0978: "Citogenesis"

[endolith]

The blatant copyright violation comes second in annoyance to the waste of time looking at sites that have nothing more to add on the subject.

Copyright violation and plagiarism are not the same thing. Also, Wikipedia is open source, so copying it is not necessarily a violation, anyway.

[A_of_s_t]

I didn't really find this funny since I just watched http://www.southparkstudios.com/full-episodes/s15e13-a-history-channel-thanksgiving last night. It's basically the same joke...

[dsawatzky]

Oh and Randall almost certainly has no book in mind whatsoever but incredibly he's made himself look knowledgeable about some unknown topic! xkcd is just a lot of hot air these days.

-- citation needed

[Ronsonic]

How is it that these threads never go a full page before someone posts that the latest strip further proves what a douche / snob / a--hole / d--k / jerk Randall is?

I'd think it was time to stop reading the things and just move on in life.

Having typed it, JMOIL might be a useful standard internets abbreviation.

[fennec]

As a one-time Wikipedia contributor who was briefly appointed to the Arbitration Committee by the Wikimedia Foundation* (and proceeded to do nothing with it and drift away from the site over the next few years)...**

It's all very true, has happened before on a variety of occasions, and you need to stay on your toes. So do Wikipedia editors.

That said, you ought to also be on your toes about other things as well. Wikipedia is a particularly easy case, but there are other cases of falsehood/misstatement/misconception propagation by people who are more interested in justifying their preconceptions than in actual reasoning. Think... politics.

All politics.
Including both your politics and my politics.


(* Earlier I lost a six-way election by two votes and proceeded to schedule a useful internship with my summer instead, so I was busy.)
(** Appeal to authority.)

[uioped1]

There are no “facts”, there is no “objectivity” and most of all, there is no “absolute reality”... Because the “absolute truth/reality” on which their “objectivity/neutrality” is based, just as much does not exist.

Prove it.

[Yu_p]

As for justifying the comic: I think I was aware of such problems before reading the comic. But I'm sure that I wasn't aware of it sufficiently to be careful about repeating the mistake myself.

One may wonder to what extent these things happened before Wikipedia (or, indeed, before the internet). Probably it was more rare, but fact invention and lazy reporting are hardly something new.
It is beautifully explained in today's strip though.

At least in genealogy it has been a well-known phenomenon since long before the Interweb. For instance, here in Norway, there are a lot of unsubstantiated lineage claims back to the old Viking kings, and a lot of «sources» just citing each other.

Well, isn't the only requirement for cyclic references (putting malicious intent aside) the possibility to update reference lists? Basically this means, that any source, that has citations added after its original publication becomes unreliable. And that is, if we assume complete and correct references in all sources.

the solution is clearly that journalists must provide references and citations when writing articles, and such references should only be accepted by wikipedia as fact if the citation trail doesn't ever lead back to wikipedia

I'm not sure here, but isn't finding all cycles in graphs an NP-hard problem? [citation needed]Unrealistic. This basically equals forbidding newspapers as references.

Basically in order to avoid the problem, for any citation it must be guaranteed, that references point to the past and that the root of the reference traceback must provide reliable statements. For citing books this means, that the specific edition of a book must be cited, for web pages (including wikipedia), that the state at a specific point in time must be cited, though I doubt that this is even reliably possible in the internet (e.g. I may cite http://en.wikipedia.org/w/index.php?title=Firefox&oldid=460164605 , but who guarantees that in 20 years that link will still give me the same information as it does today?).

As for newspapers... When reporting on daily incidents, the reliability is probably on the same level as any researchers original results, but whenever a newspaper cites something (obviously without giving a complete reference list), it basically becomes an unreliable source -- and the German minister is a nice example where it was hard even to notice that there was a citation involved.

[Icalasari]

Well, now I'm going to check all the sources for mythological creatures on Wikipedia

I wonder if I can manage to fix damage caused by this phenomena?

Probably not

[shiznits]

So your claim is that this isn't an ad hominem attack because "ad hominem" means "attacking the point by attacking the man", and you're not attacking the point, you're just attacking the man?

That's right! Here, I am criticising the author and his approach. Is that somehow unreasonable? The unapologetic, staunch defence for R in these forums is remarkable. He can't seem to put a foot wrong and the moment someone criticises his style or methods they are lambasted.

Why is it so remarkable? What you did is like walking into a Mormon Temple, loudly insulting their "Prophet", and then expecting the congregation not to be angry ... or to not stone you.

[Bounsy]

Why isn't anyone in here as agitated by the "was was" in panel 1 as me?

If it was a mistake, then I find it a highly appropriate to leave it in the comic as it is. After all, the user typing it into Wikipedia is the one making up "facts" in the first place, so a little bad grammar is further demonstration of the lack of concern he has for correctness.

[leifbk]

At least in genealogy it has been a well-known phenomenon since long before the Interweb. For instance, here in Norway, there are a lot of unsubstantiated lineage claims back to the old Viking kings, and a lot of «sources» just citing each other.

Well, isn't the only requirement for cyclic references (putting malicious intent aside) the possibility to update reference lists? Basically this means, that any source, that has citations added after its original publication becomes unreliable. And that is, if we assume complete and correct references in all sources.

You'll regularly find that the original «source« is a very vague hypothesis, which will be presented as fact a couple of iterations down the line.

[wildgunman]

There's a famous story that a former member of the New Yorker's famously diligent fact-checking department tells about this kind of thing.

There was an army faction located in some country in Africa or South America that the New Yorker was writing a piece on. They were trying to come up with an estimate for the size of the army and the fact checking department went so far as to contact the army general in that country. The general admitted that he didn't actually know the size. Figuring that if the leader of the army didn't know, then a reasonable guestimate would suffice, they printed it as around 2,000. Some time after the article went to print, the general called the woman back and told her that the number was 2,000. No one there had done a count or anything, but reliable media sources now confirmed that that was, in fact, the correct size.

[Brimhack]

The book he's referring to is Outliers, by Malcolm Gladwell.

The chapter in question is Chapter 3 where he lays out the "10,000 Hour Rule", which is the cornerstone of most of the book.

You can read about it here: http://en.wikipedia.org/wiki/Outliers_(book)

[BeagleFury]

I was about to post almost the same identical thoughts as you expressed in your reply, BaReFOOt. Glad you beat me to it.

You did mess up though... FIRST... you need to create the WIKI article... THEN... you can start to provide the references and evidence... FINALLY ... it can become an objective reality truth.

@userxp: Sorry, I have to disagree here. I’ll keep it inside the spoiler tag.

(... read the spoiler yourself in the original message ...)

[tallshipper]

A couple of years ago I coined a word to refer to the related idea that reading a Wikipedia page is sufficient for the R & D environment where I work.
A " Wikispert " is thus a self-proclaimed expert on a given topic based on reading a single source of unregulated internet information such as Wikipedia. I encourage use of this word.

Full Disclosure: I am myself a heavy user of Wikipedia, and an occasional contributor. I consider the work of Jimmy Wales and the Wiki team to be an outstanding contribution to the overall knowlege base of humanity, even if it is not 100% correct all of the time. I have rather less respect for Wikisperts.

[enginerd22]

This would actually be "factogenesis". Or perhaps "de novo factogenesis". It's the creation of a "fact" out of nowhere. Although it's the creation of both a fact and a citation, so maybe it should be "de novo factocitogenesis".

[fredfnord]

This would actually be "factogenesis".

Technically, 'factoidogenesis'. Since a factoid is (or, at least, originally was) "a questionable or spurious—unverified, incorrect, or fabricated— statement presented as a fact, but with no veracity."

...at least, according to Wikipedia.

[finity]

Sometimes I wonder if I am Randall. Or, if he is me. I went to bed early last night, but feel like I didn't get any sleep at all. Perhaps I was up recording my thoughts in stick figure fashion...

[rmsgrey]

Even chronology wouldn't protect against cyclic citations - from time to time, one academic will cite preliminary results from another ("to be published at a future date") - by the time that future paper actually gets published, there can be papers citing papers which cite the original paper, allowing the person who came up with the original preliminary results to cite seemingly independent papers as support for his otherwise dubious result...

[fahrbot]

Congratulation. You've just described how things work at Fox News, where opinions become references become facts as they cycle between their commentators and reporters. Jon Stewart would be proud!

[MuAlphaDeltaTauOmicronMu]

I had been bothered by this because I once pressed a reference that led to... a copy of the article (I've also came across a reference that contradicted the fact that cited it...). It is very hard to stop it. The "original source" is too difficult. What we need is experts (who can be individually wrong but statistics show that collectively would end up writing the truth as it is currently perceived) editing Wikipedia. And we got them, but not enough. And then we got anti-experts (who are different from non-experts, as they think they are experts but are mostly wrong) that blow up the system.
The citation system is flawed in many ways. The other day I added a couple of sentences of information about how St. Martin is celebrated in Portugal. I know it because I have always lived there, so my sources are well... perfect (or almost). It might get chopped off in the future due to lack of reference. I don't even know if they are any sources that I could cite. It's just... truth. (the said thing is a common belief, and I grew up hearing it. So it is pretty much a fact).

[Dancinganimal256]

Randall Munroe, what have you done! By attacking Wikipedia, hordes of raging Wikipedians and deconstructionist (not sure what else to call Barefoot, maybe I should check Wikipedia =D) have filled your forums. My solution, though overly basic, is as follows: Current events - Newpapers (fine), websites (iffy); Everything else - peer-reviewed/scholarly sources only.

[Rklawton]

I'm one of those Wiki people who read and enjoyed today's strip. I've been editing Wikipedia for 9 years and am an admin there.

Today's strip really does happen, but we fix it when we catch on. We do this by citing one of our fundamental tenets: Wikipedia is NOT a reliable source. Therefore, an outside article containing a useful fact that cites Wikipedia is automatically not reliable and can be removed on sight. This breaks the chain illustrated in today's strip. Of course, this assumes that the journalist was ethical and cited his our her sources.

Sometimes the "fact" can be traced to a non-established editor and the "fact" is unique enough that we can track it in the media after the edit but find no source prior to the edit. In these cases, we assume the fact has been mirrored and give other sources precedence.

By the way, it cracks me up when media headlines boldly announce that Wikipedia is not a reliable source. Hell, we know that, and we live by it.

And if you're looking for a wee bit of fun - see if you can definitively find out the late Michael Jackson's middle name: "Joe" or "Joseph". Both version appear on legal documents...

[RogueCynic]

Why is everyone shocked to learn WikiPedia is full of inaccurate information? Dr. Stephen Colbert proved it some years ago. See:http://www.outsidethebeltway.com/wikipedia_handles_colbert_elephant_prank/ Plus, everyone knows Wikipedia is just an April Fool's joke that took on a life of its own.

[mitra]

There's a firefox plugin that somewhat addresses the problem: http://www.wikitrust.net/home
Project was started by a research group of in UC Santa Cruz a few years back, very nice approach - you can look up papers here: http://www.wikitrust.net/talks-and-papers

[chubasco]

Didn't Cheney explicitly do this with a bunch of journalists regarding the Valerie Plame affair? I seem to recall there was a whole Frontline on him (or underlings) leaking it to one journalist, then going to another journalist, and 'citing' the story that the first journalist wrote, and round & round we go...anyone else remember that?

[TomSFox]

… it is not true that “Reality is whatever refuses to go away when you stop perceiving it”.

Of course it is, you idiot!

[Steve the Pocket]

the solution is clearly that journalists must provide references and citations when writing articles, and should be fired on the spot if Wikipedia ever comes up among them.

Fixed that for you. Granted, it's still not a solution, but it's a good place to start.

I didn't really find this funny since I just watched http://www.southparkstudios.com/full-episodes/s15e13-a-history-channel-thanksgiving last night. It's basically the same joke...

Wow. Beaten to a joke by South Park in just under a week. If I didn't know any better I'd say... ah, but I do, so I won't.

[dprovan]

This is hardly a problem that Wikipedia are unaware of. All Randall's done is flag a problem that we knew about anyway and fail to propose a solution.

I thought the comic was pointing out human failings, not a problem with the specific tool Wikipedia. As it was, I didn't take it as fact, just as a funny idea that might or might not actually happen in practice. But we can see he struck a nerve.

As I said, this is a phenomenon people are aware of so why is Randall simply reiterating it and not bringing anything new to the table?

Not everything in the world is intended to solve problems you are worried about. He brought it up because it's funny. I laughed. In addition, most people do not have your specialized knowledge. I had never heard of the phenomenon before. It gets people thinking about where information comes from in general, which is always a good thing.

[jpk]

I’ll try to explain this… again:

either the entire Eiffel tower is in Paris or not.

That’s the thing: It’s not a global either-or. First of all, because of relativity theory, The Eiffel tower and Paris are at a slightly different spot in space time for everyone of us.

Sorry, but no. Unless you can show that there's some relativistic effect whereby the Eiffel Tower would not be in Paris for some observer, your argument is based on an intentional misunderstanding of Einstein. The simplest way to put this is this: Relativity doesn't mean objects' locations in space-time relative to each other are random or arbitrary. As far as I can tell, there's nothing in relativity that suggests that an object which is within a bounded space relative to one observer would not be within that same bounded space for any other observer. That is, if I place an object within a sphere, and throw observers at it at various speeds, they might report the shape of the sphere differently, but none of them will report the object being outside the sphere.
I'm not a physicist, but I'm pretty sure this is correct. I'm open to corrections, though.

[coryh]

Has Pinker fallen so low? I remember when he was a serious linguist. Gladwell, of course, is a making-shit-up machine, and for some reason people love to cite anything he mentions as gospel.

Did you just reassess you opinion of a writer based on a few guesses by people who admitted they haven't even read the work, in a thread about how repeated bullshit gets treated as true? I think you just failed critical thinking forever.

I've read Pinker's latest. I don't recall any wiki cites.

[Netreker0]

Secondly, I really wish internet forum users would once and for all learn what constitutes an ad hominem attack. I was not negating Randall's point by objecting to him personally - I was merely expressing great doubt that he actually has a book in mind.

I think that you're being incredibly disingenuous, sorely underestimating other people on the internet, or adhering to an incredibly strict definition of "ad hominem attack." To me (and most other people, I think), if you spend a post arguing against someone's point, and then you randomly insert some sort of seemingly unrelated character attack, then it looks like you're trying to make sure that when the audience is weighing the merits of the two arguments, the fact that your opponent has whatever character flaw is fresh in their minds. If I said, "Bob is wrong about particle physics. This is why. On an unrelated note, Bob is a pedophile," I'd consider that an ad hominem attack even if I don't explicitly say the words, "and you shouldn't trust anything he says about physics because everyone knows pedophiles have no moral compass and would say anything."

From the perspective of an audience unable to read your mind, I didn't see any other reason to make that last comment. The most obvious would be to say "Randall's argument has no support, note that he doesn't cite any verifiable examples--that's because they don't exist." However, you acknowledge that the problem does exist, which was essentially what the non-example would prove, so what purpose does it serve to bring it up?

Oh and Randall almost certainly has no book in mind whatsoever but incredibly he's made himself look knowledgeable about some unknown topic! xkcd is just a lot of hot air these days.

\

The only logical relationship I saw at this point was "If author is full of hot air about X, then he may be full of hot air about Y."

Of course, all of this is based on an assumption that I now realize is fundamentally wrong: that most of us are here primarily to talk about some obscure, geeky subject matter we like, to comment on the comic, or to argue about some topic or another, and not to make personal attacks, and that those few of us who are here to put others down tend to do just that, and nothing else, within the same post. When you see a passionate argument about something you are clearly invested in juxtaposed with a personal attack that has nothing to do with that argument with no explanation, you tend to try to find a relationship between the two. Depending on how you look at it, I either overestimated you or underestimated you.

Also, I may be biased because I've done this on cross before, bringing up pretty much any sort of character assassination I can under the rules of evidence to impeach witness testimony. Even if I don't say "this guy is a convicted felon, don't believe anything he says," most people will put two and two together and doubt his testimony much more than if they did not hear said character assassination. I'm guessing from your post that you face a lot of people accusing you of making ad hominem attacks when you don't intend to do so--all I have to say is that even if you sincerely mean for your argument against the idea and your personal attack against the person to be completely unrelated, when you put them in the same post most people will see it as a marginally subtle attempt to discredit the idea by discrediting the person.

[sarkeizen]

I think someone earlier on wanted to know how prevalent this phenomena is. One experience I can relate which is similar is that our institution has a president who received a degree. A PhD in <some field>. Now a number of us who have degrees in or related to that field thought this was a little weird because this person has a degree and a masters in a completely unrelated field. This, at least in the sciences is kind of rare. So some of us dig around and find that the school the president came from grants PhD's but not in any particular field furthermore it's attached to a University which has no PhD program in this field. Which makes me wonder how one can get a PhD education in some field without actually possessing any PhD's from that field.

Much to our ire this gets touted in the local press and the honorific "Dr." gets added to virtually any mention of the name of our president. What happens? Someone updates the wikipedia article and when we try to change it they consider the local media to be correct.

I've also had the pleasure of correcting completely fictitious references in Wikipedia. One was to an article in a particular edition of a textbook that most people don't have and the other appeared to be the result of progressive editing of text to make it flow. Eventually losing the meaning altogether.

[Richard.]

The forum users of this comic are incredibly good at overthinking things. I think I saw a post earlier that started talking about the truth of the universe and the ethereal consciousness of reality, and now the question of "is there a book or not a book" is at end-of-the-world omega threat level 19.

No. Just no.

Here's some tips on wikipedia:
1) It probably is accurate, but occasionally confusing. If you love to learn, read wikipedia. It's a great resource. Just don't be completely obstinate or have the comprehensive abilities of a dying sloth.
2) If you read something and it sounds shitty, it might be. Vandals are obvious. Take it with a grain of salt and consider its correctness compared with other sources, but don't explode.
3) Somewhat specifically for the science and historical articles: an article, especially on a complex subject, is written (and read, and checked) by people who truly love the subject; enough to spend time to contribute to the knowledge of anyone accessing the article.

[JustMe]

I think we've wandered from the point. This is a great illustration of what your English teacher told you - look for primary sources. It is certainly not limited to Wikipedia, and I have whiled away many hours tracing down the original source of a statement that pops up repeatedly.

My basic rule of thumb is, if it doesn't have footnotes or endnotes to tell you where the information came from, it's simply for entertainment and shouldn't be relied on. If it does have notes, you need to check the key statements - often they are taken out of context.

Growing informed opinions is hard work.

I'll definitely be using this in my Writing for Engineers class when we talk about references! (Properly credited, of course!)

[JohnHupp]

If we trace this back to "Wikipedian Protester" (http://xkcd.com/285/), does this mean that all facts on Wikipedia originally came from politicians? Thank you Michele Bachmann and Herman Cain for your distinguished contributions to collective knowledge?

[doomvox]

"This is hardly a problem that Wikipedia are unaware of. All Randall's done is flag a problem that we knew about anyway and fail to propose a solution. I know he is a comic writer and not a journalist but for some reason that irks me even more."

Have the wikipedians ("Wikipedia" itself is aware of nothing, he said pedantically) gotten around to proposing solutions? When I was last hanging around there they were still heavily into denial. E.g. Jimbo Wales, at his longnow talk was asked what he would do if the Chinese government decided to subvert wikipedia rather than block it. He told an irrelevant anecdote about dealing with a tiny group of neo-nazis.

And the solution to long-time editors like myself wandering away is supposed to be a spiffy new whizzywig editor... turning the "conflict resolution process" into something besides "the run around" is apparently still off the table.

[doomvox]

"There's a firefox plugin that somewhat addresses the problem"

But it broke immediately after the next upgrade, and everyone forgot about it.

(Someone should tell those guys about the "stable API" concept.)

[BeagleFury]

I’ll try to explain this… again:

either the entire Eiffel tower is in Paris or not.

That’s the thing: It’s not a global either-or. First of all, because of relativity theory, The Eiffel tower and Paris are at a slightly different spot in space time for everyone of us.

Sorry, but no. Unless you can show that there's some relativistic effect whereby the Eiffel Tower would ...

I'm not a physicist, but I'm pretty sure this is correct. I'm open to corrections, though.

You've missed the point.

Spoiler:

One cannot describe the concept of absolute truth or objective reality except as circular definitions to similar or identical concepts... defined but lacking any real description.

How do you know the statement "The Eiffel Tower is in Paris (France)" is true, in the absolute sense. Do you mean to tell me that it is true for the observational frame of reference 300 million light years away travelling at .999% the speed of light relative to earth? (According to Einstien, such a frame of reference omits the Eiffel Tower from reality as well as the entire human race, since it does not exist... yet.)

From a psychological perspective, claiming an absolute truth or objective reality amounts to denying the (non-zero) probability that there could have been deception, hallucination, 'dream world', high tech, magic, and any other ludicrous and ordinarily deniable explanations... you can be very certain, but can you prove a 0 percent probability? (hint: no)

While statements along the lines of the original.. "The Eiffel Tower is in Paris" is quite understandable and comprehensible by sensible people, claiming they have any bearing on an objective reality or absolute truth almost certainly stem from untestable and subjective beliefs, rather than any kind of intrinsic truth, fact, or reality.

[Parrothead]

As a fan (and believer in the value) of pop science, I just hope it wasn't one of the scientists I particularly enjoy, namely Tyson and Greene. I'll be watching the third episode of Greene's new miniseries tonight, and it would be nice not to have to suspect him of lazy citations.