On Thu, Apr 10, 2014 at 7:05 PM, Henrik Abelsson <henrik@abelsson.com> wrote:
Hi everyone,
So, quick update: I've received the server installed it and set up an instance of the code on and sent it on my hosting provider, and they've received it and are hooking it up. I'll migrate over stats.grok.se to the new server asap, hopefully over the weekend.
Hopefully it will be much faster for users, and GLAM usage in particular. I'm also feeling pretty motivated to code new features after you guys, and Toby in particular, have been so generous as to provide the stats service with hardware. So, I'm taking suggestions - anything in particular you'd like to see implemented on stats.grok.se first?As an end user who manually datamines a lot of stuff for reports I write, my wish list would be to be able to get an open office document that would let me generate a report like https://en.wikinews.org/wiki/File:Spanish_competitors_at_the_IPC_Athletics_World_Championships.pdf and https://en.wikinews.org/wiki/File:IPC_NorAmCup.pdf . Some of these tools exist in isolation or I just do not know about them. :-/ https://toolserver.org/~magnus/ts2/treeviews/ and https://tools.wmflabs.org/glamtools/glamorous.php and http://tools.wmflabs.org/glamtools/baglama.php come to mind.Desirable data is by day page views (for event tracking and understanding viewing patterns for content around a specific category) that can be looked at against the context of social media linking (total links to content on a daily basis, who is linking to it, how many followers they have or how many views the linked to content gets on a native site). It would also be useful to get some of this data mapped against total daily edits alongside daily views to get a better idea if getting people to collaborate on an article results in increased views. The total number of images added to a specific category during a certain timeframe, the number of uses of those images and the number of page views to images where those articles were used. Number of articles linking to a certain domain or URL across multiple language projects and page views to articles containing links to those. This information being easy to combine and tabulate more easily into one document for the purposes of explaining broader patterns would be useful.Sincerely,Laura Hale--twitter: purplepopple
_______________________________________________
Analytics mailing list
Analytics@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/analytics