Wikipedia, as a collaborative venture, reflects certain predilections and biases among its editors. For instance, you will find very active communities making sure that every notable Pokémon character gets their Wikipedia article. Other groups are less well represented. Somewhat more than a year ago, the physicist Jess Wade, currently a postdoctoral researcher at Imperial College, decided to systematically improve the representation on Wikipedia of women scientists in particular. (She has since expanded this to include other underrepresented minoritiers.) Wade set herself the goal of adding one biographical article per day to Wikipedia – and she has written more than 500 such biographies since then.

Some of those biographies have since been deleted as “not sufficiently notable,” and quite predictably, this has drawn criticism, and claims of bias on the part of Wikipedia editors. This prompted me to write the following How-To, originally posted as a loooong thread on Twitter, and now reproduced here (unusually for this blog, in English). Here goes.

If you're angry about the recent deletion of articles by @jesswade from #Wikipedia, there is a way to help. Of course, that way is also open to you if you are not angry about those deletions: Help to edit Wikipedia. 1/n

— Markus Pössel (@mpoessel) 3. Mai 2019

If you’re angry about the recent deletion of articles by @jesswade from #Wikipedia, there is a way to help. Of course, that way is also open to you if you are not angry about those deletions: Help to edit Wikipedia. After all, Wikipedia is “the free encyclopedia that anyone can edit”!

Only that it’s not that simple, of course. It’s easy to become enthusiastic about contributing to Wikipedia, start editing, have some really frustrating encounters with other members of the community, and then leave because who needs that kind of aggravation when they are just trying to volunteer for something?

This thread is about how to do good things on Wikipedia and avoid at least some of the frustration-inducing pitfalls.

Wikipedia is a social system

Wikipedia is a social medium, and involves a considerable number of people. It pays to understand at least some of the social dynamics. A key part of the dynamics is that, when you take the world as a whole, many people are jerks. In addition to the jerks, there are many who are thoughtless. Take these facts together with the facts that (a) Wikipedia is widely known and (b) everyone can make changes to Wikipedia at the touch of a button, you can imagine that in every minute, there is some person somewhere in the world who thinks it’s funny to just delete the page of Justin Bieber, or replace all adjectives in the article about a historical figure with assorted four-letter words, or if they are really evil, to add factors 2 and minus signs to physics or mathematics articles.

If this vandalism (as it is called on Wikipedia) were to go unchecked, the site would drift into unusability on a comparatively short time scale. You might say “Hey, that’s just the second law of thermodynamics!” but you would still be left without the currently simplest way of looking up reliably what the second law of thermodynamics actually says. Wikipedia has evolved ways to react to vandalism. Some are automated; after all, detecting suspect peaks in the frequency of four-letter-words is something a computer can do. Others are not automated; there are Wikipedia editors who have made it a substantial part of their life’s work to suss out and eliminate the more complex forms of vandalism. Those editors are routinely dealing with the nastiest attempts to subvert Wikipedia, which can’t be fun, but they do it anyway, and anyone who uses Wikipedia should be grateful to them.

Another part of the reaction to vandalism has led to an infrastructure of rules (technically “policies” and “guidelines”, but I’ll stick with the general word “rules”). Rules about what is notable. Rules about how to ensure that information in an article is reliable. (One such rule, for instance, is “No original research!” – Wikipedia is not for documenting your personal view of things; you are supposed to document what is already published in reliable references!) These rules have evolved as an approximation to consensus within the Wikipedia community. (Behind the scenes at Wikipedia, there are lots of discussions going on trying to establish consensuses [consensi?] on various matters.)

Unfortunately, Wikipedia is not that different from other kinds of societies. Even if the rules are basically good, they can also be abused. And for someone who is new to Wikipedia, coming in contact with the rules can be quite daunting. (There are rules for that too, namely “Please do not bite the newcomers!”, but in certain situations, that rule is unlikely to save the day.

How to get started

My first piece of advice for starting to edit Wikipedia would be: start in a way that is unlikely to lead to conflict. If you are a scientist, you might want to start with existing articles within your field of specialization. Even in, say, physics or chemistry, there are numerous “stubs”, minimal articles that really need more information. For a systematic way to find them, you might want to head to the appropriate WikiProject (that is, loose group of editors interested in the same topic). Here is the physics project page. Did you notice that the page name begins with “Wikipedia:”? That is because this is a “behind the scenes” page. Those are the pages where all the behind-the-scenes discussions take place, where you can find the various rules written down, and the project pages.

Not in physics? See if you can find a suitable WikiProject here in this Directory of WikiProjects. I’ll stay with the physics page as an example, though, which has a section Pages needing attention. Click on “stubs”, for instance, and you will jump to an automatically generated page of Wikipedia physics articles that could use care and feeding. Now click on a stub and start improving the article!

What if someone changes it back?

One pitfall, though. In the way we have set this up, you are editing the article because it falls into your area of expertise. So if you discover a mistake and correct it, and if some feckless other editor than changes it back, what then?

First: It would be a mistake to make your correction once more, then the other editor will change it back, and so forth. That would be what is called an “edit war”, which is against the rules (for obvious reasons). The right thing would be to go to the article’s talk page and to start a discussion.

In the image below, do you see the little tab on top where it says “Talk”? That brings you behind the scenes of that particular article, in my example: Alexei L. Efros (whoever he is, but then, that is what we can read up on in this article, right?).

Start a new discussion section on that talk page and communicate with the other editor. One pitfall, though: Your argument shouldn’t be “I am a world expert on this, have published 10 papers on this topic alone, and I’m telling you my correction is right!” – that might be true, but it is not relevant for Wikipedia.

You can probably imagine why: This just moves the problem of how to verify your information to the next level. Anyone can claim to be an expert; from having to verify whether your correction is valid, the problem has moved to verifying you are who/what you say.

You could say that it would be silly not to make use of the fact that some contributors are experts. But Wikipedia consensus has gone the other way: no special status for experts.

Which is not as bad as it sounds, though, for this reason: The standard for deciding whether your correction is right, or your counterparts counter-correction is: what do the reliable references say? Would that we were so conscientious in other areas of life.

That is how these matters are decided, and there your advantage as an expert comes to the fore again. After all, being an expert also means knowing where to find the best references, and knowing what those are. In all but the most frustrating cases (which you hopefully will not encounter), giving a convincing reference (refereed journal, review article…) should resolve the conflict.

“Citations needed”

In fact, you will want to make a habit of adding suitable inline references in the first place. That way, you will (a) be on the safe side and (b) will really contribute to improving Wikipedia. After all, Wikipedia is where people often look first, but Wikipedia by itself is nothing they can cite as an authoritative source. By adding such a reliable source ready for citing, you will make Wikipedia a much more useful tool for students and scientists alike.

Oh, one other thing. It is good form to not only edit Wikipedia without logging yourself in (although that is possible), but to make a user account. Like every other registered user, you can make a page describing yourself. Here is mine, for example: en.wikipedia.org/wiki/User:Mark…

If, as a scientist, you briefly describe your position and research interest, that should also help things running smoothly. Editors will (or should) still not believe you just because you are an expert, but just like elsewhere, most Wikipedia editors are not jerks. And a non-jerk who is no expert in physics, when confronted with a correction by an expert, will not stubbornly quote the “No experts!” rule at you, but instead be smart enough to realize your correction/contribution should not be dismissed lightly. Which shouldn’t keep them from pointing out that if you could put in a reliable reference at that point, that would be really great.

Helping to balance Wikipedia: getting there

Now, so far I have not said anything about Jess’s work and the biography of scientists who are women or come from other minorities underrepresented in a given science, or science in general. There is a reason for that, and it is up to you to decide whether or not it is a good one. If you enter, say, an internal debate on Wikipedia in order to argue why a certain article should not be deleted, without having contributed anything else to Wikipedia, it is easy for those arguing against you to dismiss you.

Such decisions are supposed to reflect a consensus within the Wikipedia community, and if you just barge in from the outside to argue, your voice will not carry much weight. You will appear to be someone who isn’t really part of the community, hasn’t really contributed to the community, yet wants to tell the community what to do. (This doesn’t mean your arguments aren’t valid, but you see my/their point, right?)

So my advice would be: Want to effect positive change in the Wikipedia community? I’m all for that. But please become part of the community first, contribute to the common project, then get involved in the more controversial issues. You will carry more weight that way.

As for what constitutes sufficient community involvement, there is (predictably) no common standard, but there are indications. For instance, your Wikipedia account will allow you to do somewhat more than you initially could (namely edit certain protected pages) if your user account is more than 30 days old and you have made more than 500 edits. So that might be a suitable goal to set for yourself before taking part in controversies.

And if you want to contribute to biography pages of women/underrepresented minorities in science during that time, do so! Get hold of a reference book of the relevant biographies (any reliable source is your friend!) and start adding what you find in there, properly referencing what you write.

After you have found your feet and gained some editing experience, it’s time for the more difficult stuff. Here goes.

Why are e.g. women scientists underrepresented?

Why are women scientists (I will concentrate on women, but the reasoning applies analogously to other underrepresented minorities) underrepresented on Wikipedia? Several reasons, but I think a fairly uncontroversial consensus is as follows:

Wikipedia is not meant to be a place for original research. Like every other article, those on women scientists are supposed to be about notable scientists, and there are supposed to be reliable references. That means: If women are underrepresented in the regular media, portrayed less often, interviewed less often, given fellowships and Nobel prizes less often, that will also make them underrepresented on Wikipedia.

You are unlikely to change Wikipedia’s consensus on that, and convince editors that Wikipedia should be at the vanguard of making women scientists more visible. Not only because of anti-feminist bias (although you will certainly encounter that in some editors), but because you would be asking Wikipedia to give up those core principles that are meant to keep Wikipedia content reliable – no original research, document only the knowledge that is out there in reliable, verifiable references.

Quite a number of Wikipedia editors who are sympathetic to the cause of ending underrepresentation of women and URM in Wikipedia would oppose such a fundamental change to what Wikipedia is, too.

Which brings us to the next bit.

Striking a balance, one biography at a time

Even by Wikipedia’s own standards, lots of women that are notable enough to have a Wikipedia article of their own, do not have one. This is what @jesswade has been doing, although I have no idea how she finds the time for this incredible amount of work, in addition to her research career: adding biographies of scientists who are women or from other underrepresented minoritiers.

I’ve already told you the bare-bones basics about Jess Wade; here is a @guardian article to read on Jess some more:

The articles by Jess follow the usual Wikipedia conventions and are well-sourced with inline references. So you might think that the whole Wikipedia community is thankful that someone is putting all this work into creating nice new articles, right? Wrong, sadly.

Why would anyone be against more women scientist biographies?

Now we come to one of the usual problems when working on issues of equality. None of the Wikipedia editors who are making it more difficult for Jess to do what she does are, of course, going to do this saying “Ugh, another woman scientist, I will delete that, and by the way, yes, I am a mysogynist.”

But then, anyone who really is a mysogynist and wants to hamper Jess’s work has ample opportunity to be more subtle. Recall the rules that Wikipedia set itself for staying reasonably reliable: Only to cover notable subjects, and to require verifiable references.

First, an aside: notability and verifiability are clearly different things. Why not make an encyclopedia that includes everything that is verifiable, regardless of notability? That has been one of the great behind-the-scenes debates for decades. If you want to learn more, look on the behind-the-scenes pages for the keywords “inclusionist” (a Wikipedian who wants to include everything) and “deletionist” (obvious, no?). Want to know more? There happens to be a Wikipedia article on it en.wikipedia.org/wiki/Deletioni…

It is true, of course, that including more, and less notable, subjects, is not a matter of insufficient space (I’ve heard the words “The Great Wikipedia Paper Shortage” used in this context). But it is true that a much larger encyclopedia is that much more difficult to keep up to date, and vandalism-free (at least to a certain extent, since the number of jerks etc. in the world is likely to be independent of the number of Wikipedia articles).

Anyway, current consensus is to apply certain notability criteria. This is where matters become more difficult, and somewhat contentious. And by now, there is a large body of rules about notability. If you want, read the guideline on notability: en.wikipedia.org/wiki/Wikipedia… Lots of details, of course, but the upshot is: If there is sufficient non-trivial coverage of a subject in reliable, independent sources, it is probably notable enough to have its own Wikipedia article.

When is a scientist (or other academic) notable?

When it comes to living scientists’ biographies, that doesn’t help us very much, though. Even most very good scientists do not rise to the level of having a book chapter written about them, or newspaper articles. This is where Wikipedia as a system is biased towards celebrities of all kinds. They’re much more in the news. When it comes to historical articles about women scientists and scientists from other underrepresented minorities, we have a similar problem. Talk about “hidden figures”.

Fortunately, Wikipedia is not as dumb as leaving it at that. In 2005/2006 a more specific guideline for notability of academics began to emerge. Here it is: Notability (Academics), abbreviated (among other abbreviations) to WP:PROF en.wikipedia.org/wiki/Wikipedia…

Basically, this is patterned after what the academic community has in terms of standards when it comes to who is and isn’t notable: prizes, fellowships, named chairs or distinguished professorships (which are straightforward to verify) and then more vague criteria such as their research having “had a significant impact in their scholarly discipline”, and their academic work having “made a significant impact in the area of higher education”. Those last criteria leave one some latitude of interpretation.

Are the criteria themselves biased?

Before I come to why that can become a problem, note that some criteria for notability are missing. Now, I may be wrong about this because Wikipedia’s rules are many, but I have not yet come across a notability guideline that specifically mentions that a person can be notable for being the first in their group to achieve a certain milestone – the first black women to contribute significantly to the discovery a new element for instance.

Naturally, this is particularly relevant for underrepresented minorities, and not having this mentioned explicitly as a criterion for notability makes it harder to prove notability for a number of members of such minorities. Again, I may be wrong, and may be @Wikipedia or @Wikimedia will be so kind as to correct me on this, but I think this might be an instance where Wikipedia is systemically biased against members of underrepresented minorities.

Vagueness allows for bias to influence decisions

Back to the criteria we do have, such as substantial contribution to science or higher education. This is another area where bias can creep in. Now we come specifically to the deleted articles. Wikipedia has a procedure for that. If an article is found to lack content that is relevant for an encyclopedia, it should be deleted. In practice, it is nominated as an “Article for Deletion”, or AfD. (For Germans, that is a particularly weird association. AfD is the far-right, anti-refugee party in Germany.)

Once an article is nominated, there is a community discussion. Usually with explicit reference to the rules (or if you are a cool Wikipedia insider, to their abbreviations, like the aforementioned WP:PROF), editors give their opinion on the matter and, at some point, cast a vote, in this case “Keep” or “Delete” for the article in question.

And this, of course, is where bias can creep in. Now, I cannot say and I am not saying that those editors who have voted to “Delete” a number of Jess Wade’s articles over the last few days are doing so because they are mysogynists, or anti-feminists, or want to stick it to that pesky, uppity women-and-URM-biography-writer Jess. And, skating dangerously close to “Not all Wikipedians!”, I don’t believe all of them are.

But on the other hand, I cannot help but notice that the Article-for-deletion discussion for the article “Bacon, Egg and Cheese Sandwich”, which resulted in a keep, has a markedly different tone than the AfD discussions of Jess’s articles. en.wikipedia.org/wiki/Wikipedia… And don’t, of course, get me started on the many, many articles on Pokémon or on comic book characters or pornographic actors and actresses.

You will find many examples where those discussing a possible article deletion are rather lenient. To be fair, as far as I can see, the much stricter conduct of AfD discussions is not only true for academics from underrepresented minorities, but for academics in general.

In any case, you can see the potential for bias: when the criteria concerning research or educational impact are somewhat vague, editors use their judgement. And a little bit of bias, unconscious bias will suffice, would be enough to have a statistical effect – and lead to URM articles being deleted more often than others. We would end up in a situation where everybody claims to have unbiased, yet somehow the outcome is biased.

Now, I do not have such statistics. But I have a solution that is, thankfully, independent of the statistics. Increase diversity among those voting on articles-for-deletion. No downside to that, right?

Note: it is very unfair to ask minoritized groups, who usually already carry a heavier burden, to also shoulder wikipedia work. This is allies' work.

— algebraic geometer (@BarbaraFantechi) 3. Mai 2019

OK, there is a downside to that, as @BarbaraFantechi pointed out here. It could lead to the usual problem of asking minorities to, in addition to everything else, shoulder Wikipedia work, and she is right. This is where allies can help.

intermediate tl;dr

So, to come back to the beginning of the thread: If you think that deleting Jess’s articles on Wikipedia shows bias, you can help. First, find your footing and contribute to the project. Once you’ve gained experience, contribute to article-for-deletion discussions. Learn about the notability criteria and strive to apply them faithfully. Try to judge an article as unbiased as you can.

While I don’t want to sound like a chain letter, I believe the following to be true: If ten people of those reading this thread, wondering how to support Jess, go this route, gain Wikipedia experience (if they hadn’t before) and take the time to participate in article-for-deletion discussions, that would go a long way towards countering whatever bias the current sample of involved editors may have. That would be step two.

Step three would be to have a closer look at the notability guidelines, see if there are biases, and if community consensus can be achieved on addressing those biases. (I have mentioned one potential example – no explicit mention of “first X to do Z” as a notability criterion.)

Now, as you can imagine, changing a Wikipedia guideline is more involved than changing an article. You need to have broad consensus, and that means that you are obligated to seek out such consensus, which includes pointing various relevant groups of Wikipedians to your proposed changes and inviting them for discussion. But chances are that, by the time you have reached that stage, you will have learned how to move quite confidently through Wikipedia, use abbreviations like NPOV or words like “wikilawyering” without even thinking, and by that stage you will definitely not need an overlong Twitter thread any more to tell you what to do.

In any case, that was my take on these matters, pointing to some things you can do if you want to help fix the underrepresentation of certain group on Wikipedia. Happy and successful editing!