How many page views does a given article get? - WikiEN-l - lists.wikimedia.org

List overview All Threads
Download

How many page views does a given article get?

Re: [WikiEN-l] What is happening...

Re: [WikiEN-l] Jimmy Wales in the...

David Gerard

5 Mar 2008 5 Mar '08

12:53 p.m.

http://stats.grok.se/ These are from new logs gathered from the squids. Enter an article and see! - d.

Reply

Show replies by date

David Gerard

5 Mar 5 Mar

1:03 p.m.

On 05/03/2008, David Gerard <dgerard(a)gmail.com> wrote:

http://stats.grok.se/ These are from new logs gathered from the squids. Enter an article and see!

BTW, the raw data is here: http://dammit.lt/wikistats/ - d.

Reply

Steve Bennett

2:46 p.m.

On 3/5/08, David Gerard <dgerard(a)gmail.com> wrote:

http://stats.grok.se/ These are from new logs gathered from the squids. Enter an article and see!

Very, very interesting. This tool has the potentially to seriously shape discussions about notability and the benefit of retaining certain articles. Going through the list of articles I started, I see a big range from less than five per day up to several hundred. I think my most popular article is [[Female urination device]]. Recently deceased people do well. Anything pop culture does well. Roads, parks, long forgotten historical figures do badly. Is it possible to get information about the source of the hits? In particular, I would like to exclude the noise of editors. When a Wikipedian looks at an article it's very different to when a random googler does so. Steve

Reply

Andrew Gray

2:53 p.m.

On 05/03/2008, Steve Bennett <stevagewp(a)gmail.com> wrote:

Is it possible to get information about the source of the hits? In particular, I would like to exclude the noise of editors. When a Wikipedian looks at an article it's very different to when a random googler does so.

(This question may betray a complete failure to comprehend our caching structure, but...) This information is taken from data served by the squids... don't logged-in users mostly bypass the squid servers in some way or another, or is that only when actively editing? -- - Andrew Gray andrew.gray(a)dunelm.org.uk

Reply

Thomas Dalton

3:27 p.m.

(This question may betray a complete failure to comprehend our caching structure, but...) This information is taken from data served by the squids... don't logged-in users mostly bypass the squid servers in some way or another, or is that only when actively editing?

As I understand it, they don't technically bypass the squids, but the squids pass their requests straight through to the main servers rather than returning cached pages. The squids should be able to gather stats on logged in users.

Reply

David Gerard

4:42 p.m.

On 05/03/2008, Thomas Dalton <thomas.dalton(a)gmail.com> wrote:

> (This question may betray a complete failure to comprehend our caching > structure, but...) > This information is taken from data served by the squids... don't > logged-in users mostly bypass the squid servers in some way or > another, or is that only when actively editing?

As I understand it, they don't technically bypass the squids, but the squids pass their requests straight through to the main servers rather than returning cached pages. The squids should be able to gather stats on logged in users.

For technical details of the Squid logging, you should probably ask on wikitech-l, where Domas can answer for everyone else as well :-) - d.

Reply

Relata Refero

6 Mar 6 Mar

3:22 p.m.

On Wed, Mar 5, 2008 at 8:16 PM, Steve Bennett <stevagewp(a)gmail.com> wrote:

On 3/5/08, David Gerard <dgerard(a)gmail.com> wrote:

http://stats.grok.se/ These are from new logs gathered from the squids. Enter an article and

see! Very, very interesting. This tool has the potentially to seriously shape discussions about notability and the benefit of retaining certain articles. Going through the list of articles I started, I see a big range from less than five per day up to several hundred. I think my most popular article is [[Female urination device]]. Recently deceased people do well. Anything pop culture does well. Roads, parks, long forgotten historical figures do badly. Steve

Not to sound pedantic about it, but "anything" pop culture? Even outdated pop culture? The point I make is that "long-forgotten historical figures" tend to be something people might actually turn to an encyclopaedia for for at least the foreseeable future, but pop culture is well, pop. Just a reminder for everyone that a month's snapshot isn't that helpful.... compare the 10-12 hits a day for Edward Grey, a relatively undistinguished Foreign Secretary of the distant past with, say, the 17 hits a day for "The Great and Powerful Turtle", a character from an ongoing, if declining, SF series.... RR

Reply

Bryan Derksen

9:28 p.m.

Relata Refero wrote:

Not to sound pedantic about it, but "anything" pop culture? Even outdated pop culture?

It's been some time since I checked the policies but notability is supposed to be permanent. If something was notable in the 1800s, then it's worthy of an article even if it isn't notable now. Otherwise most of our coverage of historical figures would be labeled "historycruft" and deleted and/or redirected to various list articles. Assuming, of course, there isn't some other double standard underlying the labeling of "cruft". :)

Reply

The Mangoe

7 Mar 7 Mar

11:59 p.m.

Saying that "notability is permanent" is a kind of wishful thinking, but it's hardly worth arguing. What I find more interesting is that one can neatly quantify the effect of various means of spotlighting articles. For the two articles I had that made "Did you know" in the period, the notice was worth 4-8 thousand views. On Thu, Mar 6, 2008 at 4:28 PM, Bryan Derksen <bryan.derksen(a)shaw.ca> wrote:

Relata Refero wrote:

Not to sound pedantic about it, but "anything" pop culture? Even outdated pop culture?

It's been some time since I checked the policies but notability is supposed to be permanent. If something was notable in the 1800s, then it's worthy of an article even if it isn't notable now. Otherwise most of our coverage of historical figures would be labeled "historycruft" and deleted and/or redirected to various list articles. Assuming, of course, there isn't some other double standard underlying the labeling of "cruft". :) _______________________________________________ WikiEN-l mailing list WikiEN-l(a)lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l

Reply

Andrew Gray

5 Mar 5 Mar

9:22 p.m.

On 05/03/2008, David Gerard <dgerard(a)gmail.com> wrote:

http://stats.grok.se/ These are from new logs gathered from the squids. Enter an article and see!

Interesting note: hits to redirects do seem to be counted. http://stats.grok.se/en/200802/USA http://stats.grok.se/en/200802/United_States_of_America http://stats.grok.se/en/200802/United_States (main article) -- - Andrew Gray andrew.gray(a)dunelm.org.uk

Reply

Brian

10:29 p.m.

This data has never been available before. Extremely cool! (And stop looking at my user page so much!!!) On Wed, Mar 5, 2008 at 2:22 PM, Andrew Gray <shimgray(a)gmail.com> wrote:

On 05/03/2008, David Gerard <dgerard(a)gmail.com> wrote:

http://stats.grok.se/ These are from new logs gathered from the squids. Enter an article and

see! Interesting note: hits to redirects do seem to be counted. http://stats.grok.se/en/200802/USA http://stats.grok.se/en/200802/United_States_of_America http://stats.grok.se/en/200802/United_States (main article) -- - Andrew Gray andrew.gray(a)dunelm.org.uk _______________________________________________ WikiEN-l mailing list WikiEN-l(a)lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l

Reply

5915

days inactive

5917

days old

wikien-l@lists.wikimedia.org

Manage subscription

10 comments

8 participants

tags (0)

participants (8)

Andrew Gray
Brian
Bryan Derksen
David Gerard
Relata Refero
Steve Bennett
The Mangoe
Thomas Dalton