Wikipedia talk:Top 25 Report/Archive 4

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1 Archive 2 Archive 3 Archive 4 Archive 5 Archive 6 → Archive 10

General Question

Upper corner Screenshot of Wikipedia The Free Encyclo & Project Page
Hi, newbie here. In an effort to improve this Top 25 list for mid-Feb '16, my revision was UnDone by User:SchreiberBike, with the following reasoning: "This is not a Wikipedia article and does not need to follow Wikipedia standards. This is 'As prepared by Milowent'." As an ordinary Wikipedia user who accessed the page through a search in Wikipedia, this seems a contradiction. Second, in this Talk section, User:JMinor_(WMF) alludes to violation of Wikipedia standards, addressing the non-neutral point of view. Nevertheless, I will temporarily suspend that thought because maybe a technical term "Project Page" is exempt from "Article" status, the latter being one of the exclusive types of page that must follow Wikipedia standards? Curious for the appropriate link to or a direct explanation? Thanks in advance for taking time to help. Jackemeyer (talk) 18:34, 5 March 2016 (UTC)
This isn't an encyclopedic article; it's really more of an op-ed piece discussing matters that aren't really of interest to anyone outside the community. A smaller version appears as an article in the Signpost every week, and the Signpost is full of opinionated material. Serendipodous 19:08, 5 March 2016 (UTC)

You can understand my confusion from the publication on the Wikipedia domain, and now I am curious about "the community". To whom do you refer? Jackemeyer (talk) 07:01, 11 March 2016 (UTC)

this group of people. Or, perhaps more specifically, the group of Wikipedia editors that follow this page and are interested in the evolution of Wikipedia itself. ~Mable (chat) 07:42, 11 March 2016 (UTC)
It's very odd that you happened upon this page by a regular Wikipedia search. This should only be possible if you search started with "Wikipedia:" or "WP:" (per the article's title and namespace). Do you remember what you were searching for? Mable (chat) 20:09, 5 March 2016 (UTC)

Good catch as I used too few words to describe my journey here -- in searching(-by-clicking) for page viewing statistics, I ended up at this list Category:Wikipedia statistics and found what appeared to be a Wikipedia page, but now I am learning that some may use the Wikipedia Commons in this way, which I am confused about. Jackemeyer (talk) 07:01, 11 March 2016 (UTC)

Looking at the top of the page of Category:Wikipedia statistics, you'll see why it brought you to non-mainspace pages. It's a maintenance category used by the project to improve itself. It might be an issue that some mainspace articles, like List of Wikipedias, are listed under this category. This confusion isn't something that happens often, though, as far as I'm aware. ~Mable (chat) 07:42, 11 March 2016 (UTC)
Looking at this section, should this issue be fixed? ~Mable (chat) 07:51, 11 March 2016 (UTC)
Regardless, the only real guideline that should be kept in mind for pages such as this one are the biography of living persons guidelines. As long as the page isn't directly offensive, it's fine if it contains opinions. The page could probably use a bit more political neutrality, but you're free to copy-edit that in if you feel the opinions are unfair. ~Mable (chat) 20:09, 5 March 2016 (UTC)

Thanks! I too thought I was free to copy-edit (which was exactly my intention), and I thought my copy-edit was far superior to what had been thrown together previously. In aiming to capture the sentiment in Wikipedia's own description, "...the "five C's" summarize the copy editor's job, which is to make the copy 'clear, correct, concise, comprehensible, and consistent.' According to one guide, copy editors should 'make it say what it means, and mean what it says'." I think a review of my work will show this, and that even if we fully disregard my intention, the outcome is very nice. Jackemeyer (talk) 07:01, 11 March 2016 (UTC)

Actually, when I've had issues with statements made in the Top 25, I've often posted a comment on the talk page and I usually get a quick response. Liz Read! Talk! 23:36, 6 March 2016 (UTC)

Agreed, and thanks for the idea -- if I do return in the future, I will do exactly this. Especially since it's a non-Wikipedia-article posing as a Wikipedia article page. I do see that the creators are friendly responders, but obviously with a far different view about the purpose of contributing to "Wikipedia: The Free Encyclopedia". Jackemeyer (talk) 07:01, 11 March 2016 (UTC)

I'm sorry but I have to take issue with the idea that this article is posing as a mainspace article; it is doing nothing of the sort, as evidenced by the fact that it has the word "Wikipedia, colon" in front of it. Serendipodous 12:07, 11 March 2016 (UTC)
I agree that it "poses" as a mainspace article just as much as any other maintenance or community page. The lay-out is non-standard for Wikipedia articles, the page can't be found through regular searching and the page is in the "Wikipedia:" mainspace. Wikipedia:Notability doesn't "pose as a Wikipedia article" either, yet it has the same qualities. ~Mable (chat) 12:39, 11 March 2016 (UTC)
Jackemeyer, you didn't fix a typo or correct punctuation. You rewrote some of the article descriptions on the list. I don't think that would be tolerated on any Signpost page where there is a main author. While the author doesn't "own" the Signpost articles they write, like any newspaper-like publication, they do have primary responsibility for the content the page contains and you shouldn't rewrite their words to suit your point of view. I also think that as an editor with 11 edits, your time might be better spent contributing your copy editing skills to article-space. You might consider joining Wikipedia:WikiProject Guild of Copy Editors if you want to focus on copy editing articles. Liz Read! Talk! 13:14, 11 March 2016 (UTC)

New report up

On with the show. Serendipodous 19:49, 7 March 2016 (UTC)

Referring to the week of 15 March, may I just say that Milowent's recreation of Unfinished portrait of Franklin D. Roosevelt is gorgeous? XD Very nicely done. I too am curious why Modern Warfare 2 made the list, I hope someone can figure that out. ~Mable (chat) 19:41, 15 March 2016 (UTC)

Does this page really need to have a flagrant anti-Trump bias?

The notes on Donald Trump are incredibly biased. I think we can do better than turning this page into a rag about the "latest exploits" about "bizarre" Donald Trump. Doorzki (talk) 09:00, 16 March 2016 (UTC)

NPOV may not apply, but toning it down might be a good idea. As a European person, I don't understand much of his political opinions (all the news I see about him is pretty ridiculous), but surely, there must be some political thing about him? On a sidenote, "bizarre" refers to the entire Presidential campaign. Does this include all the other candidates? If so, it doesn't have to be a satirical statement about Trump. ~Mable (chat) 09:22, 16 March 2016 (UTC)

I tried to tone down the sensation and remove the bias but I was reverted. NPOV or not, I don't think this page should be edited in bad faith to promote one's political opinions as it is now. If you want to blog, go set up WordPress and have at it. This behavior of the revertor is both unpleasant to people who care about Wikipedia and unfortunate to users who want to hold Wikipedia's content in high regard. Doorzki (talk) 09:37, 16 March 2016 (UTC)

This article is the expressed view of User:Milowent. To change it is to falsify his opinions. If it is to be changed, then it is to be changed by Milowent. Serendipodous 09:47, 16 March 2016 (UTC)
With that I agree, which is why I'd prefer to wait for their input. Trump isn't showing any signs of going down in this chart and I expect him to be in the top for months to come: we should decide how to describe him here in general, in a way that's both interesting and fair. I'm happy to read "what he's been up to" and any controversy he's sparked is part of the reason he is in the list. We shouldn't ignore that: He doesn't get 8 million views just because he's a popular presidential candidate. ~Mable (chat) 09:54, 16 March 2016 (UTC)
  • Donald Trump is a menace, to be sure. Even many Republicans say that regularly. No one gets 8 million views a week just by being a political candidate. But even here I believe my comments were well-warranted by the press coverage of the week at issue. I don't say "he's a menace"; I don't even think the Levandowski thing is a big deal, but it was heavily covered in the press. But if I was just to say "yeah, donald is popular gee whiz," I couldn't live with myself.--Milowent • hasspoken 15:23, 16 March 2016 (UTC)
First of all, no one should be forced to know whether or not you think Trump is a "menace." Second, it is not helpful to hear your limited, personal conclusions of press coverage which (since you are just one person and can't consume even a small fraction of all press) is biased by your choice of (possibly anti-Trump) media. I assure you there is so much press coverage of Trump that one could fill his or her day with any range or extreme of positive or negative coverage by choosing sources with various degrees of favorability towards or against Trump. What you consume is up to you and is your right, but it does not give you the capacity to perform some grand analysis of all press coverage that can result in your determining of dominant opinion. Now, that doesn't mean you can't pick out a few topics you saw often in multiple sources and linking to pages about those topics to enhance the value of Top 25. It just means you shouldn't inject your own conclusions because those will only reflect your bias. Now, you don't use the word "menace," but you do characterize Trump's activities as "shenanigans" which has a similar impact. You use the word "manhandled" to describe the alleged assault. Manhandling is battery, a considerably greater offence than assault. You call Trump's activities "exploits" and "bizaare." I think these statements speak for themselves in reflecting your bias. What really bothers me though is the childish arrogance to carve out NPOV exemptions on something like a statistics page. Whether or not it is sourced from opinionated work, the opinions should be left behind when the Top 25 are posted here. There is no context to convey the Top 25 is a piece of political commentary. Including the political bias here is deceptive to the reader. I beleive these attitudes are why so many people say Wikipedia has a liberal bias. It's attitudes like these that earn this site such an unfortunate reputation. The worst thing about what you're doing is the enormous disservice to the majority of editors who care about this site being created in good faith. No one is saying you have to praise Trump by calling him a "whiz" or anything else that's driven by any opinion. In fact, that would be just as bad. Just tell the truth and stop coloring something like a Top 25 statistics page with your (admitted) emotionally-driven bias. I'm going to look into how I can bring this up to the larger Wikipedia community. In my opinion, this attitude is an injustice to Wikipedia. I don't think people should be allowed to wreck Wikipedia because they want to be king of the snow pile. And to be clear, I've looked at previous Top 25 pages and they are worse that this one so this isn't only an attack on you. There is a serious problem of multiple editors distorting Wikipedia for the the gain of their political opinions. This is shameful. Doorzki (talk) 18:58, 16 March 2016 (UTC)
Sheesh dude. You're talking about a single article that isn't even in the mainspace and isn't read by anyone outside the community. The idea that it could in any way "wreck Wikipedia" is ludicrous. Second, if you don't like reading other people's opinions in print, don't. No one is forcing anyone to do anything. The Top 25 has had a little latitude to express personal opinions because without them it would be monumentally boring to read. If there are factual inaccuracies that cross the line from semantics to libel, then fine. That's an issue. But don't complain about people expressing their opinions. Serendipodous 19:23, 16 March 2016 (UTC)
The particular location of this article and the lack of viewership do not justify the degree of forced bias. I am not complaining about a little bit of personal opinion. The wording here is strong enough to be considered an attempt at outright political persuasion. (Previous editions are even worse.) The wording here is flagrant. Doorzki (talk) 19:56, 16 March 2016 (UTC)
Actually, the location of this article matters a great deal. It is not in the mainspace; it has a place in the Signpost; a publication that is crammed with often vitriolic personal opinion. There is no way to express a personal opinion about politics without persuading someone of something, even if it is that you are wrong, so I fail to see the distinction you are making. Serendipodous 19:59, 16 March 2016 (UTC)
Well yes, I agree with you, although one of the problems the Signpost has though is it mixes straight reporting and opinion without good demarcation. Alanscottwalker (talk) 20:15, 16 March 2016 (UTC)
The Signpost != the New York Times. It's more like the Guardian in many ways. The traffic report is awesome and has been a valued and treasured part of the Signpost' since we started syndicating it three years ago (jesus, has it really been that long?). Ed [talk] [majestic titan] 20:28, 16 March 2016 (UTC)
I'm not talking about a list in an "often vitriolic personal opinion" publication, but "Top 25 Report" located at Wikipedia:Top_25_Report, with absolutely no context to convey the intent of this content to be politically persuasive. Your justification for the bias seems to be based on the goal of preserving the source's bias in a seemingly unbiased context. That's not fair. You've taken a biased source, re-titled it to give the impression of a scientific approach, and maintain the integrity of the original bias within the distorted context. It's like taking a piece of newspaper editorial content and changing the big, bold "Opinions" title at the top to "News." If you think there is value to cloning the original content, why do you have to remove the context? Why do you have to use a title that misleads the reader? Why can't you make it clear this isn't meant as a statistical analysis but rather possible "vitriolic personal opinion?" Doorzki (talk) 20:29, 16 March 2016 (UTC)
This is ridiculous. This is a longstanding page in the Wikipedia namespace, not the mainspace. Ed [talk] [majestic titan] 20:31, 16 March 2016 (UTC)

Oh for God's sake! This article is not in the mainspace! I don't know how many different ways I can say that! And where does it suggest that this is or has ever been about pure statistical analysis? Where have I or anyone else suggested this is meant to be objective? If you want pure statistics, then you don't need a writer for that; an automated bot would do just as well. Serendipodous 20:37, 16 March 2016 (UTC)

I'm starting to wonder if this page needs something similar to what Template:Essay does for essays: make abundantly clear that the page may contain opinions. Though I suppose the phrase "As prepared by" is supposed to do this... and pages of the Signpost don't do this either even though they may also include opinions... Yeah, I really don't see the issue with how it is now. ~Mable (chat) 20:47, 16 March 2016 (UTC)
I occasionally see descriptions of subjects in the Top 10 or Top 25 that make me wince (often about gender issues), but it is clear that the descriptions are subjective opinion laced with sarcastic humor. This is not new, it's been the way the list has been written since it began over three years ago. Serendipodous and Milowent will call a movie crap if they didn't like it, they will say that Kanye West's antics are over-the-top, they will express bewilderment that some cultural pop icon or product is so popular when they think that fame is undeserved. I've seen politicians of every persuasion ridiculed. The only objective aspect of the page are the numbers. That's the design of the page.
And most importantly, Doorzki, there does exist a page that does exactly what you want (facts and just the facts) and that is the Top 5000 at User:West.andrew.g/Popular pages. I recommend you check out that page if you are looking for information about the most viewed pages on Wikipedia if you find yourself disagreeing with the authors of this page. I sincerely doubt that casual readers of Wikipedia will ever stumble across these pages and so is unlikely to affect Wikipedia's "reputation." Liz Read! Talk! 21:04, 16 March 2016 (UTC)
  • LOL to all this. If there is anyone who can handle my commentary, it is Donald Trump. It has been proven many times that I am always right in my observations; Donald Trump is a menace, and a majority of Republicans agree.[1]; people are free to disagree with my comments, however, and discuss them on the talk page here, or on the comments section of the signpost. Disagreement and discussion is fruitful. But trying to prevent the expression of opinion on a project devoted to analyzing the reasons for the popularity of wikipedia articles cannot be countenanced. Readers 1800 years from now will be enriched by our commentary just as we are enriched by True History, a satire that informs us greatly about the pop culture reading habits of the 2nd century A.D.--Milowent • hasspoken 22:06, 16 March 2016 (UTC)
Agree, generally. The only suggestion I would make, for clarity is "prepared with commentary by" in the header. Alanscottwalker (talk) 10:05, 17 March 2016 (UTC)

New report up

have at it. Serendipodous 08:41, 23 March 2016 (UTC)

Great report - I'm really loving it this week :3 No one ever figured out how Call of Duty: Modern Warfare II ended up so high in last week's report? ~Mable (chat) 10:18, 23 March 2016 (UTC)
Nope. It's always the ones that hide in plain sight that are the hardest to trace. Serendipodous 10:25, 23 March 2016 (UTC)

Garland

Hatch said Garland would be a consensus pick in 2010, not 2016.[2] This discussion of the so-called "Biden Rule" is also misleading; just because the term didn't exist before this year doesn't mean the concept didn't exist before this year. It's fine to campaign for the Garland nomination, but maybe not in Wikipedia's voice?Anythingyouwant (talk) 14:04, 23 March 2016 (UTC)

He said it in 2010, and he also said it last week. And the concept may have been in the air before this year but it can't really be called a "rule" if the guy who came up with it doesn't call it one. Sheesh. I can't wait for this election cycle to be over just so people will stop complaining about political bias. Serendipodous 16:07, 23 March 2016 (UTC)
  • As an American citizen who has followed 20+ years of Supreme Court nominations, I never heard of the "concept" of the Biden Rule.[3]. It is a politically expedient creation; calling it such is not a bias, it happens all the time.--Milowent • hasspoken 16:27, 23 March 2016 (UTC)
  • The material here at this Top 25 Report about Senator Hatch is pure editorializing, not neutral at all. This Report says: "Five days before Barack Obama named this federal judge from his home turf of Chicago his nominee for Justice of the Supreme Court, Republican senator Orrin Hatch cited him as a consensus candidate that Obama would never choose. Of course, the fact that he did choose him has not stopped Senate Republicans blocking his nomination until the next president is elected...." This is editorializing by Wikipedia. Obviously, the "of course" is a snide slam at Republicans, the implication being that Republicans went back on what Hatch had said. More importantly, this Report implies that Hatch offered on March 11 to support Garland as a "consensus" nominee, when in fact Hatch did exactly the opposite. On March 11, Hatch said: "The President told me several times he's going to name a moderate, but I don't believe him ... [Obama] could easily name Merrick Garland, who is a fine man. He probably won’t do that because this appointment is about the election. So I’m pretty sure he’ll name someone the [liberal Democratic base] wants."[4] So, obviously, when Hatch referred to who the President could "easily name", he was referring to who the president could easily nominate, not who the president could easily get confirmed. This was reiterated by Hatch himself, who also said on March 11 that refusal to now consider any Obama nominee to the high court was "the chickens coming home to roost", citing historical episodes as well as old quotations from Democratic senators to explain why.[5][6] To keep this discussion focused, I will save comments about the "Biden Rule" until the bogus information about Hatch is corrected.Anythingyouwant (talk) 18:10, 23 March 2016 (UTC)
    • Can we write with big flashy letters "OPINIONS" already? ~Mable (chat) 18:30, 23 March 2016 (UTC)
      • I just did. I frigging hate disclaimers, but this is getting ridiculous. Serendipodous 18:42, 23 March 2016 (UTC)
        • Per WP:BLP, "BLP applies to all material about living persons anywhere on Wikipedia, including talk pages, edit summaries, user pages, images, categories, lists, article titles and drafts....The BLP policy also applies to user and user talk pages. The single exception is that users may make any claim they wish about themselves in their user space...."Anythingyouwant (talk) 18:47, 23 March 2016 (UTC)
I don't really understand your problem. Hatch said that Garland was a fine man and a moderate that Obama could nominate but wouldn't. And when Obama did, the Republicans still refused to consider confirming him. That's all the article says. I never said that Hatch intended to confirm him. Serendipodous 18:55, 23 March 2016 (UTC)
Per your edit summary at 19:01, "I think the word at issue here is 'consensus'; Hatch said he would be a moderate. So yeah. Slight jump too far. But one wonders why he bothered mentioning it if he never intended to confirm him anyway". That's a big improvement, thanks. If you're wondering why Hatch bothered mentioning it, the answer is probably that Hatch contemplated voting for Garland in December 2016, if Democrats win the presidential election.[7] Also, Hatch didn't say Obama would "never" nominate Garland, he said "probably" and "pretty sure". Sorry to spoil the narrative.Anythingyouwant (talk) 19:19, 23 March 2016 (UTC)
  • The death throes of the Republican party cannot be avoided by arguing over the content of the Top 25 Report. In trying to decipher why Merrick Garland was the third-most popular article of the week, it is obviously closely related the the Republican position that they won't take up the nomination. Please stop attacking Orrin Hatch. Thank you.--Milowent • hasspoken 20:15, 23 March 2016 (UTC)
I make lots of Wikipedia edits unrelated to any death throes, and this is one of many.Anythingyouwant (talk) 20:52, 23 March 2016 (UTC)

Portals

FYI, I have started a discussion here regarding whether this page is like a portal.Anythingyouwant (talk) 01:27, 24 March 2016 (UTC)

  • We are not a portal. What do you hate about Western cultural norms of rational discourse? We are trying to analyze the popularity of wikipedia articles, not spend our time on the talk pages of American political candidates. We look at the popularity of subjects across the spectrum, which come from all countries, cultures, and interests.--Milowent • hasspoken 02:39, 24 March 2016 (UTC)

New report

Loving it as always. I'm happy to see Trump not being number one for a change, simply for the sake of variation. It feels like comics is an even stronger theme than ever this week, though I suppose I wasn't here when The Avengers was released :p Either way, nicely written! ~Mable (chat) 05:36, 29 March 2016 (UTC)

TopViews Report

This Signpost talk page discussion, Wikipedia_talk:Wikipedia_Signpost#Topviews_Analysis_live, may be of interest to statistics folks.--Milowent • hasspoken 21:35, 4 April 2016 (UTC)

A Welcome Return To Pop Culture and Death (March 27 - April 2, 2016)

New report is up. I think #23 is the most interesting - a South Korean TV show.--Milowent • hasspoken 16:33, 4 April 2016 (UTC)

Ooh, that's definitely an interesting one. Have Asian media come up before in the Top 25, besides anime? It's interesting to see Korea and China making strong usage of Wikipedia, and I think it shows that articles on Asian media should be of top quality as well ^_^ ~Mable (chat) 17:34, 4 April 2016 (UTC)
I am not aware of any South Korean TV show making the Top 25 before -- the springboard of being popular in China is the key here. In fact, I can now see the article was the most viewed article of the week on the Chinese wikipedia, using the new TopViews tool. Its #4 here, and the top 3 are the main page, search page, and bot-driven spike to Harry S. Truman.--Milowent • hasspoken 21:42, 4 April 2016 (UTC)
  • Bonus report!: Wikipedia:Top 25 Report/March 27 to April 2, 2016 China.--Milowent • hasspoken 14:33, 5 April 2016 (UTC)
    • Haha, this is great – and useful too! If we are getting a larger Chinese readerbase, we should probably improve articles popular on the Chinese Wikipedia more as well! ~Mable (chat) 14:35, 5 April 2016 (UTC)
      • A caveat about Chinese Wikipedia though, it's blocked in China, so the pageviews are skewed towards topics that are of interest to people in Taiwan and Hong Kong. You may also notice that the ranking is lacking topics from Mainland China save one, which is unthinkable considering the population of China if not for the block. _dk (talk) 17:57, 5 April 2016 (UTC)
Good point _dk, I was wondering about that, maybe I should add a note about that. Chinese_Wikipedia#Origin_of_readers says that 36.6% of readers come from China, which seemed high to me? I got the impression, though, that, Descendants of the Sun is playing in China, which is responsible for slots 1, 3, and 9.--Milowent • hasspoken 20:53, 5 April 2016 (UTC)
The latest numbers for China seem to be down to 1.3% probably because of the block. Due to the prevalence of internet streaming in the Chinese speaking world, I imagine that the Korean drama would be popular in Hong Kong and Taiwan too, and I suspect this is also why the Chinese show I am a Singer is on there. (Actually, what got me suspicious about the list in the first place was the HK drama in #5 - The Chinese audience usually doesn't really care about HK TV dramas, and this one doesn't seem to be making any waves in particular.) _dk (talk) 21:22, 5 April 2016 (UTC)

A Fortune, A Leak, A Scandal: Panama

New Report up. Thanks to Seren for tracking down #16, I have no idea how he found that.--Milowent • hasspoken 14:42, 13 April 2016 (UTC)

For the record, I searched for English language references during the time of the spike and saw the Reddit thread. Since the viewing patterns suggested a Reddit thread, I went with that. I could have been wrong, of course. Serendipodous 16:29, 13 April 2016 (UTC)

Gap Finder Tool between different language wikipedias

This looks interesting - [9] From About section: "... we have developed Wikipedia GapFinder, an app that helps you find missing content in any language for which there is a Wikipedia edition. GapFinder can help you to easily find articles to create in the language of your choice. It also lets you personalize the recommendations by providing a seed article, an article that you would like to receive similar missing article recommendations. GapFinder is a research app at the moment. By using it, you will make more content available in your local language, and help us understand how we can improve the app. Please share your feedback on the tool's talk page.".--Milowent • hasspoken 16:06, 27 April 2016 (UTC)

Also "Find, Prioritize, and Recommend: An article recommendation system to fill knowledge gaps across Wikipedia" on the Wikimedia Blog. :-) Ed Erhart (WMF) (talk) 18:07, 27 April 2016 (UTC)
Thanks for the link to that post, Ed! Now I am worried why I got de:Pädophilenbewegung on my list of German-to-English targets. It appears to be some kind of pro-pedophile movement, and I suspect the English wikipedia has nixed that framing. I guess because I don't shy from controversial topics, the algorithm could include it.--Milowent • hasspoken 11:32, 28 April 2016 (UTC)

new report is up

have at it. Serendipodous 19:58, 1 May 2016 (UTC)

Isn't Civil War out yet in the United States? I watched it in the cinema in the Netherlands yesterday, and I know someone in Brazil saw it on Saturday. ~Mable (chat) 04:57, 2 May 2016 (UTC)
Nope. Not till next week. Serendipodous 06:25, 2 May 2016 (UTC)

The Donald

It's ironic that now that it looks like he has the presidential nomination all sewed up, Trump is flagging in views. Trump fatigue? Maybe people are realizing it isn't a joke any more. Liz Read! Talk! 20:59, 7 May 2016 (UTC)

Keep in mind this list has a time delay. We won't know how our viewers take it til tomorrow. Serendipodous 21:04, 7 May 2016 (UTC)

Underdogs (may 1- 7, 2016)

New report is up!--Milowent • hasspoken 04:48, 10 May 2016 (UTC)

WP:5000 shows XHamster in third place behind Cinco de Mayo. It's almost at the 95% threshold for mobile views, but not quite... is there a reason this is excluded? Thanks  — Amakuru (talk) 15:17, 10 May 2016 (UTC)
Because it is always on this list, and is often above the threshold. Serendipodous 16:04, 10 May 2016 (UTC)

New report is up

have at it Serendipodous 19:37, 15 May 2016 (UTC)

New report: Slow Ride (May 15-21, 2016)

Now up.--Milowent • hasspoken 15:55, 24 May 2016 (UTC)

June 8

The Hess triangle popped up in this Youtube video by Matt Parker this week, which got me to look at it. The viewcount doesn't match up at all, but it may be related. ~Mable (chat) 04:58, 8 June 2016 (UTC)

  • Mable, thanks for the link! Interesting video. The article had most of its views on June 2, two days bfore the Parker video. I finally found the reddit thread: [10]! Searching reddit itself is terrible, using <site:reddit.com "hess triangle"> on Google worked.--Milowent • hasspoken 17:51, 8 June 2016 (UTC)
    • Quite an odd coincidence from my perspective. The Parker video was the first time I heard of the triangle. Glad you managed to find it, anyway :) ~Mable (chat) 19:53, 8 June 2016 (UTC)
You motivated me! It is an odd coincidence because clearly he didn't just film the video the day before or something.--Milowent • hasspoken 21:51, 8 June 2016 (UTC)

New report up

Have at it. Serendipodous 15:47, 14 June 2016 (UTC)

Yay I got the referendum

jump for joy. Serendipodous 04:34, 27 June 2016 (UTC)

  • lol. Well you were more qualified. My commentary would only contain the curses of a Scottish friend.--Milowent • hasspoken 04:08, 28 June 2016 (UTC)
    • The referendum will probably stay of interest for another week regardless, though you got all of the related articles, and all of them were an interesting read. ~Mable (chat) 12:05, 28 June 2016 (UTC)

Sports, anyone?

New report up. Seren, I think the Signpost is going to run bi-weekly in general more often than not, so we should plan to consolidate reports as we've been doing unless you have other suggestions.--Milowent • hasspoken 15:24, 6 July 2016 (UTC)

Week split

Something I thought was slightly interesting. Over the last week, the actress Caroline Aherne (who died on 2 July) got 716,991 views, which would have put her in at least 11th place... but because she died on a Saturday, the views were split across two weeks. This must happen fairly often – are there other prominent celebrities or events that have slipped off the list for the same reason? Smurrayinchester 10:47, 7 July 2016 (UTC)

It does happen, though we tend to notice it more when the the start of the week coincides with a big event, such as the death of David Bowie, which sends views into the stratosphere. The real issue is that Britain just doesn't have enough people to regularly get its topics into the Top 25. Serendipodous 11:50, 7 July 2016 (UTC)
  • This definitely happens. I questioned whether Muhammad Ali would lead the chart when his death went public on a Saturday, the last day of the report's coverage but he easily did. The bulk of views surrounding a death usually occur in the 1-3 days after it is announced, on a continual downward trend which usually tails off by 7-10 days (these are rough estimates). So a less high-profile death near the end of a week can easily not make the Top 25 in either week.--Milowent • hasspoken 12:35, 7 July 2016 (UTC)

Talk page banners?

It would be great of the talk pages of articles appearing in these reports could be added with a banner noting their appearance. Is there a way to have this happen automatically? ---Another Believer (Talk) 02:32, 7 July 2016 (UTC)

@Another Believer: We did it manually for a while (with variable combinations of posts and banners, if my memory serves me correctly). It was not appreciated by everyone and I think the most aggressive of these suggested an RFC. I remain open-minded, but see what you can find in the archive. Thanks, West.andrew.g (talk) 03:43, 7 July 2016 (UTC)
(edit conflict) We did that for a while, but there were some objections. By a narrow interpretation of what a talk page should be, a place to discuss improvements to the associated article, this didn't fit. For reference see Wikipedia talk:Top 25 Report/Archive 3#Negative feedback on talk page notice. It looked like there was a need for a formal request for comment with some wider input to see if that it would be ok. If someone has the gumption to pursue such an RfC I'd be happy to go back to posting the notices. An example of a typical notice is at: https://en.wikipedia.org/w/index.php?title=Talk:Hedy_Lamarr&diff=691246752&oldid=690509161.  SchreiberBike | ⌨  03:49, 7 July 2016 (UTC)
  • This has been in the back of my mind because I liked the talk page notices. But doing an RfC hasn't been at the top of my interest list. I think we'd need to propose a pretty limited use of the banner.--Milowent • hasspoken 12:06, 7 July 2016 (UTC)
  • Seriously? Community resistance to having talk page banners about article traffic? *eye roll* I'm not sure I want to initiate an RfC, but I would not let a vocal minority stop the addition to talk page banners. I think they are informative and helpful for users. ---Another Believer (Talk) 14:40, 7 July 2016 (UTC)

new report up

Have at it. Serendipodous 18:47, 12 July 2016 (UTC)

Can I say, I really hope the NBA season is finally over now, because every time I read the word "All-Star" I have an irresistible urge to listen to "All Star" by Smash Mouth. It never ends well. Serendipodous 20:09, 12 July 2016 (UTC)

It's over, but tonight is the 2016 Major League Baseball All-Star Game. Enjoy. – Muboshgu (talk) 20:24, 12 July 2016 (UTC)

Football pedantry

Just a quick bit of pedantry - Iceland actually went out to France in the quarter-finals. The team Les Bleus beat in the semi-finals were the heavily-favoured Germany. Smurrayinchester 07:34, 13 July 2016 (UTC)

I gave it a bit of a revision. Serendipodous 16:54, 13 July 2016 (UTC)

Esports

  • "...though it's only a matter of time before a health official is forced to remind the general public that real people do not get extra lives." What a world we live in!--Milowent • hasspoken 14:53, 14 July 2016 (UTC)
    • Keeping track of all the Pokémon Go news is a lot of fun. ~Mable (chat) 06:50, 15 July 2016 (UTC)

Let it Pokemon Go

New report up - sorry for the delay. Busy week!--Milowent • hasspoken 13:31, 21 July 2016 (UTC)

This is how I always imagine DB Cooper. Serendipodous 13:37, 21 July 2016 (UTC)
  • LOL. I figured he was always long dead, but people love a good story. Speaking of which I forgot to mention I also co-wrote the Signpost In the Media column this week. People love articles that bash Wikipedia.--Milowent • hasspoken 13:42, 21 July 2016 (UTC)
You wrote the report on the Cracked article? I love that piece, gave me a few good laughs :) ~Mable (chat) 23:04, 21 July 2016 (UTC)

Who's doing the analytics now?

User:Jan Winnicki left a message on my talk page asking me why I excluded Earth from the list, despite it having 7.1% mobile views. Now of course I should have explained myself on the exclusions list so my bad. 7.1%, though higher than the usual exclusion threshold, is still low enough to send a red signal. Normally I wouldn't go searching for analytics until the mobile count topped 10%. But the mysterious rise of Earth has become a big enough mystery to be mentioned in the Signpost. Under normal circumstances I would contact User:Ironholds to do a trace, but he hasn't edited since May. Is there anyone else with access I could talk to? Serendipodous 06:51, 22 July 2016 (UTC)

  • user:MusikAnimal (WMF) is actively working on the pageview stats tool, so I strongly suspect they will either have access or be able to point you in the direction of someone who does. Thryduulf (talk) 21:53, 22 July 2016 (UTC)
    Looks like it's the Reading team that would handle this kind of request; Specifically User:HaeB, who I understand was once the editor-in-chief for the Signpost! :) Hope he doesn't mind the ping MusikAnimal (WMF) (talk) 23:27, 22 July 2016 (UTC)
  • The exclusion of Earth is definitely sound. Its daily views hovered around 10K per day until June 15, then jumped into the range of 60K every day for no reason. It has now dropped to 40K per day in last two days. Checking an early June WP:5000[11], Earth (#1168) had 45.67% mobile views. The drop to 7% and less is proof of external non-human sources.--Milowent • hasspoken 19:24, 24 July 2016 (UTC)

Trumpapalooza July 17-23,2016

New report finally up. Have at it. 5 out of the top 10 are Trumps. Good grief.--Milowent • hasspoken 14:13, 29 July 2016 (UTC)

Thanks. Six even! And 8 out of top 14 including Marla Maples. Gap9551 (talk) 17:08, 29 July 2016 (UTC)
Whoops, I noticed the list was updated to make it 6 after the above comment by Milowent. Gap9551 (talk) 17:47, 29 July 2016 (UTC)
I was on the fence about whether to combine the D Jr. views, but I guess it makes more sense as Ujongbakuto revised it. Seren has been better at combining in past instances like this.--Milowent • hasspoken 19:48, 29 July 2016 (UTC)

New report is up

Have at it. Serendipodous 21:51, 1 August 2016 (UTC)

Nice job; I particularly enjoyed reading your take on Pokémon Go. I am curious, though: have other animated series ever made it in the top 25? I'm sure many films have, with companies like Disney frequently releasing big-budget animated films that get world-wide attention, but I can't think of anything else that would. American animated series aren't particularly grand most of the time, while I imagine anime being too niche.. ~Mable (chat) 09:15, 2 August 2016 (UTC)
Brian Griffin got a mention once, when he "died". Legend of Korra showed up on the exclusions list, but that was back when we were still thrashing around in the dark. It probably should have been included. Serendipodous 09:52, 2 August 2016 (UTC)
Ah, that's pretty cool. Thanks :) ~Mable (chat) 10:49, 2 August 2016 (UTC)
On the anime front, Attack on Titan was a mainstay on the Top 25 for much of 2013, and One-Punch Man appeared on the list at least once. Probably some others as well. _dk (talk) 12:47, 2 August 2016 (UTC)
  • I learned "blonde" is the female version of "blond." It is right in the first sentence of blond. This feels like a fundamental though antiquated fact that I have overlooked. Seren, I will add this to the traffic report for the next signpost (which will now have 3 reports, egads).--Milowent • hasspoken 15:06, 2 August 2016 (UTC)

Minor view spike on Aug 8

The Top 25 Report had 894 views on August 8, a jump from the roughly 200-500 average. Not big enough that I can easily figure out where it was mentioned, but it must have been, somewhere. It is always nice when I see people mentioning the report; I really think it should get more traction in the regular press, which loves to report data on pop culture things like movie revenues, TV ratings, etc. Yes, we are only measuring the population that most frequently uses wikipedia, but aside from website-traffic reports (alexa if that's still around, etc.), its one of the few things monitoring the relative popularity of literally EVERYTHING humans have ever done or learned.--Milowent • hasspoken 12:35, 14 August 2016 (UTC)

New report is up.

Did you know only the US is competing in the Olympics? How odd. Serendipodous 20:18, 16 August 2016 (UTC)

Nah, Singapore, India, there are other countries :p ~Mable (chat) 00:13, 17 August 2016 (UTC)

New Report Up (July 31 - Aug 6)

As Seren already knows, we have to use TopViews now only. It is a bit more cumbersome.--Milowent • hasspoken 15:33, 11 August 2016 (UTC)

Is it that time-consuming to list the article rankings? You could just drop them? Ed [talk] [majestic titan] 23:33, 11 August 2016 (UTC)
It's not that; it's sorting the false positives from the real ones. WP:5000 has the mobile percentages, TopViews doesn't. We have to do the arithmetic ourselves. Serendipodous 03:14, 12 August 2016 (UTC)
Sorry, I was responding to "as we no longer have an easy source listing the rating class of each article." The mobile views are a different beast, I know. :-/ Ed [talk] [majestic titan] 03:28, 12 August 2016 (UTC)
I agree that the false positive issue is more important. But even the ratings, you have to click on the talk page of each page, instead of just having them all laid out for you on the WP:5000. I'm lazy! I know the Top 25 could be more automated (not be the likes of me though), but this is a step back. There are definitely people who like to see the ratings information, we added them (or Seren did first) half way through 2013. Would hate to drop now, under the theory someone could do research based on the data we've compiled.--Milowent • hasspoken 04:13, 12 August 2016 (UTC)
I have 6 hours on a plane tomorrow. Looking into the new data source is on my agenda. West.andrew.g (talk) 15:05, 12 August 2016 (UTC)
Sorry Topviews is falling short, it wasn't really designed for this purpose. I do however have plans to add article assessments, and will also look into mobile percentages, but that is further down the road. The first thing I'm doing, which may not be well-received by some, is to restrict the tool to showing top views on a per-day or per-month basis instead of arbitrary ranges (such as a week). Only daily and monthly data are actually provided by the API, so the range feature is a complete hack which results in bad data for lower-ranked pages. It's going to take a bit to implement the new system, but rest assured I won't get rid of the old one until West.andrew.g has his reports working off the new data source, or while you all as a team are still relying on it :) Best — MusikAnimal talk 04:19, 14 August 2016 (UTC)
We're not blaming TopViews. TopViews wasn't designed to make us happy. The Top 25 report is and always has been based on West.Andrew.g's listing. If he can make it work again, great. If he can't, we may have to change the nature of the report. Serendipodous 05:52, 14 August 2016 (UTC)
Quick note, we know that the TopViews report, when checking for a week, only uses views when an article in the Top 1000 for a day. We are able to check an individual articles' stats for a week to see if they match TopViews (and if they are off that means the article wasn't in the top 1000 for one of more days of the week). This is cumbersome stuff, and I didn't do it for last week's report, but it just occurred to me. I am relieved to hear from MusikAnimal that the current TV hack will stay in place for a bit.--Milowent • hasspoken 12:19, 14 August 2016 (UTC)
(←) @Milowent and Serendipodous: One thing you may wish to do: When linking to Topviews (e.g. "...as determined from Topviews Analysis"), use the permalink to that week's data. By default the tool shows the latest. Just hit "Permalink" in the tool and it will update the URL to always go to the data you are viewing — MusikAnimal talk 16:58, 18 August 2016 (UTC)

New data source; documentation help needed

I'm close to having the new data source integrated. I am ingesting data beginning 2016-01-01 up until "now". As part of this process, "top 5000" reports will generate every "Sunday morning" as the process replays through time. I am going to glance at these as a sanity check. For informational purposes, our new data source is the "pagecounts-ez" set as per [12]. Underlying this is the new pageview definition. At current, I am observing the major changes to be:

  1. A 10-20% reduction in reported "desktop" traffic. This is because crawlers/spider traffic is now removed. I am not yet sure if this helpful in identifying inorganic spikes.
  2. I am seeing orders of magnitude more views for "zero" traffic (but impacts are minimal given these numbers were tiny in the first place). I suspect this was mis-reported in the old data.
  3. The new data source contains only views on existing articles. This eliminates our capability to make the "red-links" report; but does make the ingestion process speedier. The top-5000 list will probably still have a couple redlinks due to rare formatting weirdness.

Could I have folks helpful in updating the following three pages in accordance with what I stated above?:

  • WP:5000: Header needs to reflect new reality above  Done
  • WP:TOPMED: Header needs to reflect new reality above  Done
  • WP:TOPRED: Notice needs added to indicate no more updates; see page history for old reports  Done
  • WP:Top_25_Report/About: As appropriate  Done

Thanks, West.andrew.g (talk) 20:01, 19 August 2016 (UTC)

FYI. Ingestion going smoothly. Currently on MAR-07-2016. Each day taking ~20 minutes. Extrapolate from that to determine if you want to wait on the new data for this week's report. Thanks, West.andrew.g (talk) 23:40, 21 August 2016 (UTC)
  • Looks like we are up to April 2016 sometime, so I should probably use TopViews to get started. Excited to see the updated stuff!--Milowent • hasspoken 13:47, 22 August 2016 (UTC)
Done to 2016-04-16 currently; doing about 2.5 months data per 24h. West.andrew.g (talk) 17:37, 22 August 2016 (UTC)
  • Up to 2016-06-04 at the moment. I did a comparison between the original data used to create the May 29-June 4, 2016 Top 25 and the new data. The numbers aren't exact matches, but do seem to be generally close, and valid for our use. The numbers seem slightly higher in the new data, e.g., #1 Muhammed Ali went from 4,425,310 to 4,509,979. Reddit-fueled entries (like Bicyle Race and Hess Triangle in this case), which generally have lower mobile counts, aren't getting any real bump. I also note the disappearance of Mexico City from the old data (2,050,381 views at 0.71% mobile, so we didn't list it) to the new data. Though a dash appears in the new data with 4.5 million views.--Milowent • hasspoken 17:17, 23 August 2016 (UTC)
Briefly... we don't expect the numbers to be exactly the same, getting spider/robot traffic out of the loop is an enhancement. This depends on spiders/crawlers identifying themselves as such to Wikipedia's servers (probably via the UserAgent field). Intentional "attack" traffic and most mis-configured scripts are unlikely to identify themselves in this manner. "Zero" traffic could explain some of the smaller increases; it seems like it was almost completely unreported in the old data. I don't know what the "-" (dash) article represents, but it does exist as such in the underlying data. We could filter it out as an anomaly if we so choose. Thanks, West.andrew.g (talk) 03:09, 24 August 2016 (UTC)
ALL INGESTED... and back on schedule. Let's now focus on the 3 or 4 pages/headers above that needed updated to reflect changes? Thanks, West.andrew.g (talk) 13:44, 25 August 2016 (UTC)
"About" has been revised as well. Also the redlinks header has been revised. Not sure what needs to be changed about WP:5000. Serendipodous 14:36, 25 August 2016 (UTC)

What is the new source for the raw data? That needs to be updated in the WP:5000 header. Serendipodous 14:49, 25 August 2016 (UTC)

@Serendipodous: I tried to provide some detail in the first paragraph of this section. West.andrew.g (talk) 15:28, 25 August 2016 (UTC)

New report: From Phelps to Bolt (Aug 14-20, 2016)

Finally up; using the new data sourced from the WP:5000. Hooray. Read it before Seren scoops me again in 48-72 hours with the next one.--Milowent • hasspoken 21:12, 26 August 2016 (UTC)

Uh, not sure how to answer that. I was wondering, should we up the exclusion limit to below 6%? We've been getting a lot of 5 point somethings recently. Serendipodous 21:14, 27 August 2016 (UTC)
  • I think 6-7% makes sense. Anything below 15% should raise our suspicions -- the only things that seem to even possibly be below 20% these days and be legitimate are Reddit threads and Google doodles. AMGTV should definitely be excluded. Looks like a fun next report -- wrestling beat out the tail end of the Olympics, Harambe up to #6. --Milowent • hasspoken 13:43, 29 August 2016 (UTC)
  • Seren, I think the next Signpost won't publish until tomorrow (Friday) so if you finish up the Top 25 it can be inserted into the Traffic Report draft I started. Cheers.--Milowent • hasspoken 13:22, 1 September 2016 (UTC)
Sorry for the delay. I'm down with an ear infection. I will try to get it done today. Serendipodous 13:54, 1 September 2016 (UTC)
Best wishes, those are nasty and extremely unfun. Ed [talk] [majestic titan] 18:35, 1 September 2016 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Top_25_Report/Archive_4&oldid=1166845010"