there are a lot of resources based in the US that may need to be distributed on a wider base including personal/private data already collected by the WMF
For editors, but not readers. On November 8 a top foundation official tweeted that the Foundation would not store personally identifying information in reader logs. Samuel Klein retweeted the statement, but the tweet has since been deleted.
According to this recently created document:
https://wikitech.wikimedia.org/wiki/Analytics/Data/Webrequest/RawIPUsage
the Foundation keeps full reader logs with IP addresses so that "in order to allow buffer time to be able to rerun metrics due to bugs or data issues." Although the analytics team has been responsive to many of my questions, nobody has been able to give any reasons why it is more important to "rerun metrics" than to protect the privacy of readers, after repeated requests for any such reasons, or even the basis upon such a decision about the trade-off is made. Furthermore, while that document states that it's "imperative for Ops to be able to examine raw IP addresses" but after at least half a dozen requests, nobody in analytics or Ops has said whether Ops is able to use the request logs with IP addresses but not article names.