End-of-year report: Wikipedia page views

User avatar
HRIP7
Denizen
Posts: 6953
kołdry
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

End-of-year report: Wikipedia page views

Unread post by HRIP7 » Sat Jan 04, 2014 5:23 am

As we were speaking of page views the other day, they seem to have gone down this year for almost all the major European language versions of Wikipedia, reversing a long-term growth trend.

Note (13 January 2014): The graphs below are dynamically linked to their Wikimedia source. Since this thread was started, and in the wake of the press coverage the related Wikipediocracy blog post by Greg Kohs spawned, the Wikimedia Foundation has added mobile page views to the graphs.

English:

Image

German:

Image

French:

Image

Spanish:

Image

Russian:

Image

Italian:

Image

Japanese:

Image

Dutch:

Image

Polish:

Image

Portuguese:

Image

The English Wikipedia has now for three months in a row had fewer than 3,000 people making at least 100 edits a month (2,858 in September, 2,981 in October, 2,917 in November, December figures not in yet).

That particular metric had been consistently above 3,000 since March 2006, peaking at 4,797 in March 2007.
Last edited by HRIP7 on Mon Jan 13, 2014 5:34 pm, edited 3 times in total.
Reason: Note added that graphs have been updated since this post was made.

User avatar
Randy from Boise
Been Around Forever
Posts: 12083
Joined: Sun Mar 18, 2012 2:32 am
Wikipedia User: Carrite
Wikipedia Review Member: Timbo
Actual Name: Tim Davenport
Nom de plume: T. Chandler
Location: Boise, Idaho

Re: End-of-year report: Wikipedia page views

Unread post by Randy from Boise » Sat Jan 04, 2014 6:27 am

Google is different this year. It takes a little work to get to the WP page, the big colorful info box answers a lot of the lite trivia questions without having to sort through WP.

For example, picking a name randomly from the sky, Google for "Jeremy Irons" generates in the left column:

1. IMDb

2. Wikipedia

3. Jeremy Irons.com

and in the right column, a big colorful info box largely extracted from WP.

The latter probably answers a fair number of the actual search inquiries.

That wouldn't count as a WP page hit, even if it did generate the info from WP.

I strongly suspect this is what the graphs are showing.


RfB

User avatar
Kevin
Critic
Posts: 157
Joined: Sun Mar 18, 2012 1:56 am
Wikipedia User: Kevin
Wikipedia Review Member: Kevin
Actual Name: Kevin Godfrey
Location: Adelaide, Australia
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by Kevin » Sat Jan 04, 2014 7:03 am

Randy from Boise wrote:Google is different this year. It takes a little work to get to the WP page, the big colorful info box answers a lot of the lite trivia questions without having to sort through WP.

For example, picking a name randomly from the sky, Google for "Jeremy Irons" generates in the left column:

1. IMDb

2. Wikipedia

3. Jeremy Irons.com

and in the right column, a big colorful info box largely extracted from WP.

The latter probably answers a fair number of the actual search inquiries.

That wouldn't count as a WP page hit, even if it did generate the info from WP.

I strongly suspect this is what the graphs are showing.


RfB
That suggests that the infobox is all that readers really want, not the endless walls of trivia Wikipedia is so good at.

User avatar
Randy from Boise
Been Around Forever
Posts: 12083
Joined: Sun Mar 18, 2012 2:32 am
Wikipedia User: Carrite
Wikipedia Review Member: Timbo
Actual Name: Tim Davenport
Nom de plume: T. Chandler
Location: Boise, Idaho

Re: End-of-year report: Wikipedia page views

Unread post by Randy from Boise » Sat Jan 04, 2014 7:22 am

Kevin wrote:
Randy from Boise wrote:Google is different this year. It takes a little work to get to the WP page, the big colorful info box answers a lot of the lite trivia questions without having to sort through WP.

For example, picking a name randomly from the sky, Google for "Jeremy Irons" generates in the left column:

1. IMDb

2. Wikipedia

3. Jeremy Irons.com

and in the right column, a big colorful info box largely extracted from WP.

The latter probably answers a fair number of the actual search inquiries.

That wouldn't count as a WP page hit, even if it did generate the info from WP.

I strongly suspect this is what the graphs are showing.


RfB
That suggests that the infobox is all that readers really want, not the endless walls of trivia Wikipedia is so good at.

Wikipedia is a serious encyclopedia and a compendium of popular culture.

There is absolutely no doubt that most people seek the latter, looking for super-fast answers to the mundane trivia of daily life. I'll bet it's at least 5 to 1 seeking the compendium over the encyclopedia... Hang around with kids with smart phones and you will see what I mean...

The infobox saves the Google user from having to click ANYTHING after the search. Quicker is better. If somebody is looking for Jeremy Irons' birthdate or the name of that movie he was in last year, the info box will get them there. BOOM.

Google is smart.

RfB

User avatar
Zoloft
Trustee
Posts: 13984
Joined: Wed Mar 14, 2012 11:54 pm
Wikipedia User: Stanistani
Wikipedia Review Member: Zoloft
Actual Name: William Burns
Nom de plume: William Burns
Location: San Diego
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by Zoloft » Sat Jan 04, 2014 9:52 am

*squints at charts*
So Wikipedia project page views as a whole have retreated to 2011 levels.
Progress in the wrong direction.

My avatar is sometimes indicative of my mood:
  • Actual mug ◄
  • Uncle Cornpone
  • Zoloft bouncy pill-thing


User avatar
Poetlister
Genius
Posts: 25599
Joined: Wed Jan 02, 2013 8:15 pm
Nom de plume: Poetlister
Location: London, living in a similar way
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by Poetlister » Sat Jan 04, 2014 10:02 am

HRIP7 wrote:As we were speaking of page views the other day, they seem to have gone down this year for almost all the major European language versions of Wikipedia, reversing a long-term growth trend.

Japanese:
Japanese is a major European language? :blink:

Seriously, I suspect that RfB is right and this is not a real decline. While on principle I never click through to a WP page, I do often find that the infobox answers my query. Of course, if it's important I then find another site to confirm the information.
"The higher we soar the smaller we appear to those who cannot fly" - Nietzsche

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Sat Jan 04, 2014 10:16 am

Randy from Boise wrote:Google is different this year. It takes a little work to get to the WP page, the big colorful info box answers a lot of the lite trivia questions without having to sort through WP.

For example, picking a name randomly from the sky, Google for "Jeremy Irons" generates in the left column:

1. IMDb

2. Wikipedia

3. Jeremy Irons.com

and in the right column, a big colorful info box largely extracted from WP.

The latter probably answers a fair number of the actual search inquiries.

That wouldn't count as a WP page hit, even if it did generate the info from WP.

I strongly suspect this is what the graphs are showing.


RfB
The Google Knowledge Graph or "knowledge panel" was launched in May 2012.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Sat Jan 04, 2014 10:31 am

Poetlister wrote:
HRIP7 wrote:As we were speaking of page views the other day, they seem to have gone down this year for almost all the major European language versions of Wikipedia, reversing a long-term growth trend.

Japanese:
Japanese is a major European language? :blink:
I included the Japanese graph because the Japanese Wikipedia is similar to the major European-language ones in that it is the only other Wikipedia that has seen a similar volume of edits. There are obviously other major world languages that I didn't include, and in some of those page views continued growing in 2013; they are not at the same stage of their development cycle.

Ranking of Wikipedias by edit volume:

Language (number of edits)
1. English (679 million)
2. German (132 million)
3. French (101 million)
4. Spanish (77 million)
5. Russian (71 million)
6. Italian (68 million)
7. Japanese (51 million)
8. Dutch (41 million)
9. Polish (39 million)
10. Portuguese (39 million)

User avatar
lilburne
Habitué
Posts: 4446
Joined: Thu Mar 15, 2012 6:18 pm
Wikipedia User: Nastytroll
Wikipedia Review Member: Lilburne

Re: End-of-year report: Wikipedia page views

Unread post by lilburne » Sat Jan 04, 2014 11:39 am

Randy from Boise wrote:
Kevin wrote:
That suggests that the infobox is all that readers really want, not the endless walls of trivia Wikipedia is so good at.

Wikipedia is a serious encyclopedia and a compendium of popular culture.

There is absolutely no doubt that most people seek the latter, looking for super-fast answers to the mundane trivia of daily life. I'll bet it's at least 5 to 1 seeking the compendium over the encyclopedia... Hang around with kids with smart phones and you will see what I mean...
The majority of visits to wikipedia are single pages. They don't spend long on the page either. People don't visit wikipedia as an encyclopedia they visit it as they would any other blog page. Grab the factoid and move on. It matters not whether the factoid is referenced as they aren't going to look at that. It matters not that lots of the page is linked to other pages, they don't click them. Should people obtain the impression that a significant part of any page is crap the site is doomed. All you'll get is the pop culture queries and that will be covered by Google scraping the site thank you very much.

You aren't writing an encyclopedia you are providing sound bites of pop culture for a mega corp to stick adverts around. Which sort of makes those that argue that WP should have "no adverts" rather silly. The way that most people experience the information you are providing is with ads, and with snooping.

All those mobile deals they are getting into with telcos, how many of those searches are outside the view of the telcos? How many of those searches aren't being monetarized into being able to sell profiles to advertisers?

You are all working for the 'Man'.
They have been inserting little memes in everybody's mind
So Google's shills can shriek there whenever they're inclined

User avatar
thekohser
Majordomo
Posts: 13406
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by thekohser » Sat Jan 04, 2014 3:06 pm

Randy from Boise wrote:and in the right column, a big colorful info box largely extracted from WP.

The latter probably answers a fair number of the actual search inquiries.

That wouldn't count as a WP page hit, even if it did generate the info from WP.

I strongly suspect this is what the graphs are showing.
I have a feeling this is the smartest thing RfB will say all year. Well done, Tim -- seriously.
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

User avatar
neved
Gregarious
Posts: 926
Joined: Sun Sep 09, 2012 5:22 pm
Location: Here, for whatever reason, is the world. And here it stays. With me on it.

Re: End-of-year report: Wikipedia page views

Unread post by neved » Sat Jan 04, 2014 3:25 pm

HRIP7 wrote:As we were speaking of page views the other day, they seem to have gone down this year for almost all the major European language versions of Wikipedia, reversing a long-term growth trend.
.
Could you please link to the similar graphs from the last year?
Also I understand why number of editors is down, but I cannot understand why number of views are? How do you explain this kind of statistic?
"We can forgive the Arabs for killing our children. We cannot forgive them for forcing us to kill their children." Golda Meir

User avatar
neved
Gregarious
Posts: 926
Joined: Sun Sep 09, 2012 5:22 pm
Location: Here, for whatever reason, is the world. And here it stays. With me on it.

Re: End-of-year report: Wikipedia page views

Unread post by neved » Sat Jan 04, 2014 3:29 pm

Randy from Boise wrote:
Wikipedia is a serious encyclopedia



RfB
Here's my serious question: Is there any other " serious encyclopedia " that doesn't count itself as a reliable source?
"We can forgive the Arabs for killing our children. We cannot forgive them for forcing us to kill their children." Golda Meir

User avatar
neved
Gregarious
Posts: 926
Joined: Sun Sep 09, 2012 5:22 pm
Location: Here, for whatever reason, is the world. And here it stays. With me on it.

Re: End-of-year report: Wikipedia page views

Unread post by neved » Sat Jan 04, 2014 3:36 pm

lilburne wrote: The majority of visits to wikipedia are single pages. They don't spend long on the page either. People don't visit wikipedia as an encyclopedia they visit it as they would any other blog page.
Don't you think that Wikipedia would have been much more useful tool, if instead of articles it provided its readers only with references to reliable sources?
"We can forgive the Arabs for killing our children. We cannot forgive them for forcing us to kill their children." Golda Meir

User avatar
thekohser
Majordomo
Posts: 13406
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by thekohser » Sat Jan 04, 2014 3:46 pm

neved wrote:Could you please link to the similar graphs from the last year?
Also I understand why number of editors is down, but I cannot understand why number of views are? How do you explain this kind of statistic?
The graphs are multi-year. A very good explanation of the page-view decline is provided above by Randy from Boise.
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

User avatar
neved
Gregarious
Posts: 926
Joined: Sun Sep 09, 2012 5:22 pm
Location: Here, for whatever reason, is the world. And here it stays. With me on it.

Re: End-of-year report: Wikipedia page views

Unread post by neved » Sat Jan 04, 2014 4:21 pm

thekohser wrote:
neved wrote:Could you please link to the similar graphs from the last year?
Also I understand why number of editors is down, but I cannot understand why number of views are? How do you explain this kind of statistic?
The graphs are multi-year. A very good explanation of the page-view decline is provided above by Randy from Boise.
Thank you. I did not pay enough attention, and thought the info provided in the graphs is for months of 2013.
So, you believe that if Google weren't different this year, the number of views would still be growing?
"We can forgive the Arabs for killing our children. We cannot forgive them for forcing us to kill their children." Golda Meir

User avatar
Randy from Boise
Been Around Forever
Posts: 12083
Joined: Sun Mar 18, 2012 2:32 am
Wikipedia User: Carrite
Wikipedia Review Member: Timbo
Actual Name: Tim Davenport
Nom de plume: T. Chandler
Location: Boise, Idaho

Re: End-of-year report: Wikipedia page views

Unread post by Randy from Boise » Sat Jan 04, 2014 5:15 pm

HRIP7 wrote:
Randy from Boise wrote:Google is different this year. It takes a little work to get to the WP page, the big colorful info box answers a lot of the lite trivia questions without having to sort through WP.

For example, picking a name randomly from the sky, Google for "Jeremy Irons" generates in the left column:

1. IMDb

2. Wikipedia

3. Jeremy Irons.com

and in the right column, a big colorful info box largely extracted from WP.

The latter probably answers a fair number of the actual search inquiries.

That wouldn't count as a WP page hit, even if it did generate the info from WP.

I strongly suspect this is what the graphs are showing.


RfB
The Google Knowledge Graph or "knowledge panel" was launched in May 2012.
That's just about the right timing to explain this.

The boxes have expanded in size and fulsomeness since their launch also.

RfB

User avatar
Randy from Boise
Been Around Forever
Posts: 12083
Joined: Sun Mar 18, 2012 2:32 am
Wikipedia User: Carrite
Wikipedia Review Member: Timbo
Actual Name: Tim Davenport
Nom de plume: T. Chandler
Location: Boise, Idaho

Re: End-of-year report: Wikipedia page views

Unread post by Randy from Boise » Sat Jan 04, 2014 5:21 pm

neved wrote:
Randy from Boise wrote:
Wikipedia is a serious encyclopedia


RfB
Here's my serious question: Is there any other " serious encyclopedia " that doesn't count itself as a reliable source?

I would be nervous of a circularly sourced encyclopedia, if such a thing actually exists.

WP is of uneven quality. Parts of it are excellent. Parts of it are terrible. Overall, it is good to very good.

The entire notion of a "reliable source" is a goofy Wikipediaism. There are sources, some better and some worse. There are facts and purported facts, some right and some wrong.


RfB

User avatar
Randy from Boise
Been Around Forever
Posts: 12083
Joined: Sun Mar 18, 2012 2:32 am
Wikipedia User: Carrite
Wikipedia Review Member: Timbo
Actual Name: Tim Davenport
Nom de plume: T. Chandler
Location: Boise, Idaho

Re: End-of-year report: Wikipedia page views

Unread post by Randy from Boise » Sat Jan 04, 2014 5:43 pm

lilburne wrote:
Randy from Boise wrote:
Kevin wrote:
That suggests that the infobox is all that readers really want, not the endless walls of trivia Wikipedia is so good at.
Wikipedia is a serious encyclopedia and a compendium of popular culture.

There is absolutely no doubt that most people seek the latter, looking for super-fast answers to the mundane trivia of daily life. I'll bet it's at least 5 to 1 seeking the compendium over the encyclopedia... Hang around with kids with smart phones and you will see what I mean...
The majority of visits to wikipedia are single pages. They don't spend long on the page either. People don't visit wikipedia as an encyclopedia they visit it as they would any other blog page. Grab the factoid and move on. It matters not whether the factoid is referenced as they aren't going to look at that. It matters not that lots of the page is linked to other pages, they don't click them. Should people obtain the impression that a significant part of any page is crap the site is doomed. All you'll get is the pop culture queries and that will be covered by Google scraping the site thank you very much.

You aren't writing an encyclopedia you are providing sound bites of pop culture for a mega corp to stick adverts around. Which sort of makes those that argue that WP should have "no adverts" rather silly. The way that most people experience the information you are providing is with ads, and with snooping.

All those mobile deals they are getting into with telcos, how many of those searches are outside the view of the telcos? How many of those searches aren't being monetarized into being able to sell profiles to advertisers?

You are all working for the 'Man'.
Last point first — yes, we are. That's part of the deal when you create something and give it away to anybody so they can do anything. Is that ideal? No. But that is the deal.

as for

>>You aren't writing an encyclopedia you are providing sound bites of pop culture for a mega corp to stick adverts around.

I 100% agree that most people hop on, grab a fact or three about something mundane, and move along. That's not why I write there, but it's obviously the main use of WP. That's sort of like saying that esoteric academic press history books are dead because way more people read paperback romance novels, fluff biographies, puzzle books, and self-help titles. It doesn't necessarily follow. There is a serious encyclopedia there, one just has to look for it.

As for Google scraping... A couple years ago the main argument was "if only Google didn't push WP to the top of its search results, Wikipedia would die." The counter-argument was made — but not universally accepted — that Google had no interest whatsoever in killing Wikipedia, because WP produced masses of constantly updating free content for them worth hundreds of millions or billions of dollars. Wikipedia needs Google, but even more, Google needs Wikipedia.

That's very, very evident now, is it not? And yeah, the Evil Empire of Google is gonna figure out how to more effectively scrape and monetize that content, because that's what they do, being an Evil Empire and all. Like I say, they're smart.

RfB

User avatar
lilburne
Habitué
Posts: 4446
Joined: Thu Mar 15, 2012 6:18 pm
Wikipedia User: Nastytroll
Wikipedia Review Member: Lilburne

Re: End-of-year report: Wikipedia page views

Unread post by lilburne » Sat Jan 04, 2014 6:14 pm

Randy from Boise wrote:
lilburne wrote: You are all working for the 'Man'.
Last point first — yes, we are. That's part of the deal when you create something and give it away to anybody so they can do anything. Is that ideal? No. But that is the deal.
Its not a deal I'm going to sign up for. I recently removed 7,000 images from Google, not a lot in the grand scheme I know, but some of them are not otherwise available, or documented online.
Randy from Boise wrote: >>You aren't writing an encyclopedia you are providing sound bites of pop culture for a mega corp to stick adverts around.

I 100% agree that most people hop on, grab a fact or three about something mundane, and move along. That's not why I write there, but it's obviously the main use of WP. That's sort of like saying that esoteric academic press history books are dead because way more people read paperback romance novels, fluff biographies, puzzle books, and self-help titles. It doesn't necessarily follow. There is a serious encyclopedia there, one just has to look for it.
That's nothing like what I'm saying. I'm saying that inserting content from esoteric academic press history books into the middle of a romance novel, devalues both. How can one find this serious encyclopedia? If there are various serious blog posts there, then they are swamped by the trite and the nonsense.

Why would anyone treat any page on WP with any more respect than the page on "The Sea" or teh bulk of the biological and geological nonsense? When WP tells me that Richard II was king in 1345, or that Edward III divided his kingdom amongst his sons, why should I take any other history page seriously? What am I to make of Hugh Despenser being Edward II gay lover?

If I don't know a subject area then I have to take what is written in something purporting to be an encyclopedia on trust. If I can't do that for the majority of pages then it is a poor resource. Worse it is a pedlar of untruths, and ignorance.
They have been inserting little memes in everybody's mind
So Google's shills can shriek there whenever they're inclined

enwikibadscience
Habitué
Posts: 1423
Joined: Mon Oct 21, 2013 9:58 pm

Re: End-of-year report: Wikipedia page views

Unread post by enwikibadscience » Sat Jan 04, 2014 6:20 pm

neved wrote:
lilburne wrote: The majority of visits to wikipedia are single pages. They don't spend long on the page either. People don't visit wikipedia as an encyclopedia they visit it as they would any other blog page.
Don't you think that Wikipedia would have been much more useful tool, if instead of articles it provided its readers only with references to reliable sources?
But the "reliable sources" would be selected by the same editors. Last year's WikiCup winner proved beyond a doubt unable to read the reliable sources she was using. What value would a list of her reliable sources be? Zero.

Well, it might do less harm.

:blink:

User avatar
Randy from Boise
Been Around Forever
Posts: 12083
Joined: Sun Mar 18, 2012 2:32 am
Wikipedia User: Carrite
Wikipedia Review Member: Timbo
Actual Name: Tim Davenport
Nom de plume: T. Chandler
Location: Boise, Idaho

Re: End-of-year report: Wikipedia page views

Unread post by Randy from Boise » Sat Jan 04, 2014 6:31 pm

enwikibadscience wrote:
neved wrote:
lilburne wrote: The majority of visits to wikipedia are single pages. They don't spend long on the page either. People don't visit wikipedia as an encyclopedia they visit it as they would any other blog page.
Don't you think that Wikipedia would have been much more useful tool, if instead of articles it provided its readers only with references to reliable sources?
But the "reliable sources" would be selected by the same editors. Last year's WikiCup winner proved beyond a doubt unable to read the reliable sources she was using. What value would a list of her reliable sources be? Zero.

Well, it might do less harm.

:blink:
I definitely think the most important thing that WP can do is steer people towards the published literature on this or that topic. The job it does in this is not great thus far, but improving. Obviously, this is always going to be a small part of total content. That's more or less my objective when I write something: introducing a subject and then steering towards available publications.

RfB

User avatar
lilburne
Habitué
Posts: 4446
Joined: Thu Mar 15, 2012 6:18 pm
Wikipedia User: Nastytroll
Wikipedia Review Member: Lilburne

Re: End-of-year report: Wikipedia page views

Unread post by lilburne » Sat Jan 04, 2014 6:53 pm

Yep its improving playing the RA game:

Fermín Martín Piera (T-H-L)
Battle of Gainsborough (T-H-L)
List of minor planets/130601–130700 (T-H-L)
Nanofountain probe (T-H-L)
1962 in British radio (T-H-L)
Christian Evangelical Church in Minahasa (T-H-L)
Kernodle Middle (T-H-L)
Olga Sánchez Cordero (T-H-L)
SolarWorld (T-H-L)
Blaenau Gwent People's Voice (T-H-L)

Can you see your encyclopedia in any of that? Or any quality links/references?
They have been inserting little memes in everybody's mind
So Google's shills can shriek there whenever they're inclined

User avatar
Poetlister
Genius
Posts: 25599
Joined: Wed Jan 02, 2013 8:15 pm
Nom de plume: Poetlister
Location: London, living in a similar way
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by Poetlister » Sat Jan 04, 2014 10:35 pm

it is extremely difficult to know what is or is not a reliable source unless you have a reasonable amount of expertise in the subject. it is not difficult to find incorrect facts referenced on WP to apparently reliable sources (and I don't mean tabloids) that are in fact wrong because they are by apparently well-qualified people working outside their areas of expertise, such as astronomers writing about biology or geologists writing about astronomy. Obviously, in a conventional reference work that shouldn't happen (I dont say it can't, but it must be quite rare) because the editors will have expertise and will choose contributors with expertise.
"The higher we soar the smaller we appear to those who cannot fly" - Nietzsche

User avatar
SB_Johnny
Habitué
Posts: 4640
Joined: Thu Mar 15, 2012 1:26 am
Wikipedia User: SB_Johnny
Wikipedia Review Member: SB_Johnny

Re: End-of-year report: Wikipedia page views

Unread post by SB_Johnny » Sat Jan 04, 2014 11:20 pm

Randy from Boise wrote:That's very, very evident now, is it not? And yeah, the Evil Empire of Google is gonna figure out how to more effectively scrape and monetize that content, because that's what they do, being an Evil Empire and all. Like I say, they're smart.

RfB
I predict that those article snippets will start to sport sponsored hyperlinks in short order.

For Google, WP is like a semantic version of DMOZ that their crawlers can scrape and their algorithms can use to improve themselves. Of course they want to steer people there.
This is not a signature.

User avatar
lilburne
Habitué
Posts: 4446
Joined: Thu Mar 15, 2012 6:18 pm
Wikipedia User: Nastytroll
Wikipedia Review Member: Lilburne

Re: End-of-year report: Wikipedia page views

Unread post by lilburne » Sat Jan 04, 2014 11:33 pm

Lets try again random article 10X

Dondon (T-H-L) - This has an excellent external link - a 5 year improve.
:sarcasm:
1660 in Norway (T-H-L) - 18 months of stupendous improvement.
Knut Leo Abrahamsen (T-H-L) - 30 months of improvement.
Duchess Violante Beatrice of Bavaria (T-H-L) - 3 years improvement. Consisting mainly of plagiarisms, from someone who's writing style is early 20th century pompous. Who writes like this in the 21st century?
The Governor entered her domain in April 1717, taking up residence in the city-centre.
Was she a bag-lady?
Cross Creek Township, Washington County, Pennsylvania (T-H-L) - The work of 11 years.
Joe Adamov (T-H-L) - Eight years worth of improvements.
Thais clavigera (T-H-L) - 4 years.
Staten Island Mall (T-H-L) - 10 years improvements. References are mainly bus timetables.
The World of Shannara (T-H-L) - 8 years of improving. Originally seems to have been a publishers blurb. No better now.
Soissons (T-H-L) - 11 years of improving. Most of the content is from a book published in the mid 19th century.
They have been inserting little memes in everybody's mind
So Google's shills can shriek there whenever they're inclined

EricBarbour
 
Posts: 10891
Joined: Wed Mar 14, 2012 11:32 pm
Location: hell

Re: End-of-year report: Wikipedia page views

Unread post by EricBarbour » Sat Jan 04, 2014 11:51 pm

SB_Johnny wrote:For Google, WP is like a semantic version of DMOZ that their crawlers can scrape and their algorithms can use to improve themselves. Of course they want to steer people there.
Perfectly put. No matter what the resource is, if it's "free" and "open" and they can scrape the hell out of it and deny any knowledge of its problems, they love it. Google has ripped off DMOZ listings before, but apparently they hardly use it today.

One does see a whole lot of complaints about DMOZ corruption, to the point where the ED article makes jokes about paying DMOZ editors for a listing......

http://www.webmarketinggroup.co.uk/Blog ... -1158.aspx
http://moz.com/blog/wheres-the-love-dmoz
http://www.seroundtable.com/odp-links-poll-15757.html

DMOZ was declining steadily in Alexa rank, until June 2013, when it took a sudden upward jump. Perhaps they made some kind of deal with a web portal, or perhaps Alexa diddled the results -- there was no news about it at the time.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Sun Jan 05, 2014 3:14 am

Randy from Boise wrote:
HRIP7 wrote: The Google Knowledge Graph or "knowledge panel" was launched in May 2012.
That's just about the right timing to explain this.

The boxes have expanded in size and fulsomeness since their launch also.

RfB
To me it looks like the reversal in Wikipedia's page view fortunes began around March or April 2013. This is where the chart traces begin to diverge from the pattern sustained through the previous years.

If Google's knowledge panel had driven this change, I would have expected to see some visible effect in May 2012, when it was introduced, even if it had subsequently been expanded and had some more bells and whistles added to it. But I see no such change – the 2012 chart traces conform to the general pattern exhibited during previous years. And Wikipedia articles were part of the knowledge panel right from the beginning.

I am sure that the Google knowledge panel will have made some difference of course. In some cases, it may even drive traffic to Wikipedia, as Wikipedia is linked in it.

The bigger story though of course, as Lilburne points out, is that Wikipedia is being monetised through the ads Google places on its search results pages. And recall that right from the beginning, Wikipedia and its Creative Commons licence were set up to allow commercial exploitation of its content.

I'm sure that Google will look at the money they donated to Wikipedia a few years ago as a worthwhile investment. Compared to the revenue they are getting from Wikipedia, it's peanuts. And all the result of free labour, willingly given. Digital sharecropping.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Sun Jan 05, 2014 3:19 am

Note that the Knowledge Graph was initially available only in English-speaking countries. Introduction in the German version of Google was apparently in December 2012.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Sun Jan 05, 2014 3:56 am

Introduction of the Knowledge Graph in the Spanish, French, Portuguese, Japanese, Russian, and Italian versions was also in December 2012.

In May 2013, Polish, Turkish, traditional Chinese and Simplified Chinese were added. (That article mentions that Korean was also introduced some time prior to May 2013.)

Here is a complete timeline, taken from highposition.com:
16th May 2012 – Knowledge Graph Launched in the US.
In May 2012 Knowledge Graph officially rolled-out in the United States. According to Google Knowledge Graph initially contained data about more than 500 million objects and more than 3.5 billion facts about and relationships between different objects.
http://googleblog.blogspot.co.uk/2012/0 ... s-not.html

8th August 2012 – English Roll-out
Google rolls-out Knowledge Graph across every English-speaking country in the world. At the same time as the language roll-out, Google launch the Knowledge Graph Carousel available via Google.com only.
http://googleblog.blogspot.co.uk/2012/0 ... e-one.html

5th September 2012 – Carousel Global Roll-out
After initial launch in the US, the Knowledge Graph Carousel rolls out globally to all in English speaking countries.
http://insidesearch.blogspot.co.uk/2012 ... ousel.html

19th October 2012 – “People also search for” Integration
Google added explanations of connnections to the “People also search for” function allowing users to better understand the connection between the entities.
http://insidesearch.blogspot.co.uk/2012 ... _9451.html

30th November 2012 – Medications Added
Google added medical information to Knowledge Graph with detailed information about various medications, with data pulled from U.S. FDA, the National Library of Medicine, and DailyMed amongst others.
http://insidesearch.blogspot.co.uk/2012 ... y-and.html

4th December 2012 – Language Roll-Out
Knowledge Graph rolls-out support for additional languages including Spanish, French, German, Portuguese, Japanese, Russian and Italian.
Google state that Knowledge Graph now contains information on 570 million entities and contains 18 billion facts.
http://insidesearch.blogspot.co.uk/2012 ... dge_4.html
2013

8th January 2013 – Clickable Source URLs
Google implemented a clickable source link when Knowledge Graph displays ‘Web Definitions’.
http://searchengineland.com/google-fixe ... nks-144511

27th March 2013 – Android Play Movies & TV –
Knowledge Graph expands into movies, availabe via the Android Play Movies & TV app allowing users to see relevant information directly within a movie.
http://techcrunch.com/2013/03/27/google ... tv-update/

9th April 2013 – Non-Profit Information
Google adds information about non-profit organisations to Knowledge Graph.
https://plus.google.com/+google/posts/MtxDMf7J9P2

15th May 2013 – Language Roll-out & Added Stats
Knowledge Graph rolls-out additional language support including Polish, Turkish, Simplified Chinese and Traditional Chinese announced by Amit Singhal at Google IO.
Knowledge Graph now contains additional statistics allowing the system to try to predicted a users next question.
Google announce support for the Organization Schema markup.
http://techcrunch.com/2013/05/15/google ... languages/

30th May 2013 – Nutritional Information
Nutritional information added to Knowledge Graph for over 1,000 fruits, vegetables, meats and meals.
http://insidesearch.blogspot.com/2013/0 ... e-jar.html

17th June 2013 – Carousel Expansion
It was rumoured that 50 different topics are now serving the Knowledge Graph carousel. Unconfirmed by Google.
http://searchengineland.com/google-know ... hes-163562

18th June 2013 – Carousel for Local Search
The Knowledge Graph Carousel now integrates local serch data for resturants, bars and other local places on desktops.
https://plus.google.com/+google/posts/KpsbyvHUotN

26th September 2013 – Comparions and Filters
Google add the ability to compare various entities (such as Earth and Mars) providing a range of facts and figures.
Users can also filter results by various factors depending on their search query.
Google also launch a Music Knowledge Graph with YouTube, Google Now, New Carousel and more.
http://insidesearch.blogspot.co.uk/2013 ... tting.html

3rd October 2013 – Google Now Integration
Google Now integrated with Knowledge Graph.
9th October 2013 – Language Roll-Out & TV Episodes
Knowledge Graph rolls-out support for additional languages including Croatian, Serbian, Hindi, Slovak, Lithuanian, Slovenian, Persian, Catalan, and Latvian, as well as Filipino. The announcement was made by Google Philippines Country Head Narciso Reyes
Google also add TV Episode information to Knowledge Graph.
https://plus.google.com/+google/posts/daesUmX45NG

13th December 2013 – Knowledge Graph Ad Testing
Gianluca Fiorelli noticed Google testing ads in Knowledge Graph results providing an example of local car dealer ads in vehicle related panels.
http://www.seroundtable.com/google-know ... 17813.html
This ties in quite well with the page view charts now, bearing in mind that the English-language roll-out took place in two stages, and the Polish roll-out came later.

dogbiscuit
Retired
Posts: 2723
Joined: Wed Mar 14, 2012 11:32 pm
Wikipedia User: tiucsibgod

Re: End-of-year report: Wikipedia page views

Unread post by dogbiscuit » Sun Jan 05, 2014 10:11 am

Poetlister wrote:it is extremely difficult to know what is or is not a reliable source unless you have a reasonable amount of expertise in the subject. it is not difficult to find incorrect facts referenced on WP to apparently reliable sources (and I don't mean tabloids) that are in fact wrong because they are by apparently well-qualified people working outside their areas of expertise, such as astronomers writing about biology or geologists writing about astronomy. Obviously, in a conventional reference work that shouldn't happen (I dont say it can't, but it must be quite rare) because the editors will have expertise and will choose contributors with expertise.
I just had a look back at the factory farming article Intensive_animal_farming (T-H-L) and that is a pretty good example as to why you cannot use Wikipedia to even start research. A bloated article that has been edit warred over for years, it is essentially an article that views factory farming as negative, seemingly every paragraph wants to highlight the problems of factory farming, every fact that might justify why the practice has evolved has to be counterbalanced by a problem. I'm not a fan of factory farming methods, but the reality is that the general public have happily lived with the system for decades.

The article still doesn't distinguish between intensive farming methods (which can occur on normal farms, such as feeding methods that allow you to contain more cattle on a farm than the land can support naturally) and factory farms where unnatural environments are created. That is after approaching a decade of debate on the subject. As an example, that distinction was of much debate to consider whether BSE was a side-effect of factory farming or simply poor farming practices (by the farm food supply industry) - Wikipedians got into such a mess over that debate that one of the most serious health problems ever created by farming seems to have drifted from the article entirely. You can't even tell from the history as the main debate shifted in November from the old Factory Farming article which is now a re-direct.

I think that makes a really good example of why Wikipedia ultimately fails. Information that is uncontroversial can readily be found on the Internet via Google so Wikipedia adds nothing but a convenient way for Google to acquire manually summarised information from other sites. Information that needs to be editorialised so that the reader can instantly understand the information is controversial and requires thought has no safe mechanism within Wikipedia. The environment at Wikipedia is incapable of providing that.
Time for a new signature.

User avatar
lilburne
Habitué
Posts: 4446
Joined: Thu Mar 15, 2012 6:18 pm
Wikipedia User: Nastytroll
Wikipedia Review Member: Lilburne

Re: End-of-year report: Wikipedia page views

Unread post by lilburne » Sun Jan 05, 2014 11:04 am

This is the issue. As we've said for yonks, without wikipedia all the information in there would still be available, in books, blogs, academic articles, and web pages dedicated to the subject matter. WP brings nothing extra to the table at all. Some of the WP blog articles I listed above are scrapings of sorts stats from some other site. Nothing has been added by WP except perhaps push the originating site down the Google ranks.

NEWSFLASH adding "citation needed" or "dead link to somewhere" does not an encyclopedia make.
They have been inserting little memes in everybody's mind
So Google's shills can shriek there whenever they're inclined

User avatar
Poetlister
Genius
Posts: 25599
Joined: Wed Jan 02, 2013 8:15 pm
Nom de plume: Poetlister
Location: London, living in a similar way
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by Poetlister » Mon Jan 06, 2014 12:33 pm

lilburne wrote:This is the issue. As we've said for yonks, without wikipedia all the information in there would still be available, in books, blogs, academic articles, and web pages dedicated to the subject matter. WP brings nothing extra to the table at all. Some of the WP blog articles I listed above are scrapings of sorts stats from some other site. Nothing has been added by WP except perhaps push the originating site down the Google ranks.

NEWSFLASH adding "citation needed" or "dead link to somewhere" does not an encyclopedia make.
At its best (which of course is all too rare) it gathers information that is not online, or behind a paywall. It can also assemble the information into a more coherent form. I expect that most people here would agree that these rare good articles are swamped by - shall we say - less good stuff, and it is hard for most people to tell the difference.

The other issue is that by providing all this for free, the good articles can drive the good sources out of business. Thus ironically if WP were what it should be, it might create just as many problems, albeit different ones.
"The higher we soar the smaller we appear to those who cannot fly" - Nietzsche

Casliber
Gregarious
Posts: 752
Joined: Thu Jul 19, 2012 3:51 am
Wikipedia User: Casliber
Wikipedia Review Member: Casliber
Location: Sydney, Oz

Re: End-of-year report: Wikipedia page views

Unread post by Casliber » Mon Jan 06, 2014 1:10 pm

HRIP7 wrote:
Randy from Boise wrote:
HRIP7 wrote: The Google Knowledge Graph or "knowledge panel" was launched in May 2012.
That's just about the right timing to explain this.

The boxes have expanded in size and fulsomeness since their launch also.

RfB
To me it looks like the reversal in Wikipedia's page view fortunes began around March or April 2013. This is where the chart traces begin to diverge from the pattern sustained through the previous years.

If Google's knowledge panel had driven this change, I would have expected to see some visible effect in May 2012, when it was introduced, even if it had subsequently been expanded and had some more bells and whistles added to it. But I see no such change – the 2012 chart traces conform to the general pattern exhibited during previous years. And Wikipedia articles were part of the knowledge panel right from the beginning.
Not necessarily - sometimes it takes folks months to take a really obvious development on board. Some folks have old browsers it doesn't show up on. I still think there is a link.
Cas

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Mon Jan 06, 2014 1:56 pm

Casliber wrote:Not necessarily - sometimes it takes folks months to take a really obvious development on board. Some folks have old browsers it doesn't show up on. I still think there is a link.
Cas
I agree. Given that the English release was staggered, and the other languages came later, it is a good match. If you look at the Polish chart (Polish Google hat the Knowledge Graph added later), it even seems as though the downward trend kicked in a little bit later for that chart than for the others.

Overall, it seems a very good match, and a plausible mechanism for reducing Wikipedia page views.

User avatar
Randy from Boise
Been Around Forever
Posts: 12083
Joined: Sun Mar 18, 2012 2:32 am
Wikipedia User: Carrite
Wikipedia Review Member: Timbo
Actual Name: Tim Davenport
Nom de plume: T. Chandler
Location: Boise, Idaho

Re: End-of-year report: Wikipedia page views

Unread post by Randy from Boise » Mon Jan 06, 2014 5:26 pm

HRIP7 wrote:
Randy from Boise wrote:
HRIP7 wrote: The Google Knowledge Graph or "knowledge panel" was launched in May 2012.
That's just about the right timing to explain this.

The boxes have expanded in size and fulsomeness since their launch also.

RfB
To me it looks like the reversal in Wikipedia's page view fortunes began around March or April 2013. This is where the chart traces begin to diverge from the pattern sustained through the previous years.

If Google's knowledge panel had driven this change, I would have expected to see some visible effect in May 2012, when it was introduced, even if it had subsequently been expanded and had some more bells and whistles added to it. But I see no such change – the 2012 chart traces conform to the general pattern exhibited during previous years. And Wikipedia articles were part of the knowledge panel right from the beginning.

I am sure that the Google knowledge panel will have made some difference of course. In some cases, it may even drive traffic to Wikipedia, as Wikipedia is linked in it.

The bigger story though of course, as Lilburne points out, is that Wikipedia is being monetised through the ads Google places on its search results pages. And recall that right from the beginning, Wikipedia and its Creative Commons licence were set up to allow commercial exploitation of its content.

I'm sure that Google will look at the money they donated to Wikipedia a few years ago as a worthwhile investment. Compared to the revenue they are getting from Wikipedia, it's peanuts. And all the result of free labour, willingly given. Digital sharecropping.
I don't think that the current "Knowledge Panel" is the same as the original variant. The first version, as I recall, had a picture and a WP extract and that was it. Now the Knowledge Panels are big fat things with multiple images and additional information.

I could be misremembering, but I recall much smaller initial presentations, which would not "move the needle" much in terms of click-through...

RfB

User avatar
thekohser
Majordomo
Posts: 13406
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by thekohser » Mon Jan 06, 2014 6:54 pm

Randy from Boise wrote:I could be misremembering...
You're not. (At least in this case.)
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

Anthonyhcole
Habitué
Posts: 1120
Joined: Sat Apr 14, 2012 3:35 am
Wikipedia User: Anthonyhcole

Re: End-of-year report: Wikipedia page views

Unread post by Anthonyhcole » Mon Jan 06, 2014 10:08 pm

lilburne wrote: The majority of visits to wikipedia are single pages. They don't spend long on the page either.
Do you have a source for the last sentence?

User avatar
Ming
the Merciless
Posts: 2966
Joined: Wed Apr 03, 2013 1:35 pm

Re: End-of-year report: Wikipedia page views

Unread post by Ming » Mon Jan 06, 2014 10:21 pm

What would be more interesting is what pattern of pages get used. The chickenpox example from another thread, for example, pulls from the National Library of Medicine, as does every other medical condition Ming has thus far tried. Movie titles seem to bring up IMDb.

dogbiscuit
Retired
Posts: 2723
Joined: Wed Mar 14, 2012 11:32 pm
Wikipedia User: tiucsibgod

Re: End-of-year report: Wikipedia page views

Unread post by dogbiscuit » Tue Jan 07, 2014 12:50 am

Ming wrote:What would be more interesting is what pattern of pages get used. The chickenpox example from another thread, for example, pulls from the National Library of Medicine, as does every other medical condition Ming has thus far tried. Movie titles seem to bring up IMDb.
I guess Google does know not to use the medical boxes in the UK. They do not show here.
Time for a new signature.

User avatar
thekohser
Majordomo
Posts: 13406
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by thekohser » Tue Jan 07, 2014 5:25 am

Anthonyhcole wrote:
lilburne wrote: The majority of visits to wikipedia are single pages. They don't spend long on the page either.
Do you have a source for the last sentence?
Alexa says 4.5 minutes per day per visitor. I find that hard to believe, since they give about the same time for MyWikiBiz visitors, which I know to be closer to 30 seconds, if even that. You have to count all the "bounced" visitors who spend less than 5 seconds to allow the page to even fully load, before they're off to another site. Not sure Alexa's doing that.
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

User avatar
Vigilant
Sonny, I've got a whole theme park full of red delights for you.
Posts: 31489
Joined: Thu Mar 29, 2012 8:16 pm
Wikipedia User: Vigilant
Wikipedia Review Member: Vigilant

Re: End-of-year report: Wikipedia page views

Unread post by Vigilant » Tue Jan 07, 2014 8:51 am

thekohser wrote:
Anthonyhcole wrote:
lilburne wrote: The majority of visits to wikipedia are single pages. They don't spend long on the page either.
Do you have a source for the last sentence?
Alexa says 4.5 minutes per day per visitor. I find that hard to believe, since they give about the same time for MyWikiBiz visitors, which I know to be closer to 30 seconds, if even that. You have to count all the "bounced" visitors who spend less than 5 seconds to allow the page to even fully load, before they're off to another site. Not sure Alexa's doing that.
Must be the VisualEditor load cycle...
Hello, John. John, hello. You're the one soul I would come up here to collect myself.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Wed Jan 08, 2014 2:49 pm

More precise page view figures are available on this page:

http://stats.wikimedia.org/EN/TablesPag ... ojects.htm

In the table, Wikipedia (non-mobile) shows a drop of 20% from December 2012 (16.718M) to December 2013 (13.423M). That's for all languages combined.

This 3.3M drop is somewhat offset by a 1.4M rise over the same period in page views for Wikipedia (mobile), but there is still a net drop of close to 2M for the combined total.

And note that mobile users will not edit, as editing on a mobile phone is nigh-impossible.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Wed Jan 08, 2014 3:22 pm

There is another table with figures for the individual language versions, available here:

http://stats.wikimedia.org/EN/TablesPag ... mbined.htm

This shows, at the top of the table, that the English Wikipedia dropped 12% of its page views from December 2012 to December 2013. The German Wikipedia dropped 17% over the same period, the Spanish, 19%. Those are massive drops.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Wed Jan 08, 2014 3:36 pm

There are still some curious discrepancies between the graphs posted at the top of this thread, and the figures given in this table.

For example, the table says that the English Wikipedia had 8.811M page views in December 2013, and 8.215M page views in December 2011. So according to the table, page views were higher in December 2013 than they were in December 2011.

However, the graph given on the Wikipedia report card page clearly shows lower page views for December 2013 than for December 2011:

Image

Anyone understand why?

User avatar
thekohser
Majordomo
Posts: 13406
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by thekohser » Wed Jan 08, 2014 4:51 pm

HRIP7 wrote:Anyone understand why?
The Wikimedia Foundation oversaw the data.
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Wed Jan 08, 2014 6:55 pm

thekohser wrote:
HRIP7 wrote:Anyone understand why?
The Wikimedia Foundation oversaw the data.
I think I've figured it out. The graphs on the report card only show non-mobile page views. The English Wikipedia traditionally makes up approximately half of all Wikipedia page views. If you look at the Wikipedia non-mobile column on http://stats.wikimedia.org/EN/TablesPag ... ojects.htm, the pattern closely matches that in the report card graph, and the figures plotted are a pretty good match for about half of the total page views for each month (the English Wikipedia proportion of the whole).

And the December 2013 page views for Wikipedia non-mobile are indeed less than the December 2011 figure (13.423M vs. 14.138M), matching the graph. So it's the mobile page views that account for the difference.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Wed Jan 08, 2014 7:03 pm

In fact, there is a data table for non-mobile page views as well:

http://stats.wikimedia.org/EN/TablesPag ... onthly.htm

And that is indeed the data plotted on the graphs. The graphs are non-mobile only. And the English Wikipedia's (non-mobile) page views dropped by 21% from December 2012 to December 2013.

The German Wikipedia's non-mobile page views dropped by 30%. Those for the Spanish Wikipedia by 29%. Those for the Japanese Wikipedia by 25%.

If you include the mobile data, the drops are still drastic: 12% for the English Wikipedia, 17% for the German Wikipedia, 19% for the Spanish Wikipedia, 9% for the Japanese Wikipedia (which has a very high proportion of mobile use).

User avatar
thekohser
Majordomo
Posts: 13406
Joined: Thu Mar 15, 2012 5:07 pm
Wikipedia User: Thekohser
Wikipedia Review Member: thekohser
Actual Name: Gregory Kohs
Location: United States
Contact:

Re: End-of-year report: Wikipedia page views

Unread post by thekohser » Wed Jan 08, 2014 8:08 pm

Excellent sleuthing, HRIP7. Even with the mobile factor, I think most would agree that mobile engagement with the content of Wikipedia will tend to be more superficial than when engaged by a desktop monitor. So, one could even further discount the page-views via mobile platforms, if you want to be honest about how "meaningful" the interaction is.
"...making nonsensical connections and culminating in feigned surprise, since 2006..."

User avatar
Vigilant
Sonny, I've got a whole theme park full of red delights for you.
Posts: 31489
Joined: Thu Mar 29, 2012 8:16 pm
Wikipedia User: Vigilant
Wikipedia Review Member: Vigilant

Re: End-of-year report: Wikipedia page views

Unread post by Vigilant » Wed Jan 08, 2014 8:23 pm

thekohser wrote:Excellent sleuthing, HRIP7. Even with the mobile factor, I think most would agree that mobile engagement with the content of Wikipedia will tend to be more superficial than when engaged by a desktop monitor. So, one could even further discount the page-views via mobile platforms, if you want to be honest about how "meaningful" the interaction is.
I think there's probablly a new Japanese prank game show in the works entitled, "Make an edit to wikipedia on your phone using the VisualEditor."
Hello, John. John, hello. You're the one soul I would come up here to collect myself.

User avatar
HRIP7
Denizen
Posts: 6953
Joined: Thu Mar 15, 2012 2:05 am
Wikipedia User: Jayen466
Wikipedia Review Member: HRIP7
Actual Name: Andreas Kolbe
Location: UK

Re: End-of-year report: Wikipedia page views

Unread post by HRIP7 » Mon Jan 13, 2014 5:28 pm

The Wikimedia Foundation has updated the report card graphs shown in the original post at the top of this thread, by adding mobile views. (When first posted, the graphs did not include mobile views. But as the graphs are linked to their source URLs, they automatically update here as well.)

Post Reply