Hi all,
I've should have posted this earlier here.
Wikimedia France, Wikimedia UK, Wikimedia NL and Wikimedia CH funded a project to create better tools voor GLAMs to upload their collections, including metadata, to Wikimedia Commons. Simultaneously the project did research into what GLAMs need in terms of analytics of these collections to be able to justify sharing their collections on Wikipedia. The researched can be condensed to a single paragraph requirement:
'A GLAM analytics system is an open dashboard that shows all contributing institutions with monthly overviews of objects, their plays and their usage in Wiki projects. Ideally, the data is created by Kraken and presentation is handled through Limn with well-designed graphs providing exportable datasets.'
This is something that the Analytics team has adopted and will be working on in the coming period according to Diederik. They started working on importing commons data: https://mingle.corp.wikimedia.org/projects/analytics/cards/723 something that is probably important for other people on this list as well like Magnus.
That research has now been published here and includes high-level and low-level requirements:
Cheers,
Maarten
FWIW, I was told yesterday that no view count data exists within WMF, and the best we can hope for (at least in the short/medium run) is a WMF version of stats.grok.se. Which would do, of course.
On Mon, Jun 17, 2013 at 12:33 PM, Maarten Zeinstra [email protected] wrote:
Hi all,
I've should have posted this earlier here.
Wikimedia France, Wikimedia UK, Wikimedia NL and Wikimedia CH funded a project to create better tools voor GLAMs to upload their collections, including metadata, to Wikimedia Commons. Simultaneously the project did research into what GLAMs need in terms of analytics of these collections to be able to justify sharing their collections on Wikipedia. The researched can be condensed to a single paragraph requirement:
'A GLAM analytics system is an open dashboard that shows all contributing institutions with monthly overviews of objects, their plays and their usage in Wiki projects. Ideally, the data is created by Kraken and presentation is handled through Limn with well-designed graphs providing exportable datasets.'
This is something that the Analytics team has adopted and will be working on in the coming period according to Diederik. They started working on importing commons data: https://mingle.corp.wikimedia.org/projects/analytics/cards/723 something that is probably important for other people on this list as well like Magnus.
That research has now been published here and includes high-level and low-level requirements:
http://pro.europeana.eu/web/guest/pro-blog/-/blogs/europeana-glamtools-publi...http://pro.europeana.eu/web/guest/pro-blog/-/blogs/europeana-glamtools-publishes-wikimedia-recommendations?_33_redirect=http://pro.europeana.eu/web/guest/home?p_p_id=115&p_p_lifecycle=0&p_p_state=normal&p_p_mode=view&p_p_col_id=column-2&p_p_col_pos=4&p_p_col_count=6
Cheers,
Maarten *
-- Kennisland | www.kennisland.nl | t +31205756720 | m +31643053919 | @mzeinstra
Analytics mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/analytics
I saw your post. The reaction by the analytics team seems strange to me as it seems to conflict with what the team said to me at the Wikimedia Hackathon in Amsterdam.
It is true that they have not started log streams from 'upload', where view counts can be measured, but I was told that they will make including that a higher priority.
@Diederik, can you talk a little about the status of this?
CHeers,
Maarten
To make sure this is not overlooked, as I wrote earlier: we have view count data in daily/monthly aggregates.
Monthly aggregates come in two versions, with and without hourly granularity.
So there is more than grok.se, and in a leaner faster to process format, just not real time and not via an API.
http://dumps.wikimedia.org/other/pagecounts-ez/merged/
Erik Zachte
From: [email protected] [mailto:[email protected]] On Behalf Of Magnus Manske Sent: Monday, June 17, 2013 1:37 PM To: A mailing list for the Analytics Team at WMF and everybody who has an interest in Wikipedia and analytics. Subject: Re: [Analytics] GLAMTools gives recommendations to Wikimedia Analytics team
FWIW, I was told yesterday that no view count data exists within WMF, and the best we can hope for (at least in the short/medium run) is a WMF version of stats.grok.se. Which would do, of course.
On Mon, Jun 17, 2013 at 12:33 PM, Maarten Zeinstra [email protected] wrote:
Hi all,
I've should have posted this earlier here.
Wikimedia France, Wikimedia UK, Wikimedia NL and Wikimedia CH funded a project to create better tools voor GLAMs to upload their collections, including metadata, to Wikimedia Commons. Simultaneously the project did research into what GLAMs need in terms of analytics of these collections to be able to justify sharing their collections on Wikipedia. The researched can be condensed to a single paragraph requirement:
'A GLAM analytics system is an open dashboard that shows all contributing institutions with monthly overviews of objects, their plays and their usage in Wiki projects. Ideally, the data is created by Kraken and presentation is handled through Limn with well-designed graphs providing exportable datasets.'
This is something that the Analytics team has adopted and will be working on in the coming period according to Diederik. They started working on importing commons data: https://mingle.corp.wikimedia.org/projects/analytics/cards/723 something that is probably important for other people on this list as well like Magnus.
That research has now been published here and includes high-level and low-level requirements:
http://pro.europeana.eu/web/guest/pro-blog/-/blogs/europeana-glamtools-publi... http://pro.europeana.eu/web/guest/pro-blog/-/blogs/europeana-glamtools-publishes-wikimedia-recommendations?_33_redirect=http://pro.europeana.eu/web/guest/home?p_p_id=115&p_p_lifecycle=0&p_p_state=normal&p_p_mode=view&p_p_col_id=column-2&p_p_col_pos=4&p_p_col_count=6
Cheers,
Maarten
Interesting!
Could that, by any chance, be exposed on tools labs at /public/datasets/public/other/
That might help a lot with the monthly updates!
On Mon, Jun 17, 2013 at 6:54 PM, Erik Zachte [email protected] wrote:
To make sure this is not overlooked, as I wrote earlier: we have view count data in daily/monthly aggregates. ****
Monthly aggregates come in two versions, with and without hourly granularity.****
So there is more than grok.se, and in a leaner faster to process format, just not real time and not via an API. ****
http://dumps.wikimedia.org/other/pagecounts-ez/merged/****
Erik Zachte****
*From:* [email protected] [mailto: [email protected]] *On Behalf Of *Magnus Manske *Sent:* Monday, June 17, 2013 1:37 PM *To:* A mailing list for the Analytics Team at WMF and everybody who has an interest in Wikipedia and analytics. *Subject:* Re: [Analytics] GLAMTools gives recommendations to Wikimedia Analytics team****
FWIW, I was told yesterday that no view count data exists within WMF, and the best we can hope for (at least in the short/medium run) is a WMF version of stats.grok.se. Which would do, of course.****
On Mon, Jun 17, 2013 at 12:33 PM, Maarten Zeinstra [email protected] wrote:****
Hi all,****
I've should have posted this earlier here.****
Wikimedia France, Wikimedia UK, Wikimedia NL and Wikimedia CH funded a project to create better tools voor GLAMs to upload their collections, including metadata, to Wikimedia Commons. Simultaneously the project did research into what GLAMs need in terms of analytics of these collections to be able to justify sharing their collections on Wikipedia. The researched can be condensed to a single paragraph requirement:****
'A GLAM analytics system is an open dashboard that shows all contributing institutions with monthly overviews of objects, their plays and their usage in Wiki projects. Ideally, the data is created by Kraken and presentation is handled through Limn with well-designed graphs providing exportable datasets.'****
This is something that the Analytics team has adopted and will be working on in the coming period according to Diederik. They started working on importing commons data: https://mingle.corp.wikimedia.org/projects/analytics/cards/723 something that is probably important for other people on this list as well like Magnus.****
That research has now been published here and includes high-level and low-level requirements: ****
http://pro.europeana.eu/web/guest/pro-blog/-/blogs/europeana-glamtools-publi...http://pro.europeana.eu/web/guest/pro-blog/-/blogs/europeana-glamtools-publishes-wikimedia-recommendations?_33_redirect=http://pro.europeana.eu/web/guest/home?p_p_id=115&p_p_lifecycle=0&p_p_state=normal&p_p_mode=view&p_p_col_id=column-2&p_p_col_pos=4&p_p_col_count=6
Cheers,****
Maarten****
-- **
Kennisland **
| www.kennisland.nl | t +31205756720 | m +31643053919 | @mzeinstra**
Analytics mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/analytics****
Analytics mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/analytics