel
(Knowledge Sharing, Social Computing, Social Software, Current Trends)
Brian Rubineau
(Social Dynamics, Societal Networks, Simulations)
Maria Binz-Scharf
(Qualitative Methodology, Knowledge Sharing, eGovernment)
Jeff Boase
(Technology, Societal networks)
Alexander Schellong
(Admin, eGovernment, Citizen Relationship Management)
Categories
Business
Citizen Relationship Management
Commentary
Complexity
Current Trends
Events/Announcements
Innovation
Internet
Knowledge Sharing
Methodology
Networked Governance
Networks in political science
PNG Working Paper Series
Powerlaws
Reviews
SN in the news
Simulations
Small Worlds
Small group networks
Social Capital
Social Computing
Social Finance
Social Networking Platforms
Social Psychology
Social Software
Societal Networks
Technology
eGovernment
Archives
- February 2008
- January 2008
- December 2007
- November 2007
- October 2007
- September 2007
- August 2007
- July 2007
- June 2007
- May 2007
- April 2007
- March 2007
- February 2007
- January 2007
- December 2006
- November 2006
- October 2006
- September 2006
- August 2006
- July 2006
- June 2006
- May 2006
- April 2006
- March 2006
- February 2006
- January 2006
- December 2005
- November 2005
Recent Entries
- Interview: Thorsten Jacobi on the current state and trends in social software
- Online Social Network User Fatigue?
- Second Call for Papers / Funding for doctoral student training for Harvard conference on networks
- Obama, techology, and social networks
- Tapping on the wisdom of the crowd: Social network analysis software tools on Wikipedia
Recent Comments
- Re: Retrospective vs prospective data analysis
By: David Lazer
Date: Jan 30, 2006 11:17 PM - Re: Retrospective vs prospective data analysis
By: allan
Date: Jan 30, 2006 5:52 PM
Notification
« Pew Social Ties Report | Main | Follow up: Google bombs and the autonomy of search engine vendors »
30 January 2006
Retrospective vs prospective data analysis
Before I leave the NSA thread, one more thought about the utility of the data mining phone log data, recording phone calls, etc. It is important to note that pattern recognition is only as powerful as the ancillary intelligence that the government would have to complement the phone data. Obviously, in the absence of any indications of who is high risk, the signal in the data is infinitesmal.
A corollary to this is that pattern recognition of today's data will be much more effective in the future-- because the amount of complementary information will only go up. We may have no reason to pay attention to Joe, except that he was 3 degrees removed from someone in some terrorist's phonebook. However, in the future we may receive solid intelligence that Joe is a terrorist. At that point, it would be a pretty good idea to go through all of the data you collected about Joe-- if you have it sitting in a hard drive somewhere. If you haven't been recording, and collecting information from the switches, etc, it may be impossible to reconstruct afterwards. That is, "retrospective" pattern analysis is likely to be much more effective than prospective pattern analysis. Of course, this in turn points to a strategy of collecting data now and asking questions later, which again brings us back to the issues around privacy and collateral usage in aces.
Posted by David Lazer at January 30, 2006 2:29 PM
-
.
- Business (2) .
- Citizen Relationship Management (13) .
- Commentary (5) .
- Complexity (2) .
- Current Trends (40) .
- eGovernment (15) .
- Events/Announcements (51) .
- Innovation (12) .
- Internet (14) .
- Knowledge Sharing (16) .
- Methodology (28) .
- Networked Governance (14) .
- Networks in political science (1) .
- PNG Working Paper Series (6) .
- Powerlaws (4) .
- Reviews (2) .
- Simulations (1) .
- Small group networks (3) .
- Small Worlds (2) .
- SN in the news (56) .
- Social Capital (2) .
- Social Computing (9) .
- Social Finance (1) .
- Social Networking Platforms (8) .
- Social Psychology (2) .
- Social Software (22) .
- Societal Networks (33) .
- Technology (27)
Comments
This is the primary motivation / justification for the closed circuit TV monitoring trend in the UK and elsewhere (some reports have the number of CCTV cameras in the UK at 4 million). The privacy balance is found in the policies that call for the destruction of tape after n days, during which relevant footage can be analyzed in the event of an incident. This helped British investigators learn a lot about the events immediately preceding the bombing last fall.
Long term data analysis had the advantage that it can be similarly governed by good policy. Databases can be encrypted to allow for access control and access auditing to protect against unwarranted use and mining. The crypto community has plenty of tools that allow limited access under different conditions.
Of course, absent any regulations, the data is stored decentralized, under whatever protections the (ISPs / phone companies / etc) have in place. Over the past few months we've seen various reactions to these companies turning over records to investigators. The advantage of retrospective analysis is that the arguments for secrecy are weaker, and therefore a better case can be made for, if not privacy protection, then at least transparency and due process.
Posted by: allan at January 30, 2006 5:52 PM
Notably, as Laszlo Barabasi pointed out to me today, last month the EU passed the Data Retention Directive, which directs a variety of information intermediaries (phone companies, ISPs) to save phone log data, etc, for a fixed period of time to allow for retrospective analysis of communication data by the government. I have not seen an analysis of what grounds are necessary for government access to these data.
Posted by: David Lazer at January 30, 2006 11:17 PM