Tuesday, May 20, 2014

Data, Data Everywhere The Need for BIG Privacy in a World of Big Data by Ann Cavoukian #FOCI14

Ann Cavoukian, Ph.D., is the Information and Privacy Commissioner of Ontario, Canada. This  morning, she gave a talk entitles "Data, Data Everywhere: The Need for BIG Privacy in a World of Big Data." Given that I love all things related to privacy, ethics, and standards, this talk was of great interest to me. Here are some of the key points that Ann addressed.
  • - big data and privacy are complementary interests
  • - her take, "privacy by design" is a win win proposition
  • - www.privacybydesign.ca
  • - privacy = personal control, freedom of choice, informational self-determination, context is key
  • - in 2010, this landmark resolution was passed to preserve the future of privacy, and has been translated into 36 languages because people are so desperate for this information
  • - the essence of it is to change the emphasis from a win-lose model to a win-win model, replace ‘vs’ with ‘and’
  • - you must address privacy at the beginning of a program, embed it into the code at the beginning
  • - 7 principles -
    1. 1. be proactive not reactive, prevention not remedial
    2. 2. default condition needs to be privacy
    3. 3. privacy embedded into design
    4. 4. full functionality, positive sum not zero sum
    5. 5. end to end security, full lifecycle protection, from the outset, from collection to destruction at the end
    6. 6. visibility and transparency, keep it open, tell customers what you’re doing, don’t let them learn afterwar
    7. 7. respect for use privacy, keep it user centric
  • - Big data will rule the world – during the honeymoon phase, everything else must step aside, forget causality, correlation is enough
  • - Then the honeymoon phase ends – found data… digital exhaust of web searches, credit card payments, mobiles pinging the nearest phone mast; these datasets are cheap to collect but they are messy and collected for disparate purposes
  • - Big data is now in the trough of disillusionment
  • - Google flu trends used to work and now doesn’t because Google engineers weren’t interested in context but rather selecting statistical patterns in the data – correlation over causation, a common assumption in big data analysis, imputed causality which is incorrect
  • - MIT professor Alex Pentland has proposed a New Deal on Data – individuals to own their data and control how it is used and distributed
  • - data problems don’t disappear just because you are working with big data instead of small data, you can’t just forget about things like data sampling
  • - Forget big data, what is needed is good data
  • - data analytics on context free data will only yield correlations, if you add context, then you might be able to impute causality
  • - once businesses have amassed the personal information, it can be hard if not impossible for individuals to know how it will be used in the future – “A long way to privacy safeguards” New York Times Editorial
  • - people now have to resign when data breaches happen, you must address them at the beginning
  • - privacy should be treated as a business issue, not a compliance issue. gain a competitive advantage by claiming privacy, lead with it
  • - proactive costs money but reactive costs lawsuits, brand damage, loss of trust, loss of consumer confidence
  • - privacy drives innovation and creativity, privacy is a sustainable competitive advantage

Annie Pettit, PhD is the Chief Research Officer at Peanut Labs, a company specializing in self-serve panel sample. Annie is a methodologist focused on data quality, listening research, and survey methods. She won Best Methodological Paper at Esomar 2013, and the 2011 AMA David K. Hardin Award. Annie tweets at @LoveStats and can be reached at annie@peanutlabs.com.

No comments: