More than 10 years on...

[dogbiscuit]

Back in 2007, Wikipedia had a major head of steam and those who did not know better had good reason to believe the PR that Wikipedia was going to be a wonderful product. The problems with the broken management of Wikipedia were obvious and the current status seems to be that the management structures have decayed rather than refined and improved, both amongst the volunteers and the WMF.

It is now 2018, I hate to think how many millions of man hours have been poured into Wikipedia. Yet it has not got any improved quality mechanism aside from the original "many eyes" which turn out to be "blind eyes". When do the public twig that after all these years, there is never going to be a time where you can trust Wikipedia as a useful, reasonably unbiased source of information? The next time Jimbo says in response to a complaint that someone out there will sort it, will someone say "You've had 10 years to get it right and it still isn't, how can we believe you?" or simply, "Have you no shame?"

[Casliber]

Back in 2007, Wikipedia had a major head of steam and those who did not know better had good reason to believe the PR that Wikipedia was going to be a wonderful product. The problems with the broken management of Wikipedia were obvious and the current status seems to be that the management structures have decayed rather than refined and improved, both amongst the volunteers and the WMF.

It is now 2018, I hate to think how many millions of man hours have been poured into Wikipedia. Yet it has not got any improved quality mechanism aside from the original "many eyes" which turn out to be "blind eyes". When do the public twig that after all these years, there is never going to be a time where you can trust Wikipedia as a useful, reasonably unbiased source of information? The next time Jimbo says in response to a complaint that someone out there will sort it, will someone say "You've had 10 years to get it right and it still isn't, how can we believe you?" or simply, "Have you no shame?"

There is a vast increase in the level of inline references over this period. Much more of it is verifiable.

[dogbiscuit]

There is a vast increase in the level of inline references over this period. Much more of it is verifiable.

There is a vast increase in the level of inline references over this period.Much more of it is verifiable.
The casual reader still has no indication that the text reflects anything in the references, that anyone has actually checked the references, that the references have been fairly summarised, that the facts referenced are even easily identified within the reference etc..

When I read a reference book, I don't expect to then read another half a dozen other articles to see if what was written is reliable, it is part of the contract with the publishers that a proper process has been followed. I am of course welcome to read other books, but unless I find something either controversial that jars with my understanding or it is something that has inspired me, I would never consider references something that was mandatory to use to verify the accuracy of an article.

It's quite bizarre to suggest that a reference work is in its nature unreliable and it is up to the reader to do the work to find out if what they are reading makes sense. When it comes down to bias, the simple trick of keeping contrary information out of an article is unverifiable, as there will not be any suitable reference to that information.

[CrowsNest]

There is a vast increase in the level of inline references over this period. Much more of it is verifiable.

Yay for Wikipedia. They should put that in the press pack. Now for the real story behind this miracle......

This grand effort to add more inline citations was only partly due to the presence of unverifiable information. It was also a result of the Wikipediots realising their model wasn't all that great in the first place. See, original Wikipedia started out allowing references to be used in the traditional manner - as a diligent author, you were meant to select, review and summarise them into a nice article. You then place all your references in the References section. From that point on, the article is verified and verifiable, and in theory, all future changes must either reflect them even betterer, or see newly added information reflecting additional references.

This was as close as Wikipedia ever really got to aiming to be a real encyclopedia. But see, the Wikipediots, collectively and individually, are too dumb to be able to appreciate this very simple concept, and do what was necessary to make it work. First issue was sneaky vandalism, and/or the simple noobie error of adding unreferenced text. Their bright idea to stop people making additions that didn't reflect existing sources, was to demand an inline citation. So it immediately became pointless to curate general reference lists, because not one damn person on Wikipedia would ever bother to consult them when deciding if the article is still verifiable after new edits.

Second, the battlers got really involved. See, Wikipedians got no time for staid and deliberative academic style processes of consensus divination, they don't want to settle disputes by speaking from the authority which comes from having read entire works, general themes, etc, etc. No, their idea of achieving neutrality is to fight it out over single lines, even single words. Even if this is the only real knowledge of a topic they have. And so of course, the only way to do that, is to rely on inline citations.

And so it came to pass. The Wikipediots quickly manufactured some bullshit narratives about how both these things, which are ultimate failures if your aim is to bring together scholarly minded amateurs to curate a free encyclopedia, were somehow benefits, both for editors and readers. It soon got to the stage where Wikipediots didn't even remember they ever even had a system of general referencing, so they started doing stupid things like tagging articles which were 100% verifiable to the provided general references, as "unreferenced". Undoubtedly, this has seen a lot of perfectly valid material lost, since there's never really been universal agreement that even uncontentious information can remain in place while tagged as lacking an inline citation.

Don't misunderstand me. The use of inline citations has its place, but it really is a monumental waste of time to adopt inline citations as the only method, to be applied to all text that is not making an obviously true claim. And the Wikipediots will even argue there is no such thing as an obviously true statement.

One other notable reason why general referencing fell out of favour, is that on Wikipedia, they absolutely detest the idea of primary authorship. But having one person who is knowledgable of all the sources and who therefore likely wrote most of the article, is the fastest way to maintain it in the face of new additions, and the best way to ensure disputes are settled. Obviously for huge topics, this translates to a team of primary authors, but the principle is the same.

And as damaging as this cancerous level of cynicism was to the sound operation of the model, it was not without good reason, since Wikipedia has always lacked the necessary mechanisms to ensure primary authors are always mindful of the fact that even though they wrote most of the work and know most of the references, they are not the authors of the Wikipedia article in the traditional intelectual ownership sense. Their work and effort is only ever for the common good, and it doesn't entitle them to any status higher than any other editor who seeks to improve or challenge the material.

It is for this reason that many primary authors have simply drifted away from Wikipedia, exhausted by the sheer idiocy of it all. This only further reduces quality, and the evidence is all around, you cannot fail to spot articles where the people adding information haven't even read the entire article, let alone all the references.

And so here we are, in the Golden Age where supposedly Wikipedia is better because inline citations. And in a really really dumb sense, where editors and readers are just lazy schmucks, and Wikipedia is not an encyclopedia but a giant data repository, it is an improvement. But even then, the Wikipediots are in denial about the true nature of their product. As dogbiscuit said, the average reader has not one clue how to tell when text has been altered away from the inline citation, and editors are doing a really really shit job of keeping on top of such changes.

And then there's citogenisis, the gift that keeps on giving. Obviously it can still happen with general referencing, but it was inline citations that really fueled that corrosive and intractable problem, which l completely holes the reliability of Wikipedia under the water line. They never ever speak of it, it is their most dangerous secret of all.

The decline in the culture of general referencing has only fueled a far greater problem of broad, all encompassing quality control. As random drive by editors cram ever more newspapers and single journal pieces into articles to support each new edit, as they are indoctrinated to believe is the wiki-way, as the reference sections balloons to 100+ citations, it becomes virtually impossible to gain an overview of the topic. It is ridiculous when you come across an article, and realise the number of references could quite easily be reduced by a factor of ten, without leaving a single word unverified.

There are a other notable issues in this area too, made worse by the proliferation of inline citations, such as offline sourcing and judging reliability, but as always, with Wikipedia, there's never enough time to explain how shit it really is.

Wikipedia is a joke. It survives only because, as it turns out, there's a demand for a giant pile of trivial shit which is free and instantly accessible. The rise in the number of inline citations to support it, hasn't been reflected in increased page views......

https://stats.wikimedia.org/EN/PlotPageviewsEN.png

Most people who bother to follow these things know that public awareness of how Wikipedia works, shit, what it even is, is so low, that this giant effort to switch from general to inline referencing, has made no real difference to how Joe public uses it. They're still happy to take it as read that the what they find there is possibly accurate. Why? Because they're not using it as an encyclopedia. Sounds about right is good enough for them.

For the more informed reader, for the people who theoretically might be looking for accurate information, the people who are aware of every fault and trapdoor of Wikipedia that stands in the way of believing the information they might read there is true, well, it is pretty hard to argue their time has been saved or their task has been made any easier by this shift. Perhaps only those people looking for a single quick uncontentious fact, like the population of Uganda. If the Wikipediots call that progress, they were even more ignorant of the world that came before them than I ever suspected. Progress was the internet itself, Wikipedia just gets in the way, lazily, sloppily and slowly duplicating what can quite easily be found elsewhere.

[Poetlister]

When it comes down to bias, the simple trick of keeping contrary information out of an article is unverifiable, as there will not be any suitable reference to that information.

An even better trick is not to have an article on a topic at all.

[Kumioko]

There has been a modest increase in certain areas sure, but the amount of article improvement or citations added has not kept pace with the volume of new articles added that all have the same problems as the articles that came before them.

At the same time the project is seeing atrophy in every area from individuals to WikiProjects. Even the larger projects like Military history that once thrived are showing signs of struggle. There simply are not enough new editors to keep pace with those that leave and the editors are leaving at a faster and faster pace. If people really want to fix this problem, and there really is no denying the problem exists, then they need to determine why these people leave. In the past it has almost always been due to bullying, harassment and other civility problems yet nothing is done about it.

So my interpretation of the tea leaves is that the next 10 years is going to continue to reflect what the last has shown. A few trolls, self serving individuals and POV pushers like James Alexander and the other WMFers and admins are going to continue to maintain control and the community and the project are going to continue to deteriorate at an increasing pace.

[BrillLyle]

I had no idea about this history of citations.

I am more comfortable with inline citations. I wish there was a separate pane at the bottom of the edit window / entry window, that could show the citations. I think it's more of a technical barrier than loving using an inline style.

Traditional footnoting would be great.

That said, I really dislike the Harvard method of citation.

And ref name naming conventions -- which I use extensively -- it would be great to standardize those too.

[Renée Bagslint]

I don't think the problem is with the citation style. As DB and CN have pointed out, it's that you have no special reason to believe the citation supports the text. In the Good Old Days, when I published a paper or wrote a book, and made an assertion which was supported by "Smith (1999a) reports that", or "according to Jones [1]" or whatever, there was an a priori reason to believe that I had read and understood those papers, which was that my name, affiliation and contact details were on the publication and my reputation, such as it was, stood or fell by whether or not I was getting things like that right. A peer reviewer, familiar with the literature, would have read my paper before publication, as would an editor. While this process is by no means perfect, it was understood, and it worked reasonably well. The Wikipedia model is the cargo cult version. When I read a Wikipedia article, I am explicitly required to follow up the reference if I want to verify the assertion that it supports. I do not know who the author is, there is no process for checking their citations before publication, and I have no reason to believe or trust the judgement, competence or honesty of the author, other contributors or reviewers should there be any. If I'm lucky enough to be near a university or major public library I can probably do all that work for myself, if I feel like it, and every other reader will have to do the same all over again.

[dogbiscuit]

So my interpretation of the tea leaves is that the next 10 years is going to continue to reflect what the last has shown. A few trolls, self serving individuals and POV pushers like James Alexander and the other WMFers and admins are going to continue to maintain control and the community and the project are going to continue to deteriorate at an increasing pace.

Agreed.

The old guard and their followers are so immersed in their own propaganda that they won't see the problems, but also if they did admit the problems even existed, they'd have to accept that they themselves were responsible, and that (shock! horror!), some of us here were right about something.

Which is good for the Hasten The Day![tm] crowd because the characters in play are so flawed there is zero chance of any substantive change in approach.

[Kumioko]

So my interpretation of the tea leaves is that the next 10 years is going to continue to reflect what the last has shown. A few trolls, self serving individuals and POV pushers like James Alexander and the other WMFers and admins are going to continue to maintain control and the community and the project are going to continue to deteriorate at an increasing pace.

Agreed.

The old guard and their followers are so immersed in their own propaganda that they won't see the problems, but also if they did admit the problems even existed, they'd have to accept that they themselves were responsible, and that (shock! horror!), some of us here were right about something.

Which is good for the Hasten The Day![tm] crowd because the characters in play are so flawed there is zero chance of any substantive change in approach.

You're right, when you toxic and problematic admins like BU Rob 13, Fram, Floquenbeam, Jamesofur, HighinBC, Huon and a whole host of others, we don't need to do anything but sit back and watch. These so called trusted individuals are so bad for the project and the community that they are the best players on our (the hasten the day crowd) team!

[Poetlister]

If I'm lucky enough to be near a university or major public library I can probably do all that work for myself, if I feel like it, and every other reader will have to do the same all over again.

Of course, any wikipedian would say that if you find an error, you should correct it to avoid any subsequent reader being deceived. But:

* Why should you bother?
* You might just get reverted, or other errors may be introduced accidentally or deliberately.
* Why should anyone else trust you?

[BrillLyle]

I don't think the problem is with the citation style. As DB and CN have pointed out, it's that you have no special reason to believe the citation supports the text. In the Good Old Days, when I published a paper or wrote a book, and made an assertion which was supported by "Smith (1999a) reports that", or "according to Jones [1]" or whatever, there was an a priori reason to believe that I had read and understood those papers, which was that my name, affiliation and contact details were on the publication and my reputation, such as it was, stood or fell by whether or not I was getting things like that right. A peer reviewer, familiar with the literature, would have read my paper before publication, as would an editor. While this process is by no means perfect, it was understood, and it worked reasonably well. The Wikipedia model is the cargo cult version. When I read a Wikipedia article, I am explicitly required to follow up the reference if I want to verify the assertion that it supports. I do not know who the author is, there is no process for checking their citations before publication, and I have no reason to believe or trust the judgement, competence or honesty of the author, other contributors or reviewers should there be any. If I'm lucky enough to be near a university or major public library I can probably do all that work for myself, if I feel like it, and every other reader will have to do the same all over again.

That sounds like the ideal is the academic publishing route. I don't think this model is sustainable in a free digital labor model.

It is also very onerous. I mean, could an editor on Mark Wahlberg's page find this type of rigorous research? For some areas of Wikipedia this high standard is ideal, I agree. But for many this would be unavailable.

That doesn't address the existing bias of academic publishing as well. It is not a balanced or fair system.

I naively would hope that a basic but well built Wikipedia entry with solid citations would pull from many sources, be they academic journal, news, etc.

- Erika

[Kumioko]

So my interpretation of the tea leaves is that the next 10 years is going to continue to reflect what the last has shown. A few trolls, self serving individuals and POV pushers like James Alexander and the other WMFers and admins are going to continue to maintain control and the community and the project are going to continue to deteriorate at an increasing pace.

Agreed.

The old guard and their followers are so immersed in their own propaganda that they won't see the problems, but also if they did admit the problems even existed, they'd have to accept that they themselves were responsible, and that (shock! horror!), some of us here were right about something.

Which is good for the Hasten The Day![tm] crowd because the characters in play are so flawed there is zero chance of any substantive change in approach.

You're right, when toxic and problematic admins like BU Rob 13, Fram, Floquenbeam, Jamesofur, HighinBC, Huon and a whole host of others, we don't need to do anything but sit back and watch. These so called trusted individuals are so bad for the project and the community that they are the best players on our (the hasten the day crowd) team!

[Randy from Boise]

Back in 2007, Wikipedia had a major head of steam and those who did not know better had good reason to believe the PR that Wikipedia was going to be a wonderful product. The problems with the broken management of Wikipedia were obvious and the current status seems to be that the management structures have decayed rather than refined and improved, both amongst the volunteers and the WMF.

It is now 2018, I hate to think how many millions of man hours have been poured into Wikipedia. Yet it has not got any improved quality mechanism aside from the original "many eyes" which turn out to be "blind eyes". When do the public twig that after all these years, there is never going to be a time where you can trust Wikipedia as a useful, reasonably unbiased source of information? The next time Jimbo says in response to a complaint that someone out there will sort it, will someone say "You've had 10 years to get it right and it still isn't, how can we believe you?" or simply, "Have you no shame?"

There is a vast increase in the level of inline references over this period. Much more of it is verifiable.

Correct. The graphics are also an order of magnitude better.

t

[Renée Bagslint]

That sounds like the ideal is the academic publishing route. I don't think this model is sustainable in a free digital labor model.

It is also very onerous. I mean, could an editor on Mark Wahlberg's page find this type of rigorous research? For some areas of Wikipedia this high standard is ideal, I agree. But for many this would be unavailable.

That doesn't address the existing bias of academic publishing as well. It is not a balanced or fair system.

I naively would hope that a basic but well built Wikipedia entry with solid citations would pull from many sources, be they academic journal, news, etc.

The traditional model has its flaws, as I said. But it does set out to, and largely succeed in, delivering justified true belief, otherwise known as knowledge.

Erika for her part acknowledges that the traditional model is not sustanable, that for most articles rigour is not available, and that the readers only recourse is to hope. One can hope, but there seems no reason to suppose that Wikipedia, or any other free labour model, dependent entirely on volunteer effort, with no form of editorial direction, can deliver knowledge.

The only possible outcome, as we have seen, is a cargo cult version in which people pretend to write knowledge, and declare success while knowing that they have not delivered it, even on their own cargo cult terms. Fake news. Sad.

[dogbiscuit]

I don't think the problem is with the citation style... ... it's that you have no special reason to believe the citation supports the text.

In the Good Old Days, when I published a paper or wrote a book, and made an assertion which was supported by "Smith (1999a) reports that", or "according to Jones [1]" or whatever, there was an a priori reason to believe that I had read and understood those papers, which was that my name, affiliation and contact details were on the publication and my reputation, such as it was, stood or fell by whether or not I was getting things like that right. A peer reviewer, familiar with the literature, would have read my paper before publication, as would an editor. While this process is by no means perfect, it was understood, and it worked reasonably well.

The Wikipedia model is the cargo cult version. When I read a Wikipedia article, I am explicitly required to follow up the reference if I want to verify the assertion that it supports. I do not know who the author is, there is no process for checking their citations before publication, and I have no reason to believe or trust the judgement, competence or honesty of the author, other contributors or reviewers should there be any. If I'm lucky enough to be near a university or major public library I can probably do all that work for myself, if I feel like it, and every other reader will have to do the same all over again.

The traditional model has its flaws, as I said. But it does set out to, and largely succeed in, delivering justified true belief, otherwise known as knowledge.

Erika for her part acknowledges that the traditional model is not sustainable, that for most articles rigour is not available, and that the readers only recourse is to hope. One can hope, but there seems no reason to suppose that Wikipedia, or any other free labour model, dependent entirely on volunteer effort, with no form of editorial direction, can deliver knowledge.

The only possible outcome, as we have seen, is a cargo cult version in which people pretend to write knowledge, and declare success while knowing that they have not delivered it, even on their own cargo cult terms. Fake news. Sad.

I've just repeated what you've said with a tiny bit of formatting. I thought it was a really good explanation of what was in my mind. I don't think I'd quite put my finger on what the fundamental difference in the two models was. Thanks!