While Wikipedia's model seems far better than an individual publishing house's encyclopedia one where they can draw editors from millions of people throughout the world and are not bound by ads or sales to keep them afloat, in terms of editing it hasn't been working well.
Not every editor has equal power on Wikipedia. The more you have stayed on the site and the more time you spend on the site, you tend to have more say on what gets inside the articles. I have seen talk pages where the same three editors who seem to be part of the same echo chamber discussing issues preventing any alternative opinion or tone to come in the article. A behaviour very similar to Reddit where some subreddit moderators can sustain echo chambers by moderating anything not falling in line. In Wikipedia's case this often even leads to some sources getting picked over the other specially when it comes to media or books.
Is it possible to break the grasp of "editors" or is every user curated platform doomed to reach this state?
But (anecdata) I have my gripes with the way it handles evidence.
My SO has a person (died already) in her family with a (relatively) short wikipedia entry. There are factual errors in the article. We know that, because we have original documents showing that the "facts" reportet in the entry are wrong.
Due to the fact that the documents cannot be published online without breaking copyright laws we can't point to a correction.
She communicated with an editor (or whatever his role was, but he had the right to approve or deny changes to the entry) who stated he could only allow changes to the "facts" if my SO could point to an online source (a credible one he said). Else he would deny changing anything. Or she should upload the documents to wikipedia if she had copyright on them.
He didn't accept the idea to send the documents to him as proof. He made it clear that it is either the wikipedia way of determining the truth or none.
We stopped. Since then I am asking myself how much of wikipedia is actually wrong and actively or passively denied from correction by these gate keepers.
Wikipedia could offer a way to upload documents for editors to review without automatically making them publicly available. This would enable fact checking and correction.
I had heard of it before but since this personal experience with German Wikipedia's "Blockwart" mentality I will never consider them a valid organisation for donations.
Of course there are two extremes where there are problems: disputed topics (especially contemporary political topics) and niche topics receiving little to no attention.
I think for both an improvment can be made by improving the view on the editing history. Showing what kind of editors (new accounts or ong term members? Editing only few articles or across the board? etc.) do what kind of editing. Maybe even broken down to article sections ("this section sees lots of change by new editors, while that section is stable")
But not sure how that would really look like to be usable more or less intuitively.
I would like to see the problem.
That, and executive pay, is where donations go, not to keeping the lights on. The running cost to keeping the servers running is about the same as that fund. It certainly isn't the case that engineering is "done" either, as the open bug count is growing constantly and huge swathes of the software aren't maintained and don't even have maintainers[2]. Just minor things of no importance to modern didactics like, you know, functional video support and rarely used, obscure, functions like uploading.
They have more important things to do like weird elections and appointing cryptobro "disrupters" to the board.
Every now and then there are groups that try to gain control of an article or group of articles on Wikipedia (especially around big current events) and then they start looking for ways to game the system all the while bitching about how Wikipedia is 'broken'.
The way I see it: Wikipedia is imperfect, but not broken, and it has developed a pretty impressive array of defenses against being overtly gamed.
First is that "Reality has a well-known liberal bias". Which means lots of people want a "neutral point of view" which reflects their own opinion.
Secondly, there are lots of trolls and bad-faith actors. So a bureaucracy needs to be built up to deal with it in a structured and consistent (not necessarily fair) way.
Thirdly, what can incentivise people? Most editors aren't paid. Introducing money just leads to corruption. So you end up with fake Internet points which can be gamed.
Penultimately, can you appeal? I've had changes reverted, appealed, and reinstated. You can go all the way up the bureaucracy if you want to break out of the echo chamber.
Finally, fork it. If you think you have a better way of doing things - and can't make internal changes - go run your own Wiki and see how long your system lasts without falling to bias.
Increasing the number of pages should increase the number of pages that are controversial, problematic, or "echo-chambers" in absolute terms.
Are they increasing as a proportion? Is this diminishing the utility of the site in aggregate?
I know, many people will possibly agree that this problem exists - but that sort of empirical observation isn't all that useful. If you want to try a method to fix it - any method at all - you need to first have a reliable way to measure whether it actually improves things.
And for that you first need to find a way to actually measure this - in a reliable, repeatable, unbiased way. You need to find a way to measure all of Wikipedia for this sort of problem. Not just the current state - but through the change history for any arbitrary point in the past as well.
Then do some thorough data analysis, trying to learn as much about the problem as you can. That's your best bet for actually solving the problem.
"Alan Turing proved in 1936 that a general algorithm to solve the halting problem for all possible program-input pairs cannot exist."
https://en.wikipedia.org/wiki/Halting_problem
Alan Turing did no such thing, but that's wikipedia for you. On topics you know nothing about, it feels like wikipedia is great. But on topics you have some knowledge/expertise, wikipedia reveals itself to be amateurish nonsense.
Now throw in politics and it isn't good for anything but superficial knowledge in a narrow band of non-political topics. And that's not touching on the problems of limiting the sources to select news/media sources which exacerbates the problem. If you have time to waste, use waybackmachine to compare the wikipedia pages of "controversial" figures and see how noticeably those pages have been changed in a consistent manner.
The only thing you can do is set up competing wikipedias. My hope is major countries around the world set up their own wikipedias ( hopefully with english translations ) because I don't see anyone competing with wikipedia in the US.
I also don't like the maxime of Wikipedia to prefer "sourced" information, rather than "true" information. Something can be wrong, but if reputable sources claim it, then it goes in WP. I know in the English speaking world the term "truth" has gotten some problematic connotations recently, but frankly I find that disturbung. I really would prefer a site that goes for the truth, and when there is no agreement then I want them to detail the disagreement (with correct proportions to prevent "false balance") and then I want to discuss the fuck out. I feel in Wikipedia, too often the "admins" (i.e. the established editors) decide on a whim.
Finally I think deletionism is the plaque, and we need an inclusionist wikipedia. This is not just because I want to read articles about Pokemon, or because I believe disk space is cheap. I personally think there is a fundamental philosophical or political argument pro inclusionism.
I believe there is a strong and non-obvious aspect to Wikipedia evolution, that is exemplified by an anecdote: When I started doing "family history in New England" research in the 1990s, the Daughters of the American Revolution publications, in hardback volumes, was considered good sourcing, with some caveats .. In the space of five years, in the 1990s, the "facts" printed in impressive DAR fashion, were shredded, in not a few cases. There simply had never been the ability to check and cross-check, so quickly, over such vast numbers of resources. Truth in family history is evasive sometimes, but tends towards very verifiable. The Internet itself, and the eyes and minds of millions of participants, just created an information environment that had never existed; library of Alexandria on steroids, if you want to be colorful.
So now, decades into the process of Wikipedia, we find out that there are not one kind of fact, there is not one kind of editor, and the attention economy is more real than ever. Mix in human nature, and there is a sort of unsolvable situation.
so what to do? admit imperfection, don't waste pages on obviously irreconcilable situations like "the son of the current Presidential candidate of the USA and allegations", I mean sure, but not dominating the response here; and then do find the cases where the process can be evolved, like breaking up smug cliques of editors, when it needs to happen.. fairly, periodically perhaps.. whatever..
Wikipedia -- amazing feature of the entire Internet
This pretty much describes the problem with Wikipedia, but it's highly cryptic because possibly 95% of people are never going to realize this because I don't think they even know a Talk page exists for every article. I read lots of scientific articles and I always visit the Talk pages even when I haven't spotted any errors in the article content. There's tons of crazy petty shit that goes on between the people who edit Wikipedia articles. Whether something is considered a valid source for citation depends on which part of Wikipedia's inner circle you're talking to; in some articles, contributors are shouted down because "blogs aren't authoritative" even when they're the subject falls outside mainstream media and blogs are most proximal to the subject, but in other cases I've seen blogs be considered to be perfectly adequate when it's totally inappropriate. When it comes to anything even remotely tied to news and current events, you might as well skip Wikipedia and just read MSM news articles because those are really what Wikipedia largely considers "authoritative" no matter how biased the mainstream articles actually are. I've even seen totally legit edits and sources being ultimately rejected after a person answers every question of the [obstinate] editor because "nuh uh".
Sadly, I cannot edit Wikipedia myself even though I've never even tried doing so. Even if I'm not using a VPN, it just flat out rejects my IP from even attempting to sign up.
This is how Wikipedia can either fix itself or how someone can replace Wikipedia:
- Be run by an accountable benevolent dictator who can ultimately veto any form of democracy present in the system; ultimately everything hinges on management no matter how the system is configured
- Charge people money to edit Wikipedia (imo this would fix many of the content moderation problems on the internet as a whole at the cost of fewer active users)
- Be transparent about who is in charge (Wikipedia obfuscates this with a facade that suggests that users are no different from one another, which is anything but the truth in 2022)
- Refine official policies around authoritative sources, because right now any existing policies aren't working very well
Wikipedia would have to make it easy to clone parts of its websites to your own service so you can change the parts which are important to you. and then basically the most surfed-on-site wins.
but atm you either fetch the database and set it up with everything.. or .. no idea.
but nobody liked that model either.
we'd need wikipedia with more-git-like backend i think, some guy wrote "levitation" for that, but there would be still a need of setting it up as service and declare which parts are important for you and for which you redirect yourself from your service to the real wikipedia.
One solution would be to teach people to check multiple sources, and each political camp can put their own version of "truth" and people could read all the versions and decide .
I am curious what type of pages bothers people that they need to endlessly fight on attempting to edit them, my bias as a non american is that is about the americans and their political/cultural war and maybe a few nationalists trying to change historical facts on some pages.
I have heard that Wikipedia is "broken" before, but without specific examples the complaint comes across as hollow. And if you do have specific examples, then I think it gives people the chance to investigate your claims.
I believe we'd be better off with real intellectual diversity. Replace the melting-pot model with a federation, a la mastodon, where any topic can be represented by multiple articles from different wikis that have different specialties. The indexing needs to be federated as well, as different groups will want to in-/exclude different alternatives. But just relieving the single pressure point that Wikipedia represents should offer huge benefits to the ecosystem.
My idea for this specifically is to make a pijul repository for Wikipedia, and then a staking/delegation/voting mechanism for acceptance of each vertex in the hash tree. At each level of the tree, a user would be presented with versioning options based on maximum aggregation of unique voters, like in hierarchical clustering, which will roughly define the ‘tribes’ of viewpoints. It would also be possible to render from a specific hash. I would hope to build a user interface with maximum flexibility to explore across ‘versions’ and encourage voting and merges across clusters.
I’ve also considered using randomization, as in, selecting a hash to render with probability equal to the vote within the user’s cluster. This would be to prevent too much separation among the clusters.
Of course, Wikipedia isn't apolitical: since the beginning it's been full of people who consider themselves critical thinkers and skeptics interested in spreading the truth as science tells it, which is itself an ideology that is necessarily opposed to political or religious ideologies which assert some truth that can't be refuted or proven by evidence, and of course to conspiracy theorists who openly detest science and expertise. So yes, people and organizations who make enemies of Wikipedia's ideology won't get nice or even 100% fair treatment in Wikipedia articles. Anyone who sees this as some new development due to recent political currents wasn't paying attention during the 2000s. If you see it as a problem, you're free to follow in the footsteps of widely mocked sites like RationalWiki or Conservapedia and create a wiki that favors your ideology instead. I think it's telling that even people who lean more left or right wing still rely on Wikipedia rather than those places.
One hypothesis is that, yes: you're hitting on an intrinsic aspect of human nature.
A possible remedy would be to set up curated sites across the spectrum, and query a 'millepedia' front end to ask: "Who is the all-time worst political figure, and why is it $GUY_I_DISDAIN?"
For instance, this article: https://en.wikipedia.org/wiki/Biden%E2%80%93Ukraine_conspira...
>The Biden–Ukraine conspiracy theory is a series of unevidenced claims centered on the false allegation that while Joe Biden was vice president of the United States, he engaged in corrupt activities relating to the employment of his son Hunter Biden by the Ukrainian gas company Burisma.
This is not a statement of fact, but an opinion of the person writing it. The fact of corruption is the question at hand.
The Vice President's son was being paid $1,000,000/yr to sit on the board of a foreign gas company.
That isn't a disputed. Joe and Hunter Biden don't dispute this.
This article should be listing the basis of accusations against Hunter, and listing refutations to those accusations.
Instead, this reads like an opinion piece/defense/misdirection.
Wikipedia should have a culture that reflexively recoils at stuff like this. I should have no idea what the political leanings of the editors at wikipedia have, but I do, and they make it obvious.
There you'll run into less-to-no helicopter parents, and stand a much-better chance of creating most of a great article before they show up. Load up on good citation prospects for solid content.
Over to the left on any page you can see how many visits/day it's been getting - that can help you stay away from them.
-I agree with how articles are written, so Wikipedia isn't that bad - maybe still a little rough around the edges, but all things considered I trust it.
-I disagree with how articles are written, and the problems are unsolvable within Wikipedia.
Allowing different editors to one article has not resolved this difference. Given that we are approaching 20 years of Wikipedia being widely popular on the 'Net and this problem has persisted, a solution likely requires something on the technical level of the site, itself, (or, more likely, a new site outside of the Wikimedia Foundation) that changes fundamentally how it operates.
To the extent that competition exists for encyclopedic wikis, the ones I have discovered use some version of a preferred view on what articles should be published and what sources are acceptable, so they are essentially using a nearly identical process as Wikipedia, thus suffer from the same problems.
Imagine if GitHub maintained editable standards through a "volunteer" bureaucracy for what software could be published to the site; what a software can do; what language is used for a given software; deciding which libraries are permitted for use in a project; not permitting "duplicate" software, even when they share the same name (users and projects still maintain unique names). To summarize, this version of GitHub would suck and nobody would use it, but this is analogous to how Wikipedia is managed.
Since the problem revolves around how a single article is to be created, edited, and sourced, and given the difficulty just in this concept, the rather obvious solution is to have multiple articles for a given topic where each is created, edited, and sourced however the maintainers of such articles desire.
The next problem of course is this generating an incredible number of articles, each predictably with its maintainer-guard dog. The way to make this usable could be through implementing "universes" of articles. Some universes could operate on some revision of the current Wikipedia and its bureaucracy, others may be purely top-down, hand-curated lists, or could include multiple universes.
This is one sort of way to deal with what is effectively fundamental problems with observing the Universe combined with aspects of the human experience.
I believe I can have a positive contribution to Wikipedia but they never allowed me to write, since like, 10 years.
What changes have you made to Wikipedia that you’ve encountered difficulty with.
I believe I am part of such a cabal, despite my contributions on Wikipedia totaling about half a dozen in the past decade. Why would I be part of such a cabal? Well, first, I have no contact whatsoever with the other editor other than edits made with the same purpose — and that purpose is to deflect an attempt by other editors to erase the details of a (well-sourced) historical event from Wikipedia and replace it with alternative facts more amenable to certain political ideologies. I (and my fellow cabalist) have explained in detail on the talk page that the other editors were pushing a fringe position which was not backed by any reliable sources, and with this justification I have removed every edit made by those editors attempting to change the facts in the article, even to the point of removing tags they add to the article indicating that there is a dispute as to its accuracy! And yet I believe I am in the right, due to overwhelming evidence of the correctness of the article's contents, and incorrectness of the position being pushed by (what I see as) rogue editors.
The thing is, probably every single such cabal is maintained with essentially identical motivations, at least in the eyes of the cabalists. If even some fraction of us are correct, it shows that you have to choose between cabals and an encyclopedia of misinformation and propaganda. Except, of course, haha, there's no choice at all, because some portion of the cabals are pushing misinformation and propaganda. In my view the Wikipedia experiment has failed, and not merely in some contingent way but in a way that shows that the concept cannot be executed as imagined, an "encyclopedia anyone can edit". Some [substantial?] portion of the information contained will be misinformation, for the same reasons that misinformation exists at all.
Unfortunately, insofar as Wikipedia seems to have displaced many other sources of information (which, to be fair, needn't have been any more reliable — a traditional encyclopedia no doubt contained misinformation spread by the editors, with the only saving grace being a relative dearth of information limiting impact), I fear that it will prove to have been the greatest setback to the aggregation of accurate human knowledge so far in the history of humanity.
Can you point out any real issues in their articles? Any widespreaded misinformation or actual issue? Most complaints I always see are usually about disputable details, not about actual problems.
Would be interested to know any future planned scope of work. What are the current big issues needing resolving?
- Only allowing "reliable sources", and defining that based on criteria which essentially excludes any right wing publication, even when the unreliable source has irrefutable evidence of the claim (video, etc). Similarly, citing claims from 'reliable' sources that are misleading, or slightly incorrect (you don't need to look at many news articles to find an example of this)
- Using social academic sources, like gender studies, to cite things that are literally just opinions. Social sciences have a lot of these publications putting forth claims, that are basically citing each other in a circle, with the root claim being someone's opinion (usually grievance studies).
- Citing lack of 'relevance' to stifle any claim that doesn't fall under the above two. Who decides what's relevant? Well that's the trick; there's no objective criteria, so the power mods decide it.
The main Wikipedia owners then choose, like they do now, which editor they support as default.
And the reason it cannot be fixed rests on the very nature of humanity and communication, which is that many times, people tend to be prideful, stupid and inarticulate. Thus, one does not "fix" a bad Wikipedia article so much as one outlasts the bad editors who keep writing crap and/or deliberately suppress opposing views.
An example of that would be the Wiki article for "Fact", which can be viewed here: https://en.wikipedia.org/wiki/Fact Look at the talk page for this same article - and then look at the revision history for the article itself. Some literally absurd explanations for the meaning of the word "Fact" were kept in the article for far too long.
Oddly enough, there's another English Wiki out there, a "simple" one and they too have an article about "Fact", but with a much less cluttered talk page https://simple.wikipedia.org/wiki/Fact Why is this talk page (and article) it less cluttered? It's because there are fewer people there, and thus less persons pridefully defending their obvious errors.
So then how does one write the "lede" sentence/paragraph for such an article? Anyone who thinks can tell you that the word "fact" has a very broad application and a number of application-specific variations in meaning. However, a skilled writer, if allowed by the resident Wiki editors, could post a simple but apropos opening blurb. Where I do so so, I would say this "A fact is an item of information offered up as a genuine portrayal of an actuality. When offered by a bona fide reliable source, it's generally accepted as true, unless otherwise disproved".
Now, given that I'm the author of this blurb, and given that I too tend to be prideful and defensive of that which I write (and feel is true), am I the best arbiter of of the truth, precision and applicability of my blurb? The answer of course, is no, I am not. But neither is the mosh pit known as "Wikipedia".
Instead, what Wikipedia is a brute force battle royal; an extended donnybrook which plays out over weeks and months. And what do people do when caught up in a massive, extended brawl? They join forces and pile on the opposition to defeat them. And at any given moment, that's also what Wikipedia is: It's a small network of entrenched editors who stay aligned with themselves, fighting off all interlopers via the means of "Admin" status and collective self congratulatory behavior.
In fact, some of them go so far as to apply to themselves a user name which exalts them and seeks to place them beyond question. Here's a user account which is a great example of that: https://en.wikipedia.org/wiki/User:Neutrality
As reflected by this user's Contributions history, User:Neutrality https://en.wikipedia.org/wiki/Special:Contributions/Neutrali... is one of the long-time guardians of the status-quo at Wikipedia.
And as clearly evidenced by his Contributions history, by being a long-time editor who is part of the Wikipedia "Admin" club, User:Neutrality, has virtually unfettered leeway to edit, revert and opine; while at the same time, disallowing any effective rebuttal.
So no, the biases and other problems at Wikipedia cannot be fixed, not unless more people across a broader political mindset join and work their way up to Admin, so as to offset the entrenched left-leaning cabal which runs things there.
As George Carlin used to say "It's a big club, and you ain't in it"
Open source software gives a way to address those political issues. So, they exist in OSS, some open source projects get these elaborate governance structures, they invent a governance LARP with elections and democracy etc. But some OSS projects have not had this.
Cantrill offers an interesting insight about this in his talks... He points out that merge algorithms give a powerful way to fight this, because once you have a merge algorithm you can allow two realities to exist simultaneously. He says the LARP is a result of being “forkaphobic”, afraid of allowing the project to splinter into two identical projects. Whereas, he says, if you look at the Linux kernel, Torvalds can tell you you're full of shit and you can do the thing anyway on your own branch and people can be like “that's actually really interesting” and Torvalds can admit he was wrong.
So the fundamental thing you have to solve is allowing an article to exist in two independent states that get merged together at the end. And if you can solve this then you do not need a political structure to get everybody to agree because your technological process tolerates disagreement.
1. By the American political problem I mean this particular construction of right-vs-left where divisiveness reigns and nobody agrees on even basic facts... there is some of that, but it's not the problem you're having. 2. Some other problems include:
- Wikipedia articles are typically extremely long. This is necessary because they do not have hierarchy but instead kind of cross-link to their peers which prevents any other way of attaining good depths in education. Information is routinely repeated in many pages.
- Encyclopedias are biased to a sort of declarative information, whereas in most cases you really want to provide imperative information: think of what would happen to a university if we forbade labs, recitation sections, homework/exercises, practice tests, instead you must learn everything from lecture without ever assembling a circuit yourself. Who is gonna learn electrical engineering there? But Wikipedia is a fount of information about electrical engineering components, mathematical topics, there are physics articles. Sigh.
- And then in terms of accessibility, Wikipedia has not solved the hard problem of allowing users to distribute computations in some vulgar programming language (Excel, HyperCard) similarly to how they enable typesetting in a vulgar markup language. As a result there is nothing interactive, there is barely any video or audio, everything must be textbook-serializable. Which makes it hard for us to have a volunteer effort of, here is a 150-word card on a topic, here are the deeper dives you can go into, here is how you can interactively play with this concept, here is some volunteer reading it to you so that you don't need a screen reader, here are examples to test whether you have understood the idea, etc. You want the same idea to have many different perspectives.