Date: Wed, 26 Nov 2008 16:50:38 +0000 From: Christiano Moreschi moreschiwikiman@hotmail.co.uk Subject: Re: [Foundation-l] Trouble in Ireland
Amusing. I am strongly reminded of the fights over Macedonia and Liancourt Rocks. Except on those occasions nobody had any illusions that the participants on all side were any more than puerile time- wasters. Just because this lot have greater fluency in the English language doesn't mean that they're any more rational.
Thanks for not caring. That doesn't help us get to a solution. :-(
Michael Everson * http://www.evertype.com
Again, this is not the place for it. I see you've brought it up with Jimbo and elsewhere, so please stop canvassing a EN-WP dispute in places it doesn't belong.
David
2008/11/26 David Yellope sirfozzie@gmail.com:
Again, this is not the place for it. I see you've brought it up with Jimbo and elsewhere, so please stop canvassing a EN-WP dispute in places it doesn't belong.
I've been suggesting on wikien-l something like trying to work out, with actual numbers, what article readers would reasonably be expecting where. This led to suggesting an idea on wikitech-l on how to gather internal referer logs (where people go from one article to another, or what searches lead to what articles) in a way that won't breach privacy. Such information would be of great general use in seeing what readers actually want and do, rather than attempting to guess. So it's far from a fruitless discussion.
- d.
2008/11/26 David Gerard dgerard@gmail.com:
I've been suggesting on wikien-l something like trying to work out, with actual numbers, what article readers would reasonably be expecting where. This led to suggesting an idea on wikitech-l on how to gather internal referer logs (where people go from one article to another, or what searches lead to what articles) in a way that won't breach privacy. Such information would be of great general use in seeing what readers actually want and do, rather than attempting to guess. So it's far from a fruitless discussion.
You don't need to. You just rig it so that to go from one article to the other involves going via a redirect then measure the traffic on that redirect. This was I understand done with respect to some links on the main page at one point. Doesn't require dev involvement.
If we could get something that reads the searches and tabulates the most frequently not found articles, we could better target our account creation efforts.
________________________________ From: David Gerard dgerard@gmail.com To: Wikimedia Foundation Mailing List foundation-l@lists.wikimedia.org Sent: Wednesday, November 26, 2008 12:08:30 PM Subject: Re: [Foundation-l] Trouble in Ireland
2008/11/26 David Yellope sirfozzie@gmail.com:
Again, this is not the place for it. I see you've brought it up with Jimbo and elsewhere, so please stop canvassing a EN-WP dispute in places it doesn't belong.
I've been suggesting on wikien-l something like trying to work out, with actual numbers, what article readers would reasonably be expecting where. This led to suggesting an idea on wikitech-l on how to gather internal referer logs (where people go from one article to another, or what searches lead to what articles) in a way that won't breach privacy. Such information would be of great general use in seeing what readers actually want and do, rather than attempting to guess. So it's far from a fruitless discussion.
- d.
_______________________________________________ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
2008/11/27 Geoffrey Plourde geo.plrd@yahoo.com:
If we could get something that reads the searches and tabulates the most frequently not found articles, we could better target our account creation efforts.
Hence my suggestion on wikitech-l :-)
Logging referers as well as the name of the page hit wouldn't load the current logging significantly. Only logging referers that are in article space or Special:Search (with the search) would guard against most possible privacy problems. (Greg Maxwell noted that people may enter private identifying data into Special:Search which would then be logged. Possibly might warrant a warning note on Special:Search.)
What it needs now is someone writing code to log and to analyse and present the logs, and the code being approved ;-)
- d.
The thing that I am thinking of would only list things are searched over 150 times in a set period, no privacy concerns necessary
________________________________ From: David Gerard dgerard@gmail.com To: Wikimedia Foundation Mailing List foundation-l@lists.wikimedia.org Sent: Thursday, November 27, 2008 4:01:09 AM Subject: Re: [Foundation-l] Trouble in Ireland
2008/11/27 Geoffrey Plourde geo.plrd@yahoo.com:
If we could get something that reads the searches and tabulates the most frequently not found articles, we could better target our account creation efforts.
Hence my suggestion on wikitech-l :-)
Logging referers as well as the name of the page hit wouldn't load the current logging significantly. Only logging referers that are in article space or Special:Search (with the search) would guard against most possible privacy problems. (Greg Maxwell noted that people may enter private identifying data into Special:Search which would then be logged. Possibly might warrant a warning note on Special:Search.)
What it needs now is someone writing code to log and to analyse and present the logs, and the code being approved ;-)
- d.
_______________________________________________ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
wikimedia-l@lists.wikimedia.org