(Manual archive list) |
Gender balance
As the globe turns into International Women's day, you may be interested to know that the gender balance for new, preference identifying, editors has been showing a steady, if somewhat volatile, increase. The average percentage of new editors who currently identify as female in their preferences, over the last 18 months is 21.22%.
The graph shows a slightly different statistic: the ratio of female-to-male (new) editors, again using only those identified in their preferences. Here, of course, the ideological target would be 100%, rather than 50%, but the relationship is not linear.
While by no means certain, it is likely that this is an underestimate of the ratio, as females are taught not to gender identify on-line.
The provisional figures this is based on will be available on a Meta sub page shortly.
All the best: Rich Farmbrough, 22:20, 7 March 2015 (UTC).
- About how many new editors identify preference? If it is 50% or even only 30%, these figures would likely be useful. But if it's only 5% or 10%, probably much less so. Smallbones(smalltalk) 22:45, 7 March 2015 (UTC)
- The % varies over time, peaking at round about 6% in 2012, and decreasing to significantly less than 1 today , strictly speaking of accounts, not editors. Nonetheless the absolute numbers are large enough that the central limit theorem enables us to draw statistically significant conclusions. Full figures are now available. All the best: Rich Farmbrough, 23:01, 7 March 2015 (UTC).
- This is certainly an interesting stat, and even if we find out it's biased it could be useful in practice (e.g. if the bias is constant as a %age). In general, looking for new ways to estimate participation of different types of users on Wikipedia should be encouraged: we need to know something about who our editors and readers are. But I don't buy into the CLT working here. The folks who answered would have to be randomly selected for the CLT to apply. It, of course, would be nice to have a story on why the selection method introduces a bias. I'm not sure I believe this particular story, but it might work like this: most male editors assume "male" is normal, so they don't see any benefit in self-identifying as male. The only folks who self-identify "have something to prove", so they might be a mix of militant mens-rights folks, feminists, and perhaps even men self-identifying as women (expressing frustration perhaps). In any case bias could be introduced by other means of selection of who reports.
- Another means of estimating %women editors might be textual analysis. Women use different vocabulary than men. I'm no expert on this, but ... (with lots of assumptions regarding different ages, nationalities, etc) if you could analyze the text on user pages or user talk pages and split a sample of active editors into categories identified as women, as men, and likely a group that can't be identified by vocabulary. Of course you could get a lot of people angry at you doing this! My stats are getting rusty (and I've never worked with textual analysis) but it's probably done using discriminant analysis or factor analysis. Smallbones(smalltalk) 02:40, 8 March 2015 (UTC)
- The % varies over time, peaking at round about 6% in 2012, and decreasing to significantly less than 1 today , strictly speaking of accounts, not editors. Nonetheless the absolute numbers are large enough that the central limit theorem enables us to draw statistically significant conclusions. Full figures are now available. All the best: Rich Farmbrough, 23:01, 7 March 2015 (UTC).
- To be clear: is the ratio reported in the graph above is F/M or F/(F+M)? MarkBernstein (talk) 23:05, 7 March 2015 (UTC)
- The graph is, as captioned, and mentioned in the text, F/M - which I find the more useful figure. The average percentage of 21.22% is F/(F+M), reported here because that is the way the figure has been reported historically, and in which the WMF/Sue Gardner/Jimmy set targets. All the best: Rich Farmbrough, 23:11, 7 March 2015 (UTC).
- The graph is, as captioned, and mentioned in the text, F/M - which I find the more useful figure. The average percentage of 21.22% is F/(F+M), reported here because that is the way the figure has been reported historically, and in which the WMF/Sue Gardner/Jimmy set targets. All the best: Rich Farmbrough, 23:11, 7 March 2015 (UTC).
- To be clear: is the ratio reported in the graph above is F/M or F/(F+M)? MarkBernstein (talk) 23:05, 7 March 2015 (UTC)
Just an aside about this, this is an article in today's Phnom Penh Post about Khmer Wikipedia and a campaign to get more women's and women-authored articles up.Skookum1 (talk) 03:03, 8 March 2015 (UTC)
- (Website currently off-line, http error 522.) All the best: Rich Farmbrough, 17:10, 8 March 2015 (UTC).
- (Website currently off-line, http error 522.) All the best: Rich Farmbrough, 17:10, 8 March 2015 (UTC).
- These numbers are in rough alignment with a recent stat given at the December 2014 WMF metrics meeting involving results of a survey conducted of 96,000 Wikipedia users in 11 countries of the so-called Global South — conducted in 16 languages (pg. 62). A total of 47,000 people ended up completing the survey, which ran both on desktop and mobile platforms. The survey showed that 21% of readers and 20% of contributors identified themselves as female (with another 2% declining to identify either as male or female and 1% selecting "other.") (pg. 64). There is definitely a gender gap but it appears to be something in the neighborhood of 80m:20f rather than 90m:10f. Carrite (talk) 05:13, 8 March 2015 (UTC)
Cullen328 Let's discuss it 05:58, 10 March 2015 (UTC)
- There are a large number of edit-a-thons taking place all over the world this weekend. These are opportunities to do two things - welcome new editors in connection with International Women's Day, and write and improve biographies of notable women, and articles of interest to women. My wife and I will attend one such event in San Francisco tomorrow. (Sorry, Carrite, as I know that you distrust all that emanates from San Francisco). I have expanded one woman's biography today, and am drafting another woman's biography that I hope to move to article space tomorrow. Thanks, Rich Farmbrough, for the trend chart. The key is to welcome and assist new editors who are here in good faith, and also for experienced editors to edit in good faith to address our well-documented systemic biases. Cullen328 Let's discuss it 05:33, 8 March 2015 (UTC)
- I have always believed that welcoming and assisting new users is important, I am not convinced that it will make a huge difference to the gender balance, since we should retain a significant number of additional male editors - which of course is a good thing too.
- Addressing content issues is another matter, and, despite two pieces of good research, one where we still are in a state of ignorance. The proposed new "views per page" which attempts to distinguish between human and automatic page views will help with this.
- I hope to have a preliminary list of ODNB articles on women uploaded shortly - if anyone is looking for something to do, there will be many red-links and a lot of additional information avaiable onmost of the blue links.
- All the best: Rich Farmbrough, 15:50, 8 March 2015 (UTC).
- Thanks [{Ping|Carrite}} for mentioning the "Global South" report, and you are correct that the editor numbers align generally with this result. There was a survey of Internet use in the US "way back when" (i.e. around the time of the WMF/UN University survey that gave the 12% figure still cited today) which showed that readership in the US was even split M/F (and that the figures WMF/UN had promulgated were skewed by selection bias, and should have been around 16%, IIRC).
- Consequently it appears that we have two very different models, which I would interpret as follows:
- In the GN women and girls are as confident in using and enabled to use the Internet and Wikipedia as men and boys in the GN, but much less confident in editing.
- In the GS women and girls are significantly less confident/enabled to use the Internet and Wikipedia, but those that do are almost as confident in editing Wikipedia as men and boys in the GS.
- All the best: Rich Farmbrough, 15:50, 8 March 2015 (UTC).
- Despite this, Wikipedia Gender Gap activists on the Gendergap-l mailing list continue to propagate the apparent error that the female editor level of English Wikipedia is "around 10% according to the most recent editor survey." To reiterate: no, that is wrong. The actual Gender Gap is approximately 80-20. Carrite (talk) 16:41, 9 March 2015 (UTC)
- This is exactly why the Gender Gap Task Force/Gendergap-l is not to be taken seriously. Instead of starting with serious analysis of the magnitude of the problem, followed by testing of ideas of how to solve it, followed by cost-benefit analysis of putting those ideas into practice on a large scale, we time and time again see a facile assumption of the Worst Case Scenario (ignoring contrary evidence), followed by gender war rhetoric and throwing of verbal petrol bombs, accompanied by grant-writing for money-money-money for unproven programs. WMF is complicit in this by allowing the GGTF/GG-l to put the money cart before the analysis horse. Carrite (talk) 16:45, 10 March 2015 (UTC)
- My long standing recommendation for those who complain women aren't *statistically* editing enough, is for those men to stop editing. This has two benefits, first it is clear they have little of value left to contribute, either for the encyclopedia or for themselves and therefore it will move them on to greener pastures. Secondarily, the less they edit, the smaller that statistical gap will be helping to resolve their moaning. 62.254.196.200 (talk) 17:49, 9 March 2015 (UTC)
- It's not just the mailing list. The recent panel discussion at WVU used the 12% figure, this article cites 10% (and extrapolates that "over 80% ... are young, white, predominantly child-free men" actually 51% are over 30 for starters). Think Progress cites 13% "according to a recent study", presumably the UNMERIT uncorrected figures from 2010. It is not helped that out article Gender bias on Wikipedia has the 9/1/80 split in a misleading graph.
- All the best: Rich Farmbrough, 19:32, 9 March 2015 (UTC).
- All the best: Rich Farmbrough, 19:32, 9 March 2015 (UTC).
If one accepts the premise that women feel less welcome than men then it is likely that there will be a selection bias in any group of people that self selects to reveal their gender. It is very possible that the ratio of female editors is higher than it appears due to men being more likely to self identify gender than women. Still, even if you take that into account a rise is still a rise. Chillum 17:20, 9 March 2015 (UTC)
- I've never weighed in on this issue since it's of no importance to me, which is what I'm stopping by to say: "Who cares what percentage of editors are female?" Even if it is important--which I realize it is to many--you have no way to compile realistic statistics since so many editors do not reveal their gender. I sort of do--by saying I answer to "Grandma" halfway down my user page--but I'm generally referred to as "he." Doesn't bither me a bot. Who knows how many of us don't feel a need to shout, "Hey, I'm a woman; take note!" or, "Over 55 here, woot-woot!"
- The hostility I experienced as a new user had nothing to do with gender, but with newbie-biters. Most of the biters assumed I was male. Yopienso (talk) 20:15, 9 March 2015 (UTC)
As far as Grandma's question "who cares?", I think we should all care. If Wikipedia editors are predominantly young men, then our content will be predominantly focused on the interests of young men. We won't be a true global encyclopedia. We'll be leaving out many of the topics of interest to say, 75%, of the world's population.
Perhaps somebody can list or link the statistical evidence on Wikipedia's editorship. I probably haven't kept up as much as I should have. But I have the impression that there was an attempt at a census-type survey about 5 years ago, and a couple of attempts since, as well as some indirect type evidence such as Farmbrough's above. And my general impression of everything I've seen is that women make up about 10-20% of the editorship, with a large margin of error in any particular study. Stats is not something we can fake, or decide by argumentation or consensus. Solid work needs to be done. Is there any reason that the WMF does not do a proper survey of a proper random sample? Please forget the old "ask everybody" census-type approach. Trying to get everybody involves too many problems, e.g. selection bias, cost, and inability to resurvey every couple of years. Concentrating on a much smaller random sample of active editors would allow much more accurate work to be done. It would allow updates every year or two, so that we can see if we are making progress. Smallbones(smalltalk) 02:27, 10 March 2015 (UTC)
- Please provide evidence that "our content is predominantly focused on the interests of young men." I find a great deal that is of interest to me; I just ignore the stuff I'm not interested in, as I assume the young men--and everybody else--likewise do. Yopienso (talk) 06:31, 10 March 2015 (UTC)
- Indeed, it's as easy to say our content is focused on Indian corporate spam.... we are not on paper, there is not limit to the potential 'size'.... subjects with no article (and admittedly, notable women are apparently underrepresented) is a far more meaningful concept than 'subjects with an article'.... the omissions are what we should care about. Reventtalk 07:23, 10 March 2015 (UTC)
- Please name a notable woman who does not have a WP article. Yopienso (talk) 19:08, 10 March 2015 (UTC)
- There's no need to shout, Yopienso. I think you may share a common misunderstanding about the composition of the Wikipedia community. We don't need more female editors because they will create more articles about notable females. We need more female editors because we need to reflect more than just the worldview of young white males. Women may not only have different views on what is important, they may also see things from a different perspective. You've been here for a long time - you know that the same facts can be presented in a number of different ways. If you want to have a neutral encyclopedia, you want to have as much diversity in the community as you can. (By the way, to answer your question off the top of my head, Wikipedia has no biography of Leona Beldini. I'm sure there are many many more, but that's not the point.) Cheers! Mr Muffler (talk) 19:44, 11 March 2015 (UTC)
- Do you think it's mere coincidence there was no response for 24 hours and then a response in half an hour when I raised my voice? But, thanks for answering!
- I was responding to the claim "subjects with no article (and admittedly, notable women are apparently underrepresented)".
- Leona Beldini, notable woman: a petty crook. Hmmm, I guess she may be notable for being bad. None of those crooks are notable (See the lists on the article page.) except those that were independently notable. Deputy mayors don't seem to cut the mustard.Yo Pienso (talk) 20:02, 11 March 2015 (UTC)
- After so many editathons to add all the missing female scientists, artists, etc I imagine that there are far fewer missing articles than there were a few years ago. But that is only a symptom of the problem and correcting it does little or nothing to deal with the root cause (the gender imbalance in the Wikipedia community). As for deputy mayors on Wikipedia, I guess being deputy mayor of Jersey City, New Jersey is probably not as notable as being deputy mayor of New York City (like Rudy Washington) or a world renowned city such as Launceston, Tasmania (like Jeremy Ball). Anyway, you asked, so I answered. If you don't want to find out why Beldini is notable, I'm not bothered. Mr Muffler (talk) 20:25, 11 March 2015 (UTC)
- Here is a list of notable women without articles. EllenCT (talk) 07:05, 19 March 2015 (UTC)
- And we reams of pages GG women. All of the following are much more influential in gaming than the ones we have covered. Hope Cochran, Stephanie Barish, Holly Liu, Jessica Tams, Kiki Wolfkill, Lucy Bradshaw (gaming). --DHeyward (talk) 08:26, 19 March 2015 (UTC)
- There's no need to shout, Yopienso. I think you may share a common misunderstanding about the composition of the Wikipedia community. We don't need more female editors because they will create more articles about notable females. We need more female editors because we need to reflect more than just the worldview of young white males. Women may not only have different views on what is important, they may also see things from a different perspective. You've been here for a long time - you know that the same facts can be presented in a number of different ways. If you want to have a neutral encyclopedia, you want to have as much diversity in the community as you can. (By the way, to answer your question off the top of my head, Wikipedia has no biography of Leona Beldini. I'm sure there are many many more, but that's not the point.) Cheers! Mr Muffler (talk) 19:44, 11 March 2015 (UTC)
- Please name a notable woman who does not have a WP article. Yopienso (talk) 19:08, 10 March 2015 (UTC)
- Indeed, it's as easy to say our content is focused on Indian corporate spam.... we are not on paper, there is not limit to the potential 'size'.... subjects with no article (and admittedly, notable women are apparently underrepresented) is a far more meaningful concept than 'subjects with an article'.... the omissions are what we should care about. Reventtalk 07:23, 10 March 2015 (UTC)
Back to the data
I'd like to go back to the data—many thanks to User:Rich Farmbrough for making it so readily available. Looking at the data, I'm struck by a couple of things. First of all, I'm not sure that a linear model really fits here. When I tried to reproduce the simple linear regression performed by Rich, the adjusted R2 was fairly low, at 0.6484. More intuitively, the data seem to show that female editorship (as measured by available metrics) was fairly stable around 10% until 2009, when it suddenly jumped to ~20% where it has remained. I added a smoothed spline (in red), which again suggests the relationship here is not really linear. These data don't lead me to believe that female editorship is steadily increasing; instead, they lead me to ask: what happened in 2009?
I went back to the data and truncated/removed all of the data points before 2004 (bottom panel). It's a bit arbitrary, but I think it's reasonable since a) the sample sizes before 2004 are tiny, and b) 2004 marked a huge increase in Wikipedia's popularity, readership, and editorship with attendant fundamental changes in the community. Interestingly, the linear relationship becomes much stronger with the truncated data, at least as measured by R2, probably as a result of excluding a bunch of 0's from 2001–2003. But once again, the relationship appears to be bimodal rather than linear, split around the year 2009.
The final thing that jumps out about these data is that they suddenly become very noisy around mid-2013. The increase in unexplained variance is really quite striking, since variance was remarkably low up until 2013. I don't know why the data suddenly become so noisy, but this implies that we should be extremely careful about drawing inferences from short-term changes in this metric, since there's a very high risk of over-interpreting statistical noise rather than responding to actual trends.
I don't have much to add to the points raised above about the metrics. Obviously, we're making the very large assumption that the sample of editors who choose to self-identify is representative of the population of editors as a whole. I love the central limit theorem as much as the next guy, but I don't think that large sample size and the CLT can overcome inherent selection and response biases. Of course, there are no better data available, and I hugely appreciate the work that Rich and others have done with what we have—I just want to make sure we're interpreting these data with the appropriate caveats. MastCell Talk 17:54, 10 March 2015 (UTC)
- I share your doubts about a linear fit, but since my graphing program (Excel if it was on my laptop, open office if it was on my desktop) gave that trend line for free I left it in. There appear to be two jumps. I have posited that the 2009 jump is related to the introduction of edit filters, which meant far less vandalism, and consequently recent changes patrollers being more relaxed. The later jump may relate to things like Teahouse, editathons, coverage of gender balance, etc.
- The volatility is curious, but can perhaps be explained by class registrations and larger editathons. If a large number of students or volunteers are following a step-by-step process that includes registering gender, then the gender balance of that class will affect the month it is in. Possibly sock farms might influence them too, but I find that unlikely.
- As to the question of whether these are good samples, it is difficult to say, but I can think of one systemic reason it may give a high F/M ratio (classes/editathons), and two to suggest it may be low (desire to avoid identifying as female, to avoid negative consequence, and lack of desire to explore all the preferences options).
- All the best: Rich Farmbrough, 19:01, 10 March 2015 (UTC).
- This looks to me as if changes to the Preferences/User Profile entry form (or the relevant help file) may have had an effect. The following was added in early 2009:
"Gender: This option allows you to select a gender (male or female), or unspecified. This option is designed for languages where, unlike English, words, and therefore interface text, can change depending on your gender. As such, this option is not greatly useful for English users at present."
- How was gender entered before that? The comment about gender being irrelevant for English can't have been very helpful, either. Does anyone know when it was changed to the present form, asking whether the user would like to be addressed as he/she etc. Wasn't there also a (somewhat discouraging) warning that the information would be public? --Boson (talk) 19:45, 10 March 2015 (UTC) PS: I see there is now a message (in small print) "This information will be public." I believe it used to (somewhere) say something like "... but not easily found". --Boson (talk) 20:18, 10 March 2015 (UTC)
- Good find. I haven't tracked down the source of the phraseology yet, search doesn't find it in a mediawiki page. I'm checking translatewiki, but there are 24k+ mediawiki messages! Design flaw somewhere I think. All the best: Rich Farmbrough, 01:54, 11 March 2015 (UTC).
- MediaWiki:Prefs-help-gender seems relevant, but the wording looks like the old wordning you refer to. All the best: Rich Farmbrough, 01:58, 11 March 2015 (UTC).
- Good find. I haven't tracked down the source of the phraseology yet, search doesn't find it in a mediawiki page. I'm checking translatewiki, but there are 24k+ mediawiki messages! Design flaw somewhere I think. All the best: Rich Farmbrough, 01:54, 11 March 2015 (UTC).
- It appears that these graphs misrepresent the proportion of women by at least(exactly?) two orders of magnitude. The graphs' claim that less than .25%, or one in 400, new users self-identify as female rather than male is obvious nonsense. Is this just an axis labeling issue?--Noren (talk) 16:03, 11 March 2015 (UTC)
- I think a linear model might be the best one to use here. While you could construct a more complex model which fits the data better there is the problem of over fitting the data. What I would probably do is fit three straight lines, upto 2009, 2009- mid 2013 and mid 2013 onwards. It looks like there are three distinct regimes to the data, making them not really comparable. Is there any reason for the jump in 2013?--Salix alba (talk): 20:02, 11 March 2015 (UTC)
- That's what I'm getting — at least there is some sort of only partially explained major break in 2009 and that it might be most useful to set that as the zero point if one is attempting to fit the data to a meaningful trend line. I'm a little confused also why the data points scatter so much. One would think that the proportions, while dynamic, would be changing slowly but steadily rather than showing broad short-term fluctuations. Carrite (talk) 18:18, 20 March 2015 (UTC)
Wikipedians should come out of their closets
IMHO the main reason for gender disbalance is not the mythical hostility towards women on wiki (as 95+% of editors do not reveal their gender this factor could only affect less than 5% anyway). The main reason is peer pressure in the real world, there wiki editing is somehow acceptable for male students, barely acceptable for female students, weakly unacceptable for adult family men (considered a strange but harmless hobby) and totally unacceptable for adult married women with family and kids. The idea that a mother of a family spends a few hours a day not earning some wages, not attending for the kids, not tending the home but editing wiki - looks like absolutely atrocious to the society. We need good role model both males and especially females. And we especially need adult professional family males and desperately need happy professional adult females with families. How often have you seen a female wikipedian announcing birth of a child? We must press editors to reveal their identities. Unless special circumstances (e.g. life in a dictatorship couuntry, etc.) editors should work under real names. All Arbitrators/Checkusers/Oversighters should have their real life identity and basic bio revealed. All new admin applications should be accompanied by real names and biographies and, yes, I am sure most !voters would give women especially adult women with families some preferences. Alex Bakharev (talk) 07:41, 10 March 2015 (UTC)
- Yes, I'm sure that the 'men' who edit wikipedia are being fed grapes from lustrous virgins as we speak. Could someone remind me why I (p.s. as a woman) would want to reveal my gender when I don't want simply to please some gender identity demagogues of questionable competence and reputation? Why would I ask the same of a man? This suggestion is underthought and uninspiring. Perhaps the childish feminists of wikipedia, who seem to be self-guilt tripping white men need to man up and respect that one of the reasons they don't see many women editing wikipedia is because they wish to turn every woman into a juvenile. Nobody wants to be special objects. This is not acceptable, fundamentally anti-feminist and embarassing to everyone involved. 62.254.196.200 (talk) 10:06, 10 March 2015 (UTC)
- If I could do it over, I sure wouldn't pick a gender-specific user id or otherwise identify as female. I've experienced the same issues that most people complain of, plus stalking, tagteams, mansplaining, and the endless accusations that my politics, because they are actually congruent with the demographic center's preferences[1] instead of skewed to the right as the major political parties skew in their desperate attempt to garner campaign contributions from moneyed interests,[2][3][4] are somehow out of the mainstream. Male-identifying users with my point of view do not, as far as I can see, have to deal with that bullcrap. EllenCT (talk) 16:27, 10 March 2015 (UTC)
- Ya lost me with use of the pejorative "mansplaining." Don't be "femhearing" me now... Carrite (talk) 16:51, 10 March 2015 (UTC)
- I added a wikilink to aid your comprehension. EllenCT (talk) 17:49, 10 March 2015 (UTC)
- Does that mean you wiksplained it? Formerip (talk) 19:19, 10 March 2015 (UTC)
- Do you (Ellen) think that if you would indicate that your technical background makes your understanding of clumsy wiki-syntax and arcane wikirules to be complete (wish I could say so myself) you would still be bothered with offering of help? Maybe you could benefit from some opening of your real life background. Alex Bakharev (talk) 02:22, 11 March 2015 (UTC)
- Alex, I don't mind being talked down to on technical issues because I realize how difficult it is to discern what an editor does not yet understand from a few limited interactions. The reason I use the perjorative to which Carrite objects is because of the several instances where enthusiastic opposition from editors with primary sources on content disputes where I've gone to the trouble to find literature reviews never have involved a single female editor, even when there were female editors on the other side of the issue. If you can show me how to get the guys to stop countering secondary sources with COI-sponsored primary sources, on Talk:Tax policy and economic inequality in the United States for example, then I would be happy to say more about my background and life experiences. But for now, I would simply prefer to not dignify those WP:COMPETENCE and/or WP:IDHT issues with continued engagement, because the enthusiasm without regard for the rules frankly makes me wonder what other rules they are willing to break to suppress the secondary literature's conclusions with which they disagree. EllenCT (talk) 16:49, 11 March 2015 (UTC)
- Do you (Ellen) think that if you would indicate that your technical background makes your understanding of clumsy wiki-syntax and arcane wikirules to be complete (wish I could say so myself) you would still be bothered with offering of help? Maybe you could benefit from some opening of your real life background. Alex Bakharev (talk) 02:22, 11 March 2015 (UTC)
- Ah, you missed something there, I'm well aware of the pejorative neologism "mansplaining." See: Pejorative. Happy to help. Carrite (talk) 02:09, 11 March 2015 (UTC)
- Does that mean you wiksplained it? Formerip (talk) 19:19, 10 March 2015 (UTC)
- I added a wikilink to aid your comprehension. EllenCT (talk) 17:49, 10 March 2015 (UTC)
- I ,for one, am certainly glad that I have not made anything about my real life identity very public. And given the lack of concern from the arbcom about the attacks on those wikipedia editors who did have their public lives targeted because of their Wikipedia editing strikes me as solid gold proof that I would strongly advise against ANYONE doing so, particularly women or people with children. -- TRPoD aka The Red Pen of Doom 17:07, 10 March 2015 (UTC)
- I think the project as the whole is hindered by unnecessary secrecy over identity of major contributors. Corporations spend billions on so called team-building efforts so to encourage their employees to share their informal lives. Most major charities do the same. They maybe ignorant but not stupid. Currently Wikiediting is seen as a shameful activity suitable only for primary school students (preferably male). This perception hurts the project greatly. Luck of positive role models (particularly adult women) hurts us internally. Difficulty in spotting conflicts of interests arising from the identity secrecy hurts us greatly. I am editing for almost 15 years under my real name, and only once regretted, then some idiot with IP geolocated 10 km from my home was making threats about me. Still under real name helped to clear my name over accusations of COI (some very painful), so in long run was beneficial. Alex Bakharev (talk) 02:22, 11 March 2015 (UTC)
- You may consider anonymity "unnecessary". I consider my anonymity on Wikipedia the reason that my family has not been subjected to hateful death threats and SWATTING. When you share personal info with a co-worker you are not sharing it with the whole fucking world that is full of crazies who do things like target Wikipedia editors they decide they dont like.-- TRPoD aka The Red Pen of Doom 02:38, 11 March 2015 (UTC)
- I think the project as the whole is hindered by unnecessary secrecy over identity of major contributors. Corporations spend billions on so called team-building efforts so to encourage their employees to share their informal lives. Most major charities do the same. They maybe ignorant but not stupid. Currently Wikiediting is seen as a shameful activity suitable only for primary school students (preferably male). This perception hurts the project greatly. Luck of positive role models (particularly adult women) hurts us internally. Difficulty in spotting conflicts of interests arising from the identity secrecy hurts us greatly. I am editing for almost 15 years under my real name, and only once regretted, then some idiot with IP geolocated 10 km from my home was making threats about me. Still under real name helped to clear my name over accusations of COI (some very painful), so in long run was beneficial. Alex Bakharev (talk) 02:22, 11 March 2015 (UTC)
- Ya lost me with use of the pejorative "mansplaining." Don't be "femhearing" me now... Carrite (talk) 16:51, 10 March 2015 (UTC)
- If I could do it over, I sure wouldn't pick a gender-specific user id or otherwise identify as female. I've experienced the same issues that most people complain of, plus stalking, tagteams, mansplaining, and the endless accusations that my politics, because they are actually congruent with the demographic center's preferences[1] instead of skewed to the right as the major political parties skew in their desperate attempt to garner campaign contributions from moneyed interests,[2][3][4] are somehow out of the mainstream. Male-identifying users with my point of view do not, as far as I can see, have to deal with that bullcrap. EllenCT (talk) 16:27, 10 March 2015 (UTC)
- I strongly oppose this principle, and would prefer if physical world factors were excluded from the consideration of Wikipedia editors. That was one of the original goals of cyberspace, if one can remember. To break down the walls of reality and replace them with an open space for everyone, regardless of factors that one cannot control, such as the imaginary constructs of sex, race, gender, and so forth. The internet has not worked out that way, but that's what it was meant to be, and what it should strive to be. This demand for people to make a decision as to whether they consider themselves a woman or a man is offensive and absurd. Many of us, for example, consider ourselves neither. What are we to do, then? This obsession with imaginary boundaries, to translate connotations attached to physical principles into the immaterial digital realm, seems to be an attempt by certain people to hold onto a stability of existence that simply does not exist. Neoliberalism really has accomplished its goal, hasn't it? That is, to balkanise the proletariat in every possible way. Instead of allowing the internet to become a space of pleasure in unity and commonality, it has become an expression of artificial difference, controlled by corporations and other unsavoury groups.
- Regardless, I find this whole "controversy", if one can call it that, absurd. The neoliberal authorities that control this encylopaedia have no desire other than to balkanise the proletariat, even in a space that is meant to be based on collaboration. They struggle to enforce imaginary boundaries in a space where those boundaries lack visual representation. More and more, we see this battle playing out. The battle, itself, is exactly what the neoliberal authorities want to see. The more they can divide the proletariat, the less the proletariat can fight back against corporate capture. That's what the heart of the matter is. "Come out of one's closet", and be captured by the camera of corporate surveyors and data-mongers. One must recognise that one is in a corporate space, a space where one produces capital for the ruling class without reaping any benefits for oneself. The producer-consumer, the person who carries the burden of the neoliberal class. That is what this encylopaedia is, of course. A factory for content, data, &c., all of which is produced by unwitting participants for the sake of the gain of a few. We are kept occupied by both producing content and consuming it, and by doing so, entrench their dominance. That's not to say there are no other benefits to the project, and of course those are often the reasons that editors remain. However, until the proletariat can break down the imaginary barriers prescribed by the ruling class as a method of dominance, there can be no true freedom or unity in human singularity. It is our duty, then, to stop this continued balkanisation and madness. It is our duty to say no to categories that do not exist, such as "man" and "woman". It is our duty, then, to defend this space from further encroachment by corporate authorities. It may already be too late, but it does not hurt to be a martyr for one's cause. RGloucester — ☎ 19:38, 10 March 2015 (UTC)
- OK, Alex, I'll take the bait. Short answer: no. Long answer: Do we really want Wikipedians to be even more easily doxxed and targeted by online trolls? No.
- And if the situation is as you claim and women are ashamed of editing here, requiring them to divulge their real identities would only exacerbate the issues. Many people edit here under anonymous names because they would be ashamed if their real identities were known. KonveyorBelt 19:45, 11 March 2015 (UTC)
- There needs to be some accountability for behavior on WP, but with complete anonymity, it will never happen. I edit under my real name and have never had any problems online, but I attribute this to staying away from controversial articles. That's not to say Ive never had problems with other Wikipedians, both male and female, but I pick my battles. I have to laugh at the mansplaining complaint though. Essentially it is a man arguing with a woman the way he argues with other men. In reality, he is treating a woman equally to a man. Seems to me that we women need to develop counter-tactics more effective than demand that everyone act the way we want to them to.Thelmadatter (talk) 19:39, 14 March 2015 (UTC)
Why focus on editors instead of readership?
I don't understand why all the focus is on the low participation rate of women when it comes to editing. Above, someone quotes the statistics that "21% of readers and 20% of contributors" identify as women. Therefore, there is no gender imbalance when it comes to who decides to edit - women are not being scared away from contributing. The same percentage of them who come to Wikipedia to read become editors as do men. Recently the lunch conversation among my coworkers (in a high tech field) turned toward reading habits as kids. Nearly ever male reported reading an encyclopedia for fun as a child, but I was the only woman who did so. Obviously, this is only anecdotal, but along with the statistics, it suggests that if you want more women to contribute to Wikipedia, you first need to get more women to be interested in reading an encyclopedia. Deli nk (talk) 14:13, 10 March 2015 (UTC)
- You may be absolutely right. Do notice, however, that the statistic you cite is for the "Global South", in the US, for example, readership (I beleive the question was something like "have you used Wikipedia in the past month?") from a large independent survey showed roughly equal numbers of males and females using Wikipedia. Of course the way it is used may also vary by gender. All the best: Rich Farmbrough, 18:07, 10 March 2015 (UTC).
- Thanks for the response. I'm a bit surprised at the disparity in the survey results, though maybe I shouldn't be if the questions were worded very differently. I'd be interested in learning more about the way Wikipedia use might vary depending on gender. Deli nk (talk) 18:23, 10 March 2015 (UTC)
- I'm not sure which surveys other people are referring to, but I believe the main survey that showed roughly equal participation by men and women (as readers) was based on a representative sample of Americans (with an online connection and a landline telephone), which incidentally also asked about Wikipedia usage, and was carried out by an established social/market research organization (Pew), while all the other studies were based on "opt-in", self-selecting samples. Personally, I think it would be much, much better if the Foundation commissioned Pew (or another established organization that routinely performs research for people who bet money on the results) to perform a similar survey of Wikipedia readers and editors based on a representative population sample. Doing our own studies with self-selecting samples may be much cheaper but it seems a bit like looking for the lost key in the bushes, in the dark, when you know you dropped it in the road, under the streetlight. --Boson (talk) 21:27, 10 March 2015 (UTC)
- PS: I would imagine it would be simplest to add a few questions (e.g. Wikipedia editing: frequency, topic and medium) to an existing omnibus survey on Internet usage. --Boson (talk) 11:41, 11 March 2015 (UTC)
- I can confirm that the Pew 2008 research which covered readership was used to correct the Wikipedia 2008 readers and editors survey (and apologies if I referred to that as the 2010 survey elsewhere) by Shaw and Hill in 2013. All the best: Rich Farmbrough, 19:37, 14 March 2015 (UTC).
- I'm not sure which surveys other people are referring to, but I believe the main survey that showed roughly equal participation by men and women (as readers) was based on a representative sample of Americans (with an online connection and a landline telephone), which incidentally also asked about Wikipedia usage, and was carried out by an established social/market research organization (Pew), while all the other studies were based on "opt-in", self-selecting samples. Personally, I think it would be much, much better if the Foundation commissioned Pew (or another established organization that routinely performs research for people who bet money on the results) to perform a similar survey of Wikipedia readers and editors based on a representative population sample. Doing our own studies with self-selecting samples may be much cheaper but it seems a bit like looking for the lost key in the bushes, in the dark, when you know you dropped it in the road, under the streetlight. --Boson (talk) 21:27, 10 March 2015 (UTC)
- Thanks for the response. I'm a bit surprised at the disparity in the survey results, though maybe I shouldn't be if the questions were worded very differently. I'd be interested in learning more about the way Wikipedia use might vary depending on gender. Deli nk (talk) 18:23, 10 March 2015 (UTC)
Confidential poll right here :)
Throw a coin twice, if the first outcome is heads then write down your sex, if it is tails, then if the second throw is heads write down "male" for your sex and "female" if it is tails. If the fraction of females is pf, and the fraction of males is pm = 1-pf, then the expectatio value of the the fraction of "females" in the answers is 1/4 + 1/2 pf, allowing one to estimate pf. Count Iblis (talk) 19:51, 10 March 2015 (UTC)
- Female . Count Iblis (talk) 19:51, 10 March 2015 (UTC)
- I am totally confused. Jehochman Talk 02:42, 11 March 2015 (UTC)
- My coin says either "Reply hazy try again" or "You may already have won!" Short Brigade Harvester Boris (talk) 02:51, 11 March 2015 (UTC)
- My coin landed on its edge. So...? — Crisco 1492 (talk) 02:55, 11 March 2015 (UTC)
- I think my coin will land eventually, but I'm currently in a very low gravity environment, so this could take years. ϢereSpielChequers 09:24, 11 March 2015 (UTC)
- My coin turned into a six-sided die, bounced a few hundred times and landed on Jumanji. Neutron (talk) 03:33, 12 March 2015 (UTC)
- How far should I throw the coin? I'd like to get it back if possible. AtHomeIn神戸 (talk) 07:43, 12 March 2015 (UTC)
- I can't actually be bothered to toss a coin. But, as far as I can see, the bigger the sample size, the more reliable the result will be, so I thought I should at least register a response. Formerip (talk) 08:50, 12 March 2015 (UTC)
- Male, but I used a two-headed coin. --B (talk) 01:37, 16 March 2015 (UTC)
- Furry. —Jeremy v^_^v Bori! 04:40, 17 March 2015 (UTC)
Discussion
I strongly approve of the motivating principles of this poll, but I question whether anyone can know how accurate it is if it's self-selected. Why don't you perform the same poll by asking on, for example, the talk pages of the most recent N editors including IPs, so you can use the responding proportion to evaluate significance, please? EllenCT (talk) 20:44, 10 March 2015 (UTC)
- Maybe the NSA could save us a lot of time, and just tell us the answer, already! All the best: Rich Farmbrough, 02:00, 11 March 2015 (UTC).
- Good idea! Count Iblis (talk) 17:16, 11 March 2015 (UTC)
- The only information this poll can provide is the sex ratio of respondents (assuming that people understand it, and give honest answers). It can tell us nothing about the ratio for contributors in general. AndyTheGrump (talk) 03:04, 11 March 2015 (UTC)
- That was my point. If you know the proportion of respondents, you can constrain their population statistics, but not if they are self-selected. I would gladly participate in this form of anonymization if it was taking place on respondents' talk pages. EllenCT (talk) 17:50, 11 March 2015 (UTC)
- It's not confidential either. It reveals the respondent's sex with probability 3/4. 50.0.205.75 (talk) 23:25, 12 March 2015 (UTC)
- That was my point. If you know the proportion of respondents, you can constrain their population statistics, but not if they are self-selected. I would gladly participate in this form of anonymization if it was taking place on respondents' talk pages. EllenCT (talk) 17:50, 11 March 2015 (UTC)
- The only information this poll can provide is the sex ratio of respondents (assuming that people understand it, and give honest answers). It can tell us nothing about the ratio for contributors in general. AndyTheGrump (talk) 03:04, 11 March 2015 (UTC)
This actually depends on the very answer we're after, i.e. the prior probabilities of someone taking part of the poll being male or female. You can easily that if indeed most people contributing here are men, then the poll will be a lot more confidiential for them than for men. We have:
p(answer = f, sex = f) = p(sex = f) p(answer=f|sex=f) = 3/4 p(sex = f)
p(answer = m, sex = f) = p(sex = f) p(answer=f|sex=f) = 1/4 p(sex = f)
p(answer = f, sex = m) = p(sex = m) p(answer=f|sex=m) = 1/4 p(sex = m)
p(answer = m, sex = m) = p(sex = m) p(answer=m|sex=m) = 3/4 p(sex = m)
The prior probabilities for the answers are thus given by:
p(answer = f) = sum over sex of p(answer = f, sex) = 3/4 p(sex = f) + 1/4 p(sex = m)
p(answer = m) = sum over sex of p(answer = m, sex) = 1/4 p(sex = f) + 3/4 p(sex = m)
The conditional probabibilties for the sex given what someone answers are therefore:
p(sex = f|answer = f) = 3/[3 + p(sex = m)/p(sex = f)]
p(sex = m|answer = f) = 1/[ 1 + 3 p(sex = f)/p(sex = m)]
p(sex = f|answer = m) = 1/[1 + 3 p(sex = m)/p(sex = f)]
p(sex = m|answer = m) = 3/[3 + p(sex = f)/p(sex = m)]
So, if the 20% figure for female editors is correct, then:
p(sex = f|answer = f) = 3/7
p(sex = m|answer = f) = 4/7
p(sex = f|answer = m) = 1/13
p(sex = m|answer = m) = 12/13
The amount of information someone reveils by giving an "f" or "m" answer is given by the drop in the Shannon entropy:
I(f) = 3/7 log_2(3/7) + 4/7 log_2(4/7) - 0.2 log_2(0.2) - 0.8 log_2(0.8) = -0.2633 bits
I(m) = 1/13 log_2(1/13) + 12/13 log_2(12/13) - 0.2 log_2(0.2) - 0.8 log_2(0.8) = 0.3307 bits
I(f) is negative because after you give such an answer your sex is pretty much a tossup while a priori it was a relatively safe bat to assume that you were male. So, giving this answer leads to a loss of information. Count Iblis (talk) 18:27, 13 March 2015 (UTC)
Meanwhile create more articles for women
I think it could help both readers/editors to have more topics about female issues. For instance, I just now created redirect "nursing pads" (after 14 years) to link "nursing bra" as a start to answer 3 requests a week about nursing pad/pads. Perhaps start other related articles as redirects, for now, until specific details can be added to create separate articles. -Wikid77 (talk) 16:35, 16 March 2015 (UTC)
Need someone to help with a BLP issue
I need someone fluent in Danish and English to help with a BLP issue that has been brought to my attention. I'm going to post as well on BLPN here, but I'm not sure where to go on Danish Wikipedia to see if someone there would like to help.--Jimbo Wales (talk) 13:11, 16 March 2015 (UTC)
- See Category:User da and Category:Wikipedians who contribute to the Danish Wikipedia
and Category:Wikipedians in Denmark and Wikipedia:WikiProject Denmark#Members. - —Wavelength (talk) 18:32, 16 March 2015 (UTC)
- Wikipedia:Biographies of living persons (WP:BLP) has an interlanguage link to da:Wikipedia:Biografier af levende personer (da:WP:BLP), whose discussion page at this moment shows comments only in Danish.
- —Wavelength (talk) 15:36, 21 March 2015 (UTC)
Status of Kosovo
Greetings Jimbo Wales! I would ask you that the English Wikipedia and Wikimedia Commons still don't recognise the independence of Kosovo as it's at odds with the international law and the UNSC resolution 1244. VS6507 (talk) 23:02, 16 March 2015 (UTC)
- Wikipedia has no mechanism by which to recognize or not recognize anything at all. Collect (talk) 23:05, 16 March 2015 (UTC)
- I would remind VS6507 that wikipedia is supposed to reflect what sources say, rather than maintaining a political ideology. Sources say that Kosovo declared independence and is recognised by most countries. Edits like this - deliberately replacing a current map of Serbia with one that annexes Kosovo to Serbia again - are obvious POV-pushing. I am surprised that wikipedia has tolerated so much of this, from an editor with a long history of cross-wiki canvassing and POV-pushing. Just another day in the Balkans... bobrayner (talk) 23:18, 16 March 2015 (UTC)
- IKR, the Balkans. If the Wikipedia ever blows up it will probably be on account of some damn thing in the Balkans. Herostratus (talk) 02:55, 18 March 2015 (UTC)
- Bobrainer, the most extreme editor always pushing all-around Kosovo as independent, and those evil Serbs annexing Kosovo... And his already boring remark "Just another day in the Balkans"... Even in a simple comment here on Jimbos page he is unable not to push-up the idea that Kosovo is not part of Serbia anymore and Serbia annexed Kosovo. if this continues he is close to be topic banned from editing within the scope. Half of the countries of the world recognize Kosovo as independent, other half still regard it as part of Serbia, so his crying here how the editor is deliberately replacing maps is ridiculous. Basically, his complain against the editor is as just rightfull as his own insertion of maps excluding Kosovo. Oh boby boby, you really don't know where the borderline is, do you? FkpCascais (talk) 04:23, 18 March 2015 (UTC)
- IKR, the Balkans. If the Wikipedia ever blows up it will probably be on account of some damn thing in the Balkans. Herostratus (talk) 02:55, 18 March 2015 (UTC)
- I would remind VS6507 that wikipedia is supposed to reflect what sources say, rather than maintaining a political ideology. Sources say that Kosovo declared independence and is recognised by most countries. Edits like this - deliberately replacing a current map of Serbia with one that annexes Kosovo to Serbia again - are obvious POV-pushing. I am surprised that wikipedia has tolerated so much of this, from an editor with a long history of cross-wiki canvassing and POV-pushing. Just another day in the Balkans... bobrayner (talk) 23:18, 16 March 2015 (UTC)
- There can't be any one correct answer. This is a location map template. That means that it is for encyclopedic use in locating events and places, some of which occurred at some time in the past when Kosovo was unarguably part of Serbia. Because it is a template, there is no Main Version - we can have one that includes Kosovo, one that doesn't, and let each and every article writer decide which to use. That said, because the template is already in-use, I agree it should stick with whichever version was first and the copy should go the other way, to avoid unobserved alterations to the articles that use it. Wnt (talk) 14:35, 18 March 2015 (UTC)
- While we're at it, let's also review the status of Crimea, the Kuril Islands, the South Chinese Sea, Kashmir, Tibet, The Western Sahara, South Ossetia, Abkhazia, Nagorno-Karabakh, etc. etc. Count Iblis (talk) 15:59, 21 March 2015 (UTC)
- Wouldn't the most encyclopedic thing and what we basically do is simply mention in prose the de jure by world standards if different. I think frankly maps cause more problems then they solve in situations of templates on history pages and such. As opposed to most of User:Count Iblis's list (with the exception of Crimea) Kosovo has the problem that it is a VERY important part of the identity and history of the nation building of Serbia, that much is not up to discussion, it's a fact; the problem is that Kosovo is Albanian in culture and language and the people have decided they want to nation-build their people (as opposed to just being "Albanians") and to build a state. Metternich once said about the Italian peninsula that a state can build a nation but a nation can not build a state... Austria-Hungary, the USSR, Czechoslovakia, and Yugoslavia proved him wrong, and Belguim going that direction. Kuril Islands don't hold a nationality defining moment of history for Russia, India and Pakistan have no huge religious meaning to Kashmir, Tibet doesn't define to China what the nation is, same for Western Sahara and Morocco, and Georgia and Azerbaijan with the last three. Crimea is the odd person out in many reasons, it really is Russian majority and it does have meaning to Russian history as does all of Ukraine back to the times of the Rus and Kiev principality. Static maps can not represent the nuances of history, politics, culture, language, a map of North America showing the Louisiana Purchase claims the US "owned" a vast tract of land when in reality all it was, was the French selling the exclusive rights to purchase the land, a map of that period doesnt show who actually controlled the land, the Souix or the Comanche Empire for example. This problem with maps is evident all over China-related history maps where those with a pro-China POV push maps showing China ruling Tibet pretty much forever and never recognize an independent Tibet in the late 19th and early to mid 20th centuries for example. POV pushers will always find a map to push their POV.97.85.247.26 (talk) 17:06, 21 March 2015 (UTC)
New essay
The ArbCom case against me is vexatious indeed - I shall not contend against those who taste blood. The main complaint even includes my essays - so I wrote one which I hope you will appreciate WP:Wikipedia and shipwrights. It would be fun to see how others react, indeed. Warm regards, Collect (talk) 04:22, 20 March 2015 (UTC)
- Though I do not agree 100% with all the implications, it is a good read, and thank you, Collect. Shipwrights are more accepting of those who swear at the bad things in life and in the world, and at our mortal fate, than they are of those who swear hatefully at and about their co-workers. Cullen328 Let's discuss it 04:30, 20 March 2015 (UTC)
- For my many faults, swearing at co-workers is not chief among them. My chief fault - now and always - has been an unremitting belief that biographies of living persons should "do no harm" - the listing of the advert signers from Duke was a clear case to me. And I wish Gibbs (very arcane NCIS reference for everyone) good luck on his boat, as to all editors who seek to build ships which will sail straight and true. Or for another arcane reference - to those who build and sail Masefield's tall ships, and see WP:BLP as a "star to steer her by." Cheers. Collect (talk) 04:38, 20 March 2015 (UTC)
- Couldn't agree more with the do no harm issue. Of particular ugliness is when someone does not like, disagree with or has other non benign reasons for editing a BLP (or any bio for that matter) and argues non stop for inclusion of negative issues and has nothing else they add to the article. Such editors should be BLP banned and if they violate their ban, then be site banned immediately.--MONGO 16:02, 20 March 2015 (UTC)
- Me? I swear at co-workers and my team all the time. In the UK this is no big deal. We try not to be prudes. Guy (Help!) 23:59, 21 March 2015 (UTC)
- For my many faults, swearing at co-workers is not chief among them. My chief fault - now and always - has been an unremitting belief that biographies of living persons should "do no harm" - the listing of the advert signers from Duke was a clear case to me. And I wish Gibbs (very arcane NCIS reference for everyone) good luck on his boat, as to all editors who seek to build ships which will sail straight and true. Or for another arcane reference - to those who build and sail Masefield's tall ships, and see WP:BLP as a "star to steer her by." Cheers. Collect (talk) 04:38, 20 March 2015 (UTC)
- Which ArbCom case is this about? Count Iblis (talk) 15:52, 21 March 2015 (UTC)
- See Wikipedia:Arbitration/Requests/Case. The permanent link to the current version in history is [5]. P.S. If you need such info ever again for anyone, try looking in their contribution list. Regards, --Bob K31416 (talk) 16:29, 21 March 2015 (UTC)
Blast from the past
One thing led to another and I found myself doing some editing on The Big Lebowski, and in the process noticed that you created that article 14 years ago. --Bob K31416 (talk) 16:09, 20 March 2015 (UTC)
- And today he would have been templated all over the place for SUCH A STUB. (There are some at Wikipediocracy who laughably contend that there is no inherent trend at Wikipedia towards improvement...) Carrite (talk) 17:39, 20 March 2015 (UTC)
- No, that would be speedy deleted. Nyth63 22:33, 20 March 2015 (UTC)
- Just for the heck of it, I looked up in our encyclopedia that Wikipedia was 38 days old when that article was created. And hey, "the Dude abides".--Bob K31416 (talk) 01:39, 21 March 2015 (UTC)
- No, that would be speedy deleted. Nyth63 22:33, 20 March 2015 (UTC)
- Rules are different when everything is new, and 38 days old is definitely new. I have to agree with Jimbo on the The Big Lebowski creation gist. Definitely one of the few movies I've seen more than 10 times. Others are The Breakfast Club, Platoon and Breakfast at Tiffany's. Dave Dial (talk) 16:52, 21 March 2015 (UTC)
Reddit IAmA
Just wanted to say, thank you (along with Jaffer and Tretikov) for hosting the AMA on the NSA lawsuit. It's really cool to see a lot of questions being answered, and hopefully a lot of people learn more about it.
For anyone reading this, here's the live Reddit "Ask Me Anything" where Jimbo and company are answering questions related to the suit. ~SuperHamster Talk Contribs 16:25, 20 March 2015 (UTC)
- You were fabulous, Mr Wales! Rauan Kenzhekhanuly (talk) 18:56, 20 March 2015 (UTC)
The Signpost: 18 March 2015
- From the editor: A salute to Pine
- Featured content: A woman who loved kings
- Traffic report: It's not cricket
.
query about a notable organization article
Florida Center for Investigative Reporting has just been started - alas it appears to be Coatrack from the Rick Scott article with its current content and sourcing. I find no sign that the organization is generally notable, but the author is the one who filed the ArbCom case against me - so I dare not touch this one. If the organization is notable, I would like to know. Right now its "claim to fame" is its allegation that Florida employees are officially barred for saying "climate change." Including a claim that since Scott does not answer questions about the "ban" that this s somehow "notable" in itself.
I have now been quite successfully harassed from anything remotely connected to politics or BLPs on Wikipedia by the many complaints and noticeboard posts, and even a "proposal" on my UT page that we declare Florida officially a "fringe state" filled with "fringe politicians" with "fringe ideas" which I find a clear case of harassment, alas. In any event, I will not touch this "organization" but think the views of others on its inherent notability may be worthwhile. Cheers. Collect (talk) 12:23, 21 March 2015 (UTC)
- What? Making an entire state subject to WP:FRINGE? I mean, Florida might be an epicentre of bullshit, but WTF? Guy (Help!) 23:57, 21 March 2015 (UTC)
- The context was, to my reading I was part of the Wolin discussion, a hyperbolic comment about the use of analogies in a discussion about whether Sheldon Wolin and his concept of Inverted Totalitarianism were FRINGE. Here is a diff [6] of the exchange between Viriditas and Collect leading up to that comment. Jbh (talk) 00:38, 22 March 2015 (UTC)
- Exactly. It was an analogous joke based on Collect's use of similar leading questions in subject headings. As that lengthy, multi-thread discussion shows, Collect does not do well with analogies and metaphors, as he interprets them quite literally. This is one reason (of many), that I find it impossible to communicate with him. And notice, even after it was explained to him that this was a joke on three separate occasions, he is still insisting that it is a real and serious proposal. Clearly, there is more than just a simple reading comprehension problem at work here. He's been repeatedly informed it was a joke yet says above that this is a serious proposal. This is about as WP:IDHT as one can get. Viriditas (talk) 01:02, 22 March 2015 (UTC)
- Thanks to the link leading to the complaint about your harassment of Collect, which was ongoing even after he asked you to stop, which you were doing while he was blocked and now you accuse him of reading comprehension issues, a personal attack.--MONGO 01:28, 22 March 2015 (UTC)
- The context was, to my reading I was part of the Wolin discussion, a hyperbolic comment about the use of analogies in a discussion about whether Sheldon Wolin and his concept of Inverted Totalitarianism were FRINGE. Here is a diff [6] of the exchange between Viriditas and Collect leading up to that comment. Jbh (talk) 00:38, 22 March 2015 (UTC)
How to deal with half truths told in the Western media?
As I pointed out here, the mere fact that in the West we have a free press does not always lead to objective news reporting. A good example could be the Iraq WMD issue, our last Wikipedia article before we could have direct access to the truth is a textbook example of how our media can get corrupted, affecting our Wiki articles too. It's not by telling outright lies, rather by telling only selective truths in a tendentious way. The fact that something is wrong with the conclusions suggested by the news reports can often not be sourced from equally high quality reliable sources, or it's based on a critical analysis by some group that may have some agenda so it will be given less weight (already in the media, long before editors here would have to consider the matter).
Usually this problem will manifest itself in hot buttom issues, like the Iraq's WMD was, or today Russia's involvement in Ukraine or Iran's nuclear program. E.g. the fact that the referendum in Crimea was obviously held in flawed way, doesn't imply that the vast majority of the people there didn't want to join Russia, but in the media we usually only read about how flawed the poll was that was held.
In case of Iran's nuclear program, for all the talk about the number of centrifuges how far Iran is to building an nuclear device and what breakout time we would want the Iranians to have, there is nothing whatsoever one can read about how large Iran's enrichment capacity would need to be in order to provide fuel for the Bushehr reactor, if they were to choose to not renew the contract with Russia for fuel deliveries in the early 2020s. This turns out to be a simple back of the envelope calculation that any ten year old can do, but that simple calculation would all by itself tell a story that is in complete disagreement with the suggested story told by the selective mentioning of the facts we can read about.
Then Wikipedia does have the ability to push back against such pressures to some degree but I think is still leads to significant bias in our articles. E.g. in case of Iran's nuclear program, the core issues are covered, but they appear at the end of the article, the polemics are presented in a more prominent way. This reflects the way our media writes about these issues, which is ultimately due to the fact that the free media has to compete with the entertainment industry for the attention of the public. So, Netanyahu telling a scare story will always win from a professor giving a rather complicated assessment of the facts. Count Iblis (talk) 17:10, 21 March 2015 (UTC)
- And do you have a proposal as to how we deal with this? One that doesn't fall foul of WP:OR and WP:RS policy? I can't think of one offhand... AndyTheGrump (talk) 17:44, 21 March 2015 (UTC)
- That's indeed the problem one faces, you don't want a solution to thisn't issue that would also give cranks the opportunity to get in their version of "the truth" as well. We could perhaps change NPOV to SPOV, i.e. giving far more weight to scientific assessments of an issue. In case of the Iran issue, that would mean giving far more weight to what legal experts, nuclear experts etc. have published in papers on this issue. Count Iblis (talk) 18:05, 21 March 2015 (UTC)
- Strongly agree. I absolutely agree with count iblis on this issue. Wikipedia is becoming unreliable because much of even the so-called mainstream media is biased. In fact, it would be an understatement to say that mainstream media is biased. It is extremely biased would be more apt. Have you ever asked yourself why it is that the media parrots everything that comes out of the White House? Take for instance the ISIS stories online. The media would have you believe that only ISIS engages in sectarian murder, but the Iraqi government has done the same thing. Yet, how much coverage does it get when an U.S. ally does it? In order to boost neutrality we should give more leeway to alternative media. I leave it up to you Jimbo or any other knowledgeable wikipedian to alter our guideline pages. Kuwii hinugu yiri (talk) 17:54, 21 March 2015 (UTC)
- And the evidence that 'alternative media' is less biased than the mainstream can be found where? AndyTheGrump (talk) 18:02, 21 March 2015 (UTC)
- Source analysis, particularly for recent/ongoing events, is a pretty specialized skill set. One which usually requires an area expert or a subject matter expert. Because of how Wikipedia is set up almost all of the results of source analysis are considered original research. This is as it should be because there is no way to vet the skill/knowledge of the editors who want to vet the sources.
Wikipedia is a place where we document what the reliable sources say. analysis is not and should not be done here. As these 'current events' fade into the past we the report other's analysis - people whose knowledge and credentials we can verify. The bias of the news/media is the bias of the encyclopedia because our job is to document not analyze. Jbh (talk) 18:30, 21 March 2015 (UTC)
- I agree that source analysis is a specialized skill; it is, however, one which we must do. It is not our function to regurgitate errors made by the mainstream media echo chamber, it is to determine what is objectively true and to present that information in a fair and temperate manner. Unfortunately, controversial topics are frequently populated by content warriors attempting to spin reality to advance political ends, and therein lies the rub. But the abandoned doctrine of "verifiability-not-truth" is best left in the dustbin. Carrite (talk) 21:47, 21 March 2015 (UTC)
- For an excellent example of what I am saying in practice, see for example the way that "reliable sources" are handled by those tending the medical articles. It's not beneficial to recycle garbage churned out by the hacks of the press; even scientific studies are handled gingerly. Carrite (talk) 21:53, 21 March 2015 (UTC)
- If the standards laid out in WP:MEDRS were replicated elsewhere, I'm sure we'd have a more reliable encyclopaedia - but a much smaller one. And somehow I doubt that the WMF would be particularly keen on that, since 'anyone can edit' seems currently to be more of a priority than 'an encyclopedia'. Ultimately, I think it comes down to the question as to who the project is for - readers, or editors? And lowest-common-denominator standards probably make for higher donations (or do they? does anyone know what percentage of contributions come from Wikipedia editors?) AndyTheGrump (talk) 22:02, 21 March 2015 (UTC)
- I would jump for joy if Wikipedia were limited to high quality academic sources and vetted quality media. If the Daily Mail and similar sensationalistic sources were not used. But who vets the media and what is their POV? If people looked into issues and understood the concept of national narrative, echo chambers and could tell what they are not being told etc. That is just not going to happen because once you try to write rules for source analysis you provide a venue for ever more drama. Proper source analysis goes far beyond NPOV. It means editors must examine and understand their own POV down to such basics as 'my government does not lie' and 'those people I hate might have a point from their perspective'. They would also need to discuss each other's POVs if a frank and open manner and be receptive to having their POV challenged directly by other editors. There would be no simple understandable, rules to fall back on - and most conflicts here seem to be about rules not principles. At best you would end up with SPORTSRS, POLITICSRS, INTERNATIONALRELATIONSRS, WARANDCONFLICTRS etc. each with #CURRENT and #HISTORICAL sections.
Specifying all these 'bestRS' might be possible if the Foundation engaged area experts to identify/specify/quantify the sources for a given area but understanding and applying those rules would place Wikipedia beyond the reach of 'anyone can edit'. That might not be bad, as articles become better they require more specialized knowledge but I do not see it happening. Jbh (talk) 22:47, 21 March 2015 (UTC)
- I would jump for joy if Wikipedia were limited to high quality academic sources and vetted quality media. If the Daily Mail and similar sensationalistic sources were not used. But who vets the media and what is their POV? If people looked into issues and understood the concept of national narrative, echo chambers and could tell what they are not being told etc. That is just not going to happen because once you try to write rules for source analysis you provide a venue for ever more drama. Proper source analysis goes far beyond NPOV. It means editors must examine and understand their own POV down to such basics as 'my government does not lie' and 'those people I hate might have a point from their perspective'. They would also need to discuss each other's POVs if a frank and open manner and be receptive to having their POV challenged directly by other editors. There would be no simple understandable, rules to fall back on - and most conflicts here seem to be about rules not principles. At best you would end up with SPORTSRS, POLITICSRS, INTERNATIONALRELATIONSRS, WARANDCONFLICTRS etc. each with #CURRENT and #HISTORICAL sections.
- If the standards laid out in WP:MEDRS were replicated elsewhere, I'm sure we'd have a more reliable encyclopaedia - but a much smaller one. And somehow I doubt that the WMF would be particularly keen on that, since 'anyone can edit' seems currently to be more of a priority than 'an encyclopedia'. Ultimately, I think it comes down to the question as to who the project is for - readers, or editors? And lowest-common-denominator standards probably make for higher donations (or do they? does anyone know what percentage of contributions come from Wikipedia editors?) AndyTheGrump (talk) 22:02, 21 March 2015 (UTC)
- See The Infography.—Wavelength (talk) 23:20, 21 March 2015 (UTC)
- I'm all for citing alternative media - there's nothing in WP:RS that says we can't, provided they have an editorial process. I realize de facto that PressTV or RT may get harsher treatment than a comparable British outlet, but that's not written into policy. Even within 'mainstream' sources, almost the first hit I got for bushehr centrifuges is [7], which seems to make this argument, but in reverse; they say explicitly that Iran's potential needs in these cases are to be ignored. It's not at all clear to me that you can't get the information across that you want to, under existing policy, without back-of-the-envelopes. Wnt (talk) 23:16, 21 March 2015 (UTC)
(edit conflict)Regarding the comment, "Because of how Wikipedia is set up almost all of the results of source analysis are considered original research." — That's only true if the analysis goes into the article. Original research is acceptable if it's presented on the talk page with the purpose of deciding what from reliable sources should go into the article. At the end of the lead paragraph of the policy No original research is the following,
- "(This policy of no original research does not apply to talk pages.)"
Also, from the section Verifiability does not guarantee inclusion of the Verifiability policy,
- "While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article. The onus to achieve consensus for inclusion is on those seeking to include disputed content."
So if an editor thinks there is a problem with incorrect or misleading information in a source, it doesn't have to be put in the article, unless a consensus of editors decide that they want to put it in. This is the way editing should work, but for one reason or another it may not work that way in Wikipedia, or it may work that way. It depends on the editing environment of a particular article. --Bob K31416 (talk) 23:27, 21 March 2015 (UTC)
- "though it cannot hope to be useful or informative on all matters, it does make the reassuring claim that where it is inaccurate, it is at least definitively inaccurate. In cases of major discrepancy it was always reality that's got it wrong". This has always been the case with Wikipedia, pretty much by design. Guy (Help!) 23:54, 21 March 2015 (UTC)
- and that's from the Hitchhiker's Guide to the Galaxy, don't cha know. --Bob K31416 (talk) 00:37, 22 March 2015 (UTC)
- The last time we trusted the mainstream media we got the Iraq war based on a now disproven theory of WMD's. Now the warhawks want to meddle us into another conflict this time with the 17th largest country in the world. We can start by creating a Wikipedia:List of fearmongering sensationalists in order to discredit these BS warhawks. I'm getting sick of these endless wars. Kuwii hinugu yiri (talk) 00:53, 22 March 2015 (UTC)
Accounts and watchlists
I have six questions (including one double question) about accounts and watchlists.
- (1) How many accounts have been registered on English Wikipedia?
- (2) How many accounts on English Wikipedia have no edits?
- (3) How many editless accounts on English Wikipedia have watchlists?
- (4) How many editless accounts on English Wikipedia have been accessed after they were registered?
- (5) How many watchlists of editless accounts on English Wikipedia have been accessed?
- (6) How frequently have watchlists of editless accounts on English Wikipedia been accessed (individually and collectively)?
—Wavelength (talk) 23:34, 21 March 2015 (UTC) and 23:41, 21 March 2015 (UTC)