Interesting study <span class="Apple-style-span" style="font-family: arial, sans-serif; font-size: 18px; border-collapse: collapse; ">Sohan. We could all benefit from such a study and I wish there was a way to collect data without privacy concerns.</span><div>

<font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><br></span></font></div><div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;">Unfortunately, it isn&#39;t as easy to anonymize data. Often when you think you&#39;ve done it right someone comes along and fairly easily de-anonymizes it. Check out the links below (they all refer to the same Netflix dataset but with varying degrees of detail).</span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><br></span></font></div><div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;">- When 2+2 Equals A Privacy Question (New York Times): <a href="http://www.nytimes.com/2009/10/18/business/18stream.html?_r=4">http://www.nytimes.com/2009/10/18/business/18stream.html?_r=4</a> </span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><br></span></font></div><div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;">- How To Break Anonymity of the Netflix Prize Dataset:</span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><a href="http://www.cs.utexas.edu/~shmat/shmat_oak08netflix.pdf">http://www.cs.utexas.edu/~shmat/shmat_oak08netflix.pdf</a></span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><br></span></font></div><div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;">- AOL had a major fiasco a few years back when it also released &quot;anonymized&quot; search records of its users: <a href="http://en.wikipedia.org/wiki/AOL_search_data_scandal">http://en.wikipedia.org/wiki/AOL_search_data_scandal</a></span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><br></span></font></div><div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;">- New Study Looks at Re-Identification Risks of Hospital Pharmacy Prescription Records</span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><a href="http://www.cheo.on.ca/english/3100-09-10-14.shtml">http://www.cheo.on.ca/english/3100-09-10-14.shtml</a> (The journal article referenced is here: <a href="http://www.cheo.on.ca/english/pdf/news_CJHP-El-Emam.pdf">http://www.cheo.on.ca/english/pdf/news_CJHP-El-Emam.pdf</a>)</span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><br></span></font></div><div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;">I don&#39;t know you can get meaningful data for your study without running into significant privacy risks. These guys (Artificial Intelligence for Development <a href="http://ai-d.org/">http://ai-d.org/</a>) seem to have a large dataset of mobile phone data from Kenya. I don&#39;t know how they got it and how they deal with privacy issues. They may be of some use to you.</span></font></div>

<div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;"><br></span></font></div><div><font class="Apple-style-span" face="arial, sans-serif" size="5"><span class="Apple-style-span" style="border-collapse: collapse; font-size: 18px;">Saidi</span></font></div>

<div><font class="Apple-style-span" face="&#39;Times New Roman&#39;"><span class="Apple-style-span" style="font-size: medium;"><br></span></font></div><div>On Thu, Nov 19, 2009 at 2:23 PM, <a href="mailto:rsohan@gmail.com">rsohan@gmail.com</a> <span dir="ltr">&lt;<a href="mailto:rsohan@gmail.com">rsohan@gmail.com</a>&gt;</span> wrote:</div>

<div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div class="gmail_quote"><div class="im">On Thu, Nov 19, 2009 at 5:33 PM, Jared Koyier <span dir="ltr">&lt;<a href="mailto:jaredkoyier@gmail.com" target="_blank">jaredkoyier@gmail.com</a>&gt;</span> wrote:<br>

<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I tend to think logs are a little raw and any ISP that keeps such might consider them to be client confidential.<br></blockquote><div><br></div></div><div>Raw? What do you mean? The logs are the definitive source on what actually happened.</div>


<div><br></div><div>As to the confidentiality argument -- I agree.  However, I point out it is exceedingly trivial to anonymize logs without losing any contextual information; and I would be happy to provide scripts to anyone who doesn&#39;t want to do it themselves.</div>


<div><br></div></div>
<br>_______________________________________________<br>
Skunkworks mailing list<br>
<a href="mailto:Skunkworks@lists.my.co.ke">Skunkworks@lists.my.co.ke</a><br>
<a href="http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks" target="_blank">http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks</a><br>
------------<br>
Skunkworks Rules<br>
<a href="http://my.co.ke/phpbb/viewtopic.php?f=24&amp;t=94" target="_blank">http://my.co.ke/phpbb/viewtopic.php?f=24&amp;t=94</a><br>
------------<br>
Other services @ <a href="http://my.co.ke" target="_blank">http://my.co.ke</a><br>
Other lists<br>
-------------<br>
Announce: <a href="http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks-announce" target="_blank">http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks-announce</a><br>
Science:  <a href="http://lists.my.co.ke/cgi-bin/mailman/listinfo/science" target="_blank">http://lists.my.co.ke/cgi-bin/mailman/listinfo/science</a><br>
kazi:     <a href="http://lists.my.co.ke/cgi-bin/mailman/admin/kazi/general" target="_blank">http://lists.my.co.ke/cgi-bin/mailman/admin/kazi/general</a><br></blockquote></div><br></div>