Hi Franz,
I am CCing your email to the public Analytics list so that others may respond too. In short, those links don’t work because releasing the data was a mistake. You can see the Update from 2012/9/20 at the top of the post that explains why.
There isn’t an easy way to safely release user search queries without potentially compromising private user data. I don’t think we even collect this data at the moment. Since that blog post, our search architecture has changed, and the new one doesn’t have the ability to collect the queries easily. I believe https://phabricator.wikimedia.org/T103505 https://phabricator.wikimedia.org/T103505 is a ticket to start doing this, but I don’t know if there is a real timeline to make this happen.
-Andrew
On Jul 20, 2015, at 08:39, franz guenthner fguenthner@gmail.com wrote:
Dear Andrew Otto,
I have a question about the announcement concerning search logs on this page:http://blog.wikimedia.org/2012/09/19/what-are-readers-looking-for-wikipedia-... http://blog.wikimedia.org/2012/09/19/what-are-readers-looking-for-wikipedia-search-data-now-available/. The links given there don't seem to work.
Have you been able to make this service accessible ?
I am working on a new search device for Wikipedia where as much of the searchable information (at the fact level) as possible would be presented in an autosuggest mode; it would be helpful to have an overview of the query complexity of queries to Wikipedia. If you know of any other sources for Wikipedia query logs any information would be highly apprecated.
Thank you in advance for your hellp.
Prof. Franz Guenthner University of Munich
Heyo, search analyst here.
A big +1 to Andrew's statement. To my knowledge there's no timeline on that ticket and I would expect Search to be involved in those discussions. Since we haven't.. ;p.
Depending on how you're measuring query complexity or what you're looking for it may be possible to perform that analysis at our end without compromising or releasing user data. Feel free to reach out to me offlist if that is of interest.
On 21 July 2015 at 12:36, Andrew Otto aotto@wikimedia.org wrote:
Hi Franz,
I am CCing your email to the public Analytics list so that others may respond too. In short, those links don’t work because releasing the data was a mistake. You can see the Update from 2012/9/20 at the top of the post that explains why.
There isn’t an easy way to safely release user search queries without potentially compromising private user data. I don’t think we even collect this data at the moment. Since that blog post, our search architecture has changed, and the new one doesn’t have the ability to collect the queries easily. I believe https://phabricator.wikimedia.org/T103505 is a ticket to start doing this, but I don’t know if there is a real timeline to make this happen.
-Andrew
On Jul 20, 2015, at 08:39, franz guenthner fguenthner@gmail.com wrote:
Dear Andrew Otto,
I have a question about the announcement concerning search logs on this page:http://blog.wikimedia.org/2012/09/19/what-are-readers-looking-for-wikipedia-.... The links given there don't seem to work.
Have you been able to make this service accessible ?
I am working on a new search device for Wikipedia where as much of the searchable information (at the fact level) as possible would be presented in an autosuggest mode; it would be helpful to have an overview of the query complexity of queries to Wikipedia. If you know of any other sources for Wikipedia query logs any information would be highly apprecated.
Thank you in advance for your hellp.
Prof. Franz Guenthner University of Munich
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics