Wikipedia’s Sexism Toward Female Novelists

[thekohser]

Hey Greg,
You want me to post your comment for you?

Sure, if you want it to get rejected, and then have your other two comments pulled. Be my guest.

[Volunteer Marek]

Are all subcategories wrong?

Possibly. They are certainly an important arena for edit-warriors, because you is either in the category or you ain't, no NPOV applies. So if someone can get you into a pejorative category (like "Conspiracy theorists"), you are screwed.

While I'm pretty sure there's some background axe-grinding on Hersch's part in the above comment (LaRouche stuff getting put into the Conspiracy theory cat), he is right in general. Back in the day, when the "nationalist" wars were raging on Wikipedia there were actually accounts which specialized in POV-pushing via wacky categorization. It was actually becoming a niche specialization, because you didn't need sources and you wouldn't get reported to drama boards since the edits looked "trivial". So you put "country X" or "person X" in category "Sex crimes" or "Nazis" or "Ugly looking countries or people that killed a bunch of people and hate old ladies and kick their dogs a lot while listening to Justin Bieber" (or whatever inane categories Wikipedians manage to invent) or simply "Bad category Y" and you don't have to justify it in any way because it's not like somebody's gonna put a {{citation needed}} tag after the category. All you need is some flimsy excuse. And it does show up in google searches. Points!

And of course even these kinds of agenda driven categorizations get completely lost in the sea of bots and twinkle users and bot-like-editors trying to run up their edit counts by endlessly categorizing and re-categorizing stuff for no apparent purpose. It's like you have a thousand people digging ditches and then filling them back up just so they can look like they're working (the edit count insanies), so of course the wacko or two (the POV pusher) sneaks in a dead body or three and gets them safely buried along the way.

Personally I try to ignore the whole categorization mess as much as possible unless something really obnoxious pops up on my radar.

(on the same topic - how do you deal with crazy POV articles kept in user space, because if they were in article space they'd get edited/deleted/NPOVed, but they're safe in user space and you still get the google juice?)

[Mason]

I think editors who do things like that should be banned much more quickly and firmly than our usual relaxed approach to banning.--Jimbo Wales (talk) 17:40, 25 April 2013 (UTC)

It is quite remarkable how quick Jimbo was to throw that poor sap under the bus.

[Vigilant]

I think editors who do things like that should be banned much more quickly and firmly than our usual relaxed approach to banning.--Jimbo Wales (talk) 17:40, 25 April 2013 (UTC)

It is quite remarkable how quick Jimbo was to throw that poor sap under the bus.

What I find remarkable is that Jimbo doesn't pay attention forever, things go to shit, he swoops in, makes a knee jerk pronouncement and then flies off into the sunset waiting for the archive bot to sweep his half rendered shit under the rug.

[HRIP7]

Sue Gardner's argument is like saying a broken-down car "worked" because you complained about the manufacturer in the press.

And anyway, Wikimedia cars are never finished. So don't complain about the absence of brakes.

LA Observed have another write-up of the debacle, focusing on revenge editing.

Kevin Roderick, LA Observed, Revenge editing is a big blemish on Wikipedia

Such troll attacks are often caught by the larger Wikipedia community — but not always. For a stark example of the problem, look at this rant by one of the Wikipedia "editors" who took umbrage at Filipacchi's New York Times op-ed about the sexist way of organizing novelists into "novelists" and "women novelists." He froths about Judith Miller and Saddam Hussein and the New York Times to justify his dishonest editing, and basically promises to plague her through the years.

The bloody p.o.s. New York Times supposedly employs fact checkers, but they have allowed this incompetent woman to libel Wikipedia not once, but two times. They owe Wikipedia two separate retractions. They have no journalistic integrity whatsofuckingever. They are nothing better than a blog, a barrel full of dog feces offered to the world as the “truth.” There is one thing you are wrong about, however. This incident is never going to be forgotten. Not by anyone involved in it. Retribution will be taken five, ten, fifteen, twenty years from now. That’s just the way people seem to be, unfortunately. It is the way these things work, and that’s something about the world which many of us actively dislike, and are working hard to change. The documented fact is that this woman has sent thugs after certain Wikipedia editors. This is no slight affair, I am afraid.

The New York Times has a vested interest in trying to undermine Wikipedia. For one thing, the Times has only 600,000 digital subscribers, which makes it a piece-of-shit website in terms of numbers. On Sundays, its biggest day, the Times adds another 1.4 million readers in its paper edition, for a total of 2 million. Meanwhile, HUNDREDS OF MILLIONS OF PEOPLE are reading you and me on Wikipedia EVERY DAY. You can see why the Times feels it has a very very short and stubby and ugly little penis compared with us. This is the real reason why they want to run baseless articles slamming us. Because we are the future and they are already the distant past.

If losers like this are allowed to do anything for Wikipedia, it's a shame.

[Hersch]

Mighty Ming, may I beg an indulgence?
Why are there plainly pejorative categorizations on wikipedia to begin with?

Because people do things that are stupid and/or evil? If one accepts the notion of categorization/tagging/whatever, there are going to be people/things that get identified in ways that reflect badly on them, because they did or represent bad things.

When you get right down to it, this why we really need Wikipedia. The average person is never going to figure out for him- or herself who the stupid and/or evil people are. They rely upon the more sophisticated ones, the mouth-breathing basement-dwellers who really know where the action is happening at, the Wikipedians. Those are the people who are smart enough to know which public figures are the stupid and/or evil ones. and Wikipedia provides them with the bully pulpit from which they may instruct the ignorant masses.

[HRIP7]

And of course even these kinds of agenda driven categorizations get completely lost in the sea of bots and twinkle users and bot-like-editors trying to run up their edit counts by endlessly categorizing and re-categorizing stuff for no apparent purpose. It's like you have a thousand people digging ditches and then filling them back up just so they can look like they're working (the edit count insanies), so of course the wacko or two (the POV pusher) sneaks in a dead body or three and gets them safely buried along the way.

I like the way you put that.

I said much the same thing on Gendergap the other day.

Compare it to the weaknesses of the current category system. 98% of editors
don't know what they are doing. Categories and subcategories are applied
inconsistently all the time. Nobody has an overview of the entire tree
structure, or even a major branch of it. Something that is a subcategory of
American novelists today may stop being one tomorrow, just by dint of a
single edit, and no one would be the wiser (unless they keep hundreds of
categories on their watchlist). The category tree (or weave, as categories
can have several parents) changes daily, with categories created, renamed,
recategorised, and deleted. There are incessant arguments about how to
name, categorise and diffuse categories, and about perceived iniquities.
Wiki-gnomes spend days working and undoing each other's work. It's insane.

[Hex]

What I find remarkable is that Jimbo doesn't pay attention forever, things go to shit, he swoops in, makes a knee jerk pronouncement and then flies off into the sunset waiting for the archive bot to sweep his half rendered shit under the rug.

+1

[lilburne]

What I find remarkable is that Jimbo doesn't pay attention forever, things go to shit, he swoops in, makes a knee jerk pronouncement and then flies off into the sunset waiting for the archive bot to sweep his half rendered shit under the rug.

+1

He simply understands how web2.0 works. Its what all of them do.

[Poetlister]

you is either in the category or you ain't, no NPOV applies. So if someone can get you into a pejorative category (like "Conspiracy theorists"), you are screwed.

Wrong. You can edit war without limit about these things - "I have a reliable source that he's ginger haired" - "It's not reliable" - "Yes it is" "OK, but it doesn't say in so many words that he's ginger, only that he's carrot-haired". Or even the ultimate put down: "It may be true, but it has no relevance to the article so should be removed".

[Hersch]

you is either in the category or you ain't, no NPOV applies. So if someone can get you into a pejorative category (like "Conspiracy theorists"), you are screwed.

Wrong. You can edit war without limit about these things - "I have a reliable source that he's ginger haired" - "It's not reliable" - "Yes it is" "OK, but it doesn't say in so many words that he's ginger, only that he's carrot-haired". Or even the ultimate put down: "It may be true, but it has no relevance to the article so should be removed".

The point, Mr. Outsider, is that there is no compromise solution to disputes about inclusion in a category. There are simply winners or losers.

[Ming]

And as anyone could have guessed, the problem category will remain and will also be repopulated into the parent category.

[Sweet Revenge]

And as anyone could have guessed, the problem category will remain and will also be repopulated into the parent category.

I guess Obiwankenobi (T-C-L) didn't get the news, because he's issuing hidden commands to the contrary on Amanda Filipacchi (T-H-L):

<!-- Note: Since this article is already in two by-century, diffusing sub-categories of American novelists, it should not be placed in American novelists -->

[Zoloft]

And as anyone could have guessed, the problem category will remain and will also be repopulated into the parent category.

I guess Obiwankenobi (T-C-L) didn't get the news, because he's issuing hidden commands to the contrary on Amanda Filipacchi (T-H-L):

<!-- Note: Since this article is already in two by-century, diffusing sub-categories of American novelists, it should not be placed in American novelists -->

You're solving the wrong problem!

[Ming]

I'm voting for death wish.

[Sweet Revenge]

The dinosaurs are bickering about the deck chairs while the meteor descends upon the titanic:

No, the decision is that we will not remove articles from [[:Category:American novelists]] to place them in [[:Category:American women novelists]]. There is no decision against moving them to [[:Category:20th-century American novelists]] or [[:Category:American historical novelists]]. Those are legitmate diffusing categories and there is absolutely no reason not to diffuse. If you think we should not have [[:Category:20th-century American novelists]], you are free to take it to a CfD, but while it exists it is a 100% legitimate diffusion category, and there is no reason to fight diffusion.[[User:Johnpacklambert|John Pack Lambert]] ([[User talk:Johnpacklambert|talk]]) 03:16, 3 May 2013 (UTC)

(Why don't we have a smiley for intentionally mixed metaphors??)

[Zoloft]

The dinosaurs are bickering about the deck chairs while the meteor descends upon the titanic:

No, the decision is that we will not remove articles from [[:Category:American novelists]] to place them in [[:Category:American women novelists]]. There is no decision against moving them to [[:Category:20th-century American novelists]] or [[:Category:American historical novelists]]. Those are legitmate diffusing categories and there is absolutely no reason not to diffuse. If you think we should not have [[:Category:20th-century American novelists]], you are free to take it to a CfD, but while it exists it is a 100% legitimate diffusion category, and there is no reason to fight diffusion.[[User:Johnpacklambert|John Pack Lambert]] ([[User talk:Johnpacklambert|talk]]) 03:16, 3 May 2013 (UTC)

(Why don't we have a smiley for intentionally mixed metaphors??)

[Hex]

The dinosaurs are bickering about the deck chairs while the meteor descends upon the titanic...

(Why don't we have a smiley for intentionally mixed metaphors??)

How about:



Or indeed for any time you realize you may well be making no sense at all.

[Cedric]

The dinosaurs are bickering about the deck chairs while the meteor descends upon the titanic...

(Why don't we have a smiley for intentionally mixed metaphors??)

How about:



Or indeed for any time you realize you may well be making no sense at all.

I got it.

[HRIP7]

For those interested, Magnus Manske explains a little bit about the technical challenges with category intersections here.

[lilburne]

I believe that there are well traversed techniques to optimize SQL queries on a database.

[Vigilant]

I believe that there are well traversed techniques to optimize SQL queries on a database.

It's a shame they're trying to recreate, with terrible programming talent, a flat, text-based database search index.

[Ming]

For those interested, Magnus Manske explains a little bit about the technical challenges with category intersections here.

Oh, good God. This is classic "anyone could guess that this is problem people have tried really hard to solve, but rather than look for a solution I will display my own technical 'competence' by blathering on about how hard it is" marginal competency speaking. Hmmm, let's see:


Take a basic relational database with a really simple table of two columns: an index to the article, and a category tag. You index both tables. OK, so now you want to apply a set of four category tags to get an intersection, so you write a four-way inner join of the table to itself. I think even MySql's optimizer is smart enough to look at the tag index and start with the tag that returns the smallest row set and then follow the join across to get a set of rows to reduce by checking each against the other criteria. Or it might get a list for each of the tags and then collate them by article index. No, this isn't a hard problem.

[HRIP7]

For those interested, Magnus Manske explains a little bit about the technical challenges with category intersections here.

Oh, good God. This is classic "anyone could guess that this is problem people have tried really hard to solve, but rather than look for a solution I will display my own technical 'competence' by blathering on about how hard it is" marginal competency speaking. Hmmm, let's see:


Take a basic relational database with a really simple table of two columns: an index to the article, and a category tag. You index both tables. OK, so now you want to apply a set of four category tags to get an intersection, so you write a four-way inner join of the table to itself. I think even MySql's optimizer is smart enough to look at the tag index and start with the tag that returns the smallest row set and then follow the join across to get a set of rows to reduce by checking each against the other criteria. Or it might get a list for each of the tags and then collate them by article index. No, this isn't a hard problem.

That assumes that everyone using the search system uses it because they are interested in the result. Looking at it from Wikimedia's side, wouldn't they be concerned that if they had this system, and someone were to write a script that performs several thousand searches for intersections of top level categories ("man", "living person", "American" etc.) each day, they could bring the server down?

[lilburne]

For those interested, Magnus Manske explains a little bit about the technical challenges with category intersections here.

Oh, good God. This is classic "anyone could guess that this is problem people have tried really hard to solve, but rather than look for a solution I will display my own technical 'competence' by blathering on about how hard it is" marginal competency speaking. Hmmm, let's see:


Take a basic relational database with a really simple table of two columns: an index to the article, and a category tag. You index both tables. OK, so now you want to apply a set of four category tags to get an intersection, so you write a four-way inner join of the table to itself. I think even MySql's optimizer is smart enough to look at the tag index and start with the tag that returns the smallest row set and then follow the join across to get a set of rows to reduce by checking each against the other criteria. Or it might get a list for each of the tags and then collate them by article index. No, this isn't a hard problem.

That assumes that everyone using the search system uses it because they are interested in the result. Looking at it from Wikimedia's side, wouldn't they be concerned that if they had this system, and someone were to write a script that performs several thousand searches for intersections of top level categories ("man", "living person", "American" etc.) each day, they could bring the server down?

You cache the results of popular queries. Most large sites update caches during low access periods, or at set periods {every 12 hrs, once a day, once a week} depending on volatility of the dataset.

[Anroth]

What he said ^^ This really is well-trodden ground technically. Most DBA's could arrive at an elegant solution.

Most sysadmins could hack it together ugly

[TungstenCarbide]

For those interested, Magnus Manske explains a little bit about the technical challenges with category intersections here.

Oh, good God. This is classic "anyone could guess that this is problem people have tried really hard to solve, but rather than look for a solution I will display my own technical 'competence' by blathering on about how hard it is" marginal competency speaking. Hmmm, let's see:


Take a basic relational database with a really simple table of two columns: an index to the article, and a category tag. You index both tables. OK, so now you want to apply a set of four category tags to get an intersection, so you write a four-way inner join of the table to itself. I think even MySql's optimizer is smart enough to look at the tag index and start with the tag that returns the smallest row set and then follow the join across to get a set of rows to reduce by checking each against the other criteria. Or it might get a list for each of the tags and then collate them by article index. No, this isn't a hard problem.

That assumes that everyone using the search system uses it because they are interested in the result. Looking at it from Wikimedia's side, wouldn't they be concerned that if they had this system, and someone were to write a script that performs several thousand searches for intersections of top level categories ("man", "living person", "American" etc.) each day, they could bring the server down?

You cache the results of popular queries. Most large sites update caches during low access periods, or at set periods {every 12 hrs, once a day, once a week} depending on volatility of the dataset.

and set aside a couple of db computers for this work.

[Vigilant]

For those interested, Magnus Manske explains a little bit about the technical challenges with category intersections here.

Oh, good God. This is classic "anyone could guess that this is problem people have tried really hard to solve, but rather than look for a solution I will display my own technical 'competence' by blathering on about how hard it is" marginal competency speaking. Hmmm, let's see:


Take a basic relational database with a really simple table of two columns: an index to the article, and a category tag. You index both tables. OK, so now you want to apply a set of four category tags to get an intersection, so you write a four-way inner join of the table to itself. I think even MySql's optimizer is smart enough to look at the tag index and start with the tag that returns the smallest row set and then follow the join across to get a set of rows to reduce by checking each against the other criteria. Or it might get a list for each of the tags and then collate them by article index. No, this isn't a hard problem.

That assumes that everyone using the search system uses it because they are interested in the result. Looking at it from Wikimedia's side, wouldn't they be concerned that if they had this system, and someone were to write a script that performs several thousand searches for intersections of top level categories ("man", "living person", "American" etc.) each day, they could bring the server down?

You cache the results of popular queries. Most large sites update caches during low access periods, or at set periods {every 12 hrs, once a day, once a week} depending on volatility of the dataset.

and set aside a couple of db computers for this work.

We're not going to work through RDB optimization.
Let's just say that many companies have studied this problem extensively as part of their primary product development efforts and none of htem ahve chosen to use a flat, text based representation of data.

[Hex]

Bonus points to Ming for excellent first use of the Matrix smiley.

We're not going to work through RDB optimization.
Let's just say that many companies have studied this problem extensively as part of their primary product development efforts and none of htem ahve chosen to use a flat, text based representation of data.

Graph databases (T-H-L) are appropriate for this kind of thing, no?* Which would require a lot of rewiring of MediaWiki innards. Which should happen, in that case.


* I was never much of a mathematician/computer scientist (ie graph theory (T-H-L)) so the answer may well in fact be "no". But I'm happy to be set straight.

[EricBarbour]

This is classic "anyone could guess that this is problem people have tried really hard to solve, but rather than look for a solution I will display my own technical 'competence' by blathering on about how hard it is" marginal competency speaking.

I'm not sure if it's a question of competency, or if Mr. Manske realizes what a political morass it would be to get
his Magical Fellow Wikipedians to agree on a set of rules for categories. So perhaps he doesn't want to address it.

Like the Manual of Style, they should have decided on categorization before writing articles. But no, they
started generating a mountain of content first, in a burst of Bitcoin-like blind enthusiasm, and years later started
to argue over article format constantly--and they are still arguing. Just another little thing Jimbo facilitated.

[HRIP7]

Magnus has done some more work on intersections. Whatever he's done, it now seems a couple of orders of magnitude faster.

[EricBarbour]

Magnus has done some more work on intersections. Whatever he's done, it now seems a couple of orders of magnitude faster.

Just what he said: "Recurses full depth; avoids loops". In other words, the previous version of CatScan had problems handling
circular categories and recursion. It was buggy.