Hugo Hacker News

Nobel scientist's deletion from Wikipedia points to wider bias

tptacek 2021-08-19 16:52:06 +0000 UTC [ - ]

What happened here was atypical for Wikipedia "deletionism" debates.

Normally, when you read a story like this, someone created an article and someone else petitioned to have it deleted (with an "AfD") and there's a whole debate log you can read, and you could try to assess people's biases from the comments they left in the AfD.

Here, though, no article was ever created (prior to 2018, when Strickland was awarded her Nobel). Instead, a beginner editor wrote a draft of an article through the "Articles for Creation" ("AfC") process. AfC drafts are, apparently, reviewed by admins before being promoted to the "mainspace" of real articles. You can read the draft here:

https://en.wikipedia.org/w/index.php?oldid=842614385

This is a stub article --- merely the beginning of a real article. And that's fine; it's how lots of good articles start out. But it's easy to see how an admin churning through AfC stub drafts might not have clearly understood the notability of the subject. The article links mostly to the professor's own research group. There are hundreds of thousands of academics whose position in the encyclopedia could also be established with links to their research groups!

What I think is probably more important here though is that the original article, the one we talk about as being "deleted", was created with a special process where the author asked for permission to have it added. That's not how Wikipedia ordinarily works; you have to opt in to the permissions process. The author could just as easily have simply created a Strickland article in the mainspace, and the process that deleted it would have been more formal, would have included appeals, and there'd be a log of the reasoning. It likely would have survived. The author of the draft essentially asked for that not to happen, by using the AfC process.

It seems important to add here: had it been an actual article, and had it been deleted through a speedy deletion or a full-blown AfD process, it would be more difficult to get a Strickland article onto WP in the future (not "difficult", just "more difficult"). But there wasn't, so, in the entire span of time prior to Strickland's Nobel, anybody could have created a stub article for her and her work. From what I can tell, nobody did. That, to me, is the more telling phenomenon here, not the draft deletion.

crazygringo 2021-08-19 17:05:29 +0000 UTC [ - ]

> There are hundreds of thousands of academics whose position in the encyclopedia could also be established with links to their research groups!

Yup, that's really the crux of it. Wikipedia isn't meant for every tenured professor to have a page, the same way an actor with only one IMDB credit doesn't get a page either. And sadly that stub article doesn't seem to clearly explain why this particular professor does.

In hindsight "Nobel prize laureate" is obviously notable... but it's the job of someone creating a page to make the notability rationale clear.

rob_c 2021-08-19 17:04:34 +0000 UTC [ - ]

Not sure I actually agree.

If articles such as this get stricken off and she hadn't won a novel prize someone would just cite "deleted as before" when someone attempts to start a new article again.

AHappyCamper 2021-08-19 17:08:24 +0000 UTC [ - ]

What I don't like about this article (besides the very obvious conclusions that it starts out with and then hits you over the head with) is that it doesn't attempt to reach any wikipedia mod or reviewer for a comment. It's only the author's side and the author's narrative. Also, correlation does not imply causation. The fact that women's pages get deleted more often doesn't mean that there is a gender bias against women.

JohnWhigham 2021-08-19 16:53:32 +0000 UTC [ - ]

Some critics say it was gender bias, while others say it was a problem with notability, a gauge editors use to determine if a topic deserves a Wiki page. Wikipedia editors must be able to verify facts about any Wiki entry against published reliable sources, from publications to the press.

Actual crux of the issue. Stop trying to turn every fucking thing into a gender war.

Grollicus 2021-08-19 16:42:04 +0000 UTC [ - ]

Why does this article about notability deletions not link the notability criteria of wikipedia?

https://en.wikipedia.org/wiki/Wikipedia:Notability_(people)#...

jessaustin 2021-08-19 16:55:08 +0000 UTC [ - ]

"Notability" is an idiosyncratic, illogical, fringe, inconsistently-enforced criterion, and no one cares about it other than a tyrannical minority of wiki editors.

2021-08-19 16:53:42 +0000 UTC [ - ]

rob_c 2021-08-19 16:44:00 +0000 UTC [ - ]

I'd say it's a wiki go edit it but I refuse to given my contributions have been kicked in the past.

krisoft 2021-08-19 16:52:56 +0000 UTC [ - ]

Are you sure you are talking about the same thing? The article is not on a wiki.

fsckboy 2021-08-19 16:35:59 +0000 UTC [ - ]

misleading title, the scientist Donna Strickland had had her page deleted before she won the Nobel prize and therefore did not have a page at the time she won the prize.

jessaustin 2021-08-19 16:48:14 +0000 UTC [ - ]

The current page was created in 2014. [edit: Although it looks like everything before 2018 got deleted in 2018? I don't understand all the notation on the history page. [0]] This is after she served in various roles (including president) at The Optical Society, and of course after developing chirped pulse amplification, for which she won the Nobel, during the 1980s.

[0] https://en.wikipedia.org/w/index.php?title=Donna_Strickland&...

tptacek 2021-08-19 16:55:19 +0000 UTC [ - ]

Are we looking at different things? I just wrote a whole long comment on this after poking around on WP about it, and I'd be glad to know if I got things completely scrambled. Because what I see is a page created in 2018, and a _lot_ of intra-WP coverage of the process that led to there not being a page before 2018, all of which squares with what this article says.

Ensorceled 2021-08-19 16:41:30 +0000 UTC [ - ]

No, that's the point, a woman doing such great work that she won the Nobel Prize had her profile deleted as "not notable".

BeetleB 2021-08-19 17:01:11 +0000 UTC [ - ]

People are assuming quite a bit.

If an article about a professor is written and has virtually no references, then it's not the admin's job to do research to see if she is notable or not. It should be tagged as "needs references" and if they are not provided, eventually deleted.

In her particular case, the article didn't even exist - only a draft did - which was not approved because it had no references. The rejection notice specifically suggests resubmitting with references.

The story/study about bias is interesting, but they picked a poor example to showcase.

moonchrome 2021-08-19 17:03:05 +0000 UTC [ - ]

I don't know the specifics of what makes someone notable in Wikipedia standards - but Wikipedia is obviously not the Nobel committee - sounds perfectly reasonable to me.

rob_c 2021-08-19 16:44:44 +0000 UTC [ - ]

Thank you for saying it.

youeseh 2021-08-19 16:41:09 +0000 UTC [ - ]

Easy enough to fix: "Nobel scientist's deletion from Wikipedia before rise to prominence points to wider bias"

I wish news outlets would stick to reporting the news.

Ensorceled 2021-08-19 16:47:03 +0000 UTC [ - ]

You are missing the point of the article. This woman WAS prominent AND doing notable work AND was still deleted. The Nobel Prize, for the most part, recognizes notable work by prominent scientists.

MichaelGroves 2021-08-19 17:03:20 +0000 UTC [ - ]

Right, the Nobel Prize didn't make her notable, it recognized her existing notability. What made her notable was the work she did before she got the prize. Wikipedian editors incorrectly failed to recognize her notability, because their byzantine bureaucracy is riddled with blind-spots.

tptacek 2021-08-19 16:57:58 +0000 UTC [ - ]

Repeating myself from upthread here, but it doesn't look like this person was deleted from Wikipedia at all. A newcomer asked if there should be a Strickland article, and a volunteer on the project said "based on the article you came up with, it doesn't look like there should be". The newcomer didn't have to ask; they could have (and probably should have) just posted the article directly, which is the ordinary process.

jhgb 2021-08-19 16:55:04 +0000 UTC [ - ]

Prominent to optics experts before the award, clearly, not to Wikipedia editors. It's easy to tell who's prominent in distant past (it's the people in history textbooks), but if your guide to recent prominence is things like lists of awards or frequent mentions in the media, you may be getting some corner cases from time to time.

not2b 2021-08-19 17:06:00 +0000 UTC [ - ]

Wikipedia editors need to fix that. The notability criteria are skewed, tending to think that every minor person in the pop culture is notable, but regularly dumping scientists and local elected officials (the latter is a problem when they run for higher office and people can't find information about them).

They could relax the standard for deletion and it wouldn't hurt anything. If some editors think someone is prominent and others don't, leave the article in place, even if the "not prominent" side has a majority. If the article doesn't meet standards, tag it as a stub or with the "multiple issues" tag.

cool_dude85 2021-08-19 17:02:34 +0000 UTC [ - ]

Well, when you're wrong, and especially when you're exceptionally wrong as in this "corner case", it makes a lot of sense to try to understand what kind of process failure and institutional biases may have played a part in the bad decision. That's what the article's about.

jhgb 2021-08-19 17:06:36 +0000 UTC [ - ]

One of the problems with binary classifiers is that you can't simultaneously make low errors in both directions. That's not "an institutional failure", that's practically a law of nature. You have to live with the fact that sometimes you're wrong. What errors to make is the core of the feud between deletionists and inclusionists.

cool_dude85 2021-08-19 16:43:17 +0000 UTC [ - ]

Before rise to prominence? Presumably she did not publish the work that won her the Nobel and win it on the same day.

Surely publishing Nobel-level work in your field means you have already "risen to prominence."

jhgb 2021-08-19 16:59:35 +0000 UTC [ - ]

How do you tell before such an award is granted whether someone's work is Nobel-worthy or not? There's plenty of already-published works right now that will earn dozens of Nobels to some people in the next decade -- care to tell us which works we should be looking for?

cool_dude85 2021-08-19 17:06:00 +0000 UTC [ - ]

Luckily for me I'm not an admin on wikipedia, so I don't have to decide whose work is notable enough to justify a page. That's their problem.

And it's fine for them to get heat when they make bad decisions, or when people notice that those decisions are systematically biased. I don't have to solve the problem to say it's bad.

rob_c 2021-08-19 16:46:11 +0000 UTC [ - ]

Modern research doesn't consist of magical researchers discovering new things in isolation.

This is not some marvel movie.

If someone has won the Nobel prize they've been changing and contributing in a huge way to their field of interest for many years.

cf100clunk 2021-08-19 16:45:36 +0000 UTC [ - ]

Or better yet: "You'll read this if we say 'Nobel scientist' instead of just 'Academic', right?"

Anyway, as an admitted pedant, even I didn't make a big deal of the title.

codezero 2021-08-19 16:40:30 +0000 UTC [ - ]

This is covered in the article. Also, most Nobel winners are notable for their work long before winning the actual prize.

ilamont 2021-08-19 16:39:33 +0000 UTC [ - ]

pretty soon scientists are going to have to start doing what celebrities and corporations have been doing for years: paid swarms of contributors and PR teams who write articles that are too robust to be deleted, and negative facts/citations are kept to a minimum.

leephillips 2021-08-19 16:58:45 +0000 UTC [ - ]

Misogyny is just one of Wikipedia’s numerous faults. But why pay attention? To me this has about the same import as someone somberly pointing out that an informative graffito was erased from the bathroom wall of their favorite restaurant.

rob_c 2021-08-19 17:02:04 +0000 UTC [ - ]

I feel sorry for your bathroom wall being covered in factual information covering a variety of topics. Then again you never have to buy a newspaper...

I like comparing Wikipedia to graffiti as it's heading in this direction at times but I'd think of it more like a notice board outside a town hall covered in signs for the next witch burning rather than a bathroom stall.

andyxor 2021-08-19 16:36:27 +0000 UTC [ - ]

most of wikipedia content is in public domain, how hard is it to "fork" wikipedia and start a new version without the army of their power editors and biased "admins", i.e. take an initial snapshot of the content as a seed but get rid of the centralized bureaucracy and censorship.

Edit: looks like there is IPFS version https://blog.ipfs.io/24-uncensorable-wikipedia/

kwhitefoot 2021-08-19 16:55:10 +0000 UTC [ - ]

It's easy to fork it but next to impossible to maintain.

jimbob45 2021-08-19 16:39:34 +0000 UTC [ - ]

Sure but those power editors and admins do a lot of heavy lifting that you're taking for granted. Someone's gotta be updating those articles and who's to say the editors you get for your fork will be any less partial than the ones on Wikipedia?

rob_c 2021-08-19 16:56:42 +0000 UTC [ - ]

Could some fork drop the topical, media and "hot" topics and polish the hard sciences?

Or historical facts which should be presented as being unbiased and isn't something which is now changing (unless we find an ancient dvd player in the pyramids).

If this existed and Wikipedia worked with this "factopedia" to pull from it I'd very happily contribute.

Potentially a wiki with some basic checks for having at least an undergraduate in the field of question?

andyxor 2021-08-19 16:49:07 +0000 UTC [ - ]

i think it's better to have multiple versions of wikipedia articles

the distribution would be via something like a search portal on IPFS with ranking of different article versions to appear in search results similar to google Pagerank, based on popularity, or create an explicit mechanism of social voting similar to Reddit.

yes it might also be biased but at least there will be no single source of "truth" and people can make their own mind which version to believe, and be able to do something like "sort by controversial" as they do on Reddit so see different opinions

bena 2021-08-19 16:39:25 +0000 UTC [ - ]

And trade them for an entirely different army of weird and biased editors?

Wikipedia has the same problem every major "community-driven" site does, those with the most time dictate the culture. And since there is no real compensation for that time, no one whose time is worth anything is going to be a major factor.

darepublic 2021-08-19 16:29:49 +0000 UTC [ - ]

Glad I got this entry in during the early days of wikipedia: https://en.wikipedia.org/wiki/Eduardo_Corrochio.

It has been wittled down numerous times over the years but the skeleton of the man's life remains.

sombremesa 2021-08-19 16:35:04 +0000 UTC [ - ]

Based on TFA, this page would've been long gone had this been a woman.

rob_c 2021-08-19 16:42:58 +0000 UTC [ - ]

Typical

Wrote up a long description about how the equipment I used for my master's worked (fleshing out a stub article). Only to have some moderator (being extremely polite here) revert the whole thing because it was "uncited". When challenged I ended up getting temp banned for telling them "I know how it works I just built one". Haven't trusted or contributed since it's moving closer to a collective "what we want to know" information service which makes me want to invest more in an actual encyclopedia.

crazygringo 2021-08-19 16:58:42 +0000 UTC [ - ]

You're being downvoted because everyone knows Wikipedia is built on citations. There's no way for any reader or editor to know whether you know how it works or not.

So Wikipedia is built primarily around not even primary sources but secondary sources for a good reason.

Yes it may lose out in some valuable knowledge such as yours in the process -- but it achieves a greater good of minimizing Wikipedia being flooded by incorrect or even harmful information.

In the absence of an omniscient impartial arbitrer, requiring citations of secondary sources is a pretty good rule.

the_third_wave 2021-08-19 16:52:18 +0000 UTC [ - ]

> it's moving closer to a collective "what we want to know" service

I'd say it is moving towards a "what we want you to think" corpus, with politically motivated activists taking over from objective editors. Those activists will use the language of editors, calling upon the myriad of policies put in place when Wikipedia was growing, while bending and changing that language towards their own goals.

vidarh 2021-08-19 16:56:49 +0000 UTC [ - ]

While I get this is frustrating, original research is explicitly against Wikipedia rules.

tedsanders 2021-08-19 16:52:00 +0000 UTC [ - ]

I know it's frustrating to write true things and have them deleted, but I hope you understand how bad it would be to generalize the rule "accept uncited/unverified claims as true if the account who wrote them says it's true."

I say this as someone who spent weeks writing up a new Wikipedia article with 50 citations and then had it graded a C because I used too many primary sources instead of secondary sources. I empathize with you, but I also empathize with the editors.

rob_c 2021-08-19 16:59:06 +0000 UTC [ - ]

Yes and no.

I'd argue having an unpolished article in need of work is better than a stub as it encourages interaction.

Not to mention being told to stop using a paywall for citing a journal (which is now out of print) as a source.

rob_c 2021-08-19 16:50:54 +0000 UTC [ - ]

At this point I'm willing to go to go toe on this on a typical debate perspective.

If I'm being voted down what is so great about the ignorance of the masses here?

Wikipedia has some serious (and in come cases concerningly anti-science) biases. Which is honestly love for them to address at a community level.

2021-08-19 16:58:10 +0000 UTC [ - ]