User talk:Milimetric (WMF)
I have replied to you on my talk page, in the future, you'll see "My new messages(1)" at the top left.
I'm sorry for the misunderstanding, I delete many userpages like this all the time. I had nothing against you and I never will have anything against you. Please take my apology.--Jasper Deng (talk) 04:26, 23 September 2012 (UTC)
Question about analytics
Hey Tobias, sorry, no I can't help much with that question. I'm more on the engineering side of things and don't have great analytical knowledge yet. I would try the analytics-l mailing list maybe. DAndreescu
- Aah! Sorry about that, totally unintentional, thanks for fixing.
I have two questions. It would be nice, if you could answer them.
- ) Why gets the reportcard content deleted after two years (e.g. rc_very_active_editors_count.csv)?
- ) Why are there small inconsistencies when you compare the data of an older date at that date which this data at today's date (Github)?
2012/06/01 11776 3260 1175 987 764 669 491 468 323 281 (earlier) 2012/06/01 11797 3217 1158 1001 755 666 485 458 319 279 (now)
- Hi Kopiersperre. Reportcard data is not deleted after two years, the data source is wikistats (for the example you asked about, see this page for data going back to 2008: http://stats.wikimedia.org/EN/SummaryEN.htm). The reason some graphs and datafiles in the reportcard don't have data going back further is because we migrated to this dashboard in 2012, and we didn't import too much historical data. As for the small inconsistencies, the reason is because pages that get deleted today have edits that sometimes happened a long time ago. The way we chose to report is to modify history and not count those edits after the page was deleted. There has been a lot of discussion about this, and I personally think it's important to see the data both ways. But the reportcard has to choose one way and stay consistent, and we think this is the most honest way to report the number of edits. Hope that helps :) Milimetric (WMF) (talk) 14:04, 11 September 2015 (UTC)
Please provide feedback on suggested improvements to the Code of Conduct
Thanks to everyone who’s helped work on the Code of Conduct so far.
These are important sections, so please take a look at the proposed changes. I apologize that this feedback arrived later than planned, but I think this will create a better document.
If you prefer to give your opinion privately, feedback via e-mail is welcome at conduct-discussionwikimedia.org.
Please participate in discussion on updated "Marginalized and underrepresented groups" text
Thanks for participating in the earlier discussion on Talk:Code of Conduct/Draft#Marginalized and underrepresented groups. There was some support and some issues were raised. I've attempted to propose a better draft to address the issues.
Wikistats 2.0 Design Preview
Hello! Thank you for leaving feedback on the Wikistats - Future per report page. We read it carefully and designed a set of wireframes that reflects your priorities. We’re now looking for feedback on the design, and we’d love your input. We have key questions that touch on different sections, and links for feedback here: Request for feedback/Round 1 (if you prefer email, write to
email@example.com and please include “Wikistats Design” in the subject). Please comment by Monday, February 13th so we can include your feedback and iterate. Thank you very much!
Your Wikistats 2.0 Team, --16:09, 3 February 2017 (UTC)
Wikistats 2.0 Prototype Feedback
Hello! Thank you for leaving feedback on the Wikistats - Future per report and Feedback/Round 1 pages. We used it to build an interactive prototype of the next version of Wikistats. We're now looking for feedback on the visual design, and we'd love your input. We have some follow-up questions but mostly an open discussion at Request for feedback/Round2 (if you prefer email, write to
firstname.lastname@example.org and please include "Wikistats Design" in the subject). Please comment by Monday, May 1st so we can include your feedback into the first release of Wikistats 2.0. Thank you very much!
Your Wikistats 2.0 Team, -13:38, 20 April 2017 (UTC)
Recent conversation at enWS about stats
Hi. Thx for the most recent prod about wikistats. The following conversation was one recently had by me at enWS with a fellow staff member with a series of questions of my trying to understand how and where a sister wiki of some size might look to understand from whence appear our readership. Rather than leave it away in a corner I am generously <ahem> sharing it.. I have no idea how a typical or sets of typical visitors arrive and progress through a site like English Wikisource. — billinghurst sDrewth 14:11, 20 April 2017 (UTC)
<paste> == Dumb question about visitors ==
Is there a means to find out how most of our link referrals arriving at our works? ie. Do we get direct referrals from Google and other search engines? Do we get links followed from WP? Do we have people following links once arrived from the main page? Or arriving and then running searches? Where shoudl I ask such naïve questions. At this point there I cannot find the basic tutorial for sites to understand their visitors. It all seems disorganised and all at levels above my stupidity. — billinghurst sDrewth 13:42, 10 April 2017 (UTC)
- Somebody knows the answer to this question, but unfortunately it isn't me. I heard something about referrals a year or two ago. Let me see what I can find out for you... Whatamidoing (WMF) (talk) 23:36, 10 April 2017 (UTC)
- Some links and some ideas for getting more information:
- http://discovery.wmflabs.org/external/ has some stats for a few big sources (e.g., Google Search). I don't know if it's possible to make that tool give you Wikisource-specific information.
- m:Research:Wikimedia referrer policy (by Dario in Analytics) is about outbound traffic, but it has a little information about/links to some work on inbound traffic.
- https://lists.wikimedia.org/mailman/listinfo/analytics is the best place to ask questions like this, and at least one similar discussion is alleged to be present in the archives. It's probably worth a search.
- But it's possible to file a Phab task to request a custom query from Analytics, to find out more relevant information, e.g., for a particular project. Such requests are granted on a "space available" basis, so it's probably worth spending a few minutes thinking about what you really want to know the most.
- Hobbitschuster or Ryan at Wikivoyage may have some ideas about what would be interesting to learn. Perhaps we could pull together a couple of Phab tasks for both of the projects. Whatamidoing (WMF) (talk) 20:07, 11 April 2017 (UTC)
- Some links and some ideas for getting more information:
- Thanks for the question. We do have this task https://phabricator.wikimedia.org/T112284 which we want to do sometime relatively soon. It would build a table where we could ask better questions about referrers. As for Wikistats 2, the first release won't have any such metrics because the data's won't be ready in time. We agree it's an interesting question, we just have many other things to work on before we get there. For an idea of the kind of work we do, check out https://phabricator.wikimedia.org/tag/analytics Milimetric (WMF) (talk) 16:55, 20 April 2017 (UTC)
Hi Dan, you must be very busy with Wikistats 2.0 (amongst other things) and my questions are not urgent. I'm a newby editor and I'm still trying to get a feel for the 'big picture' (Meta-Wiki Strategy, Plans and Projects). I searched for 'Analytics' and went around in circles for hours, looking at the current tools/reports and for places to suggest improvements to these. By some strange path, I finally discovered the Wikistats 2.0 Design project. This will deliver a huge improvement on what we have now! I'll put any feedback/suggestions I have on the project pages.
My background is that I've managed some online-communities/websites (using Google Analytics) and I spent some years as an'Application Portfolio Manager'. So the challenge of licensing/developing/customizing apps to better meet business and user needs is familiar to me. I also spent more than 10 years in Organizational Change and Communication to support business programs and projects that included new IT-systems to replace existing ones.
From my newby perspective, my impression is that most innovations (functional/technical) seem to be 'home-grown'. I get this impression from the large collection of individual tools and gadgets, the wishlists, etc. I'm not sure of the extent to which this true for Wikistats 2.0. I have no criticism of the project or its result. I'm trying to understand how projects like these get off the ground and what the major decisions are along the way. For example, the [page on Analytics software] has a whole list of alternatives. To what extent were these considered or used as the basis for Wikistats 2.0? Is this evaluation described anywhere? Is the question of whether to make/buy (or get for free) usually a standard part of the decision-making process for innovations? Thanks in advance! Kind Regards, Mikemorrell49 (talk) 17:28, 29 November 2017 (UTC)