Wikipedia's New Definition of Truth

Hugh Pickens writes "Simson Garfinkel has an interesting essay on MIT Technology Review in which he examines the way that Wikipedia has redefined the commonly accepted use of the word 'truth.' While many academic experts have argued that Wikipedia's articles can't be trusted because they are written and edited by volunteers who have never been vetted, studies have found that the articles are remarkably accurate. 'But wikitruth isn't based on principles such as consistency or observability. It's not even based on common sense or firsthand experience,' says Garfinkel. What makes a fact or statement fit for inclusion is verifiability — that it appeared in some other publication, but there is a problem with appealing to the authority of other people's written words: many publications don't do any fact checking at all, and many of those that do simply call up the subject of the article and ask if the writer got the facts wrong or right. Wikipedia's policy of 'No Original Research' also leads to situations like Jaron Lanier's frustrated attempts to correct his own Wikipedia entry based on firsthand knowledge of his own career. So what is Wikipedia's truth? 'Since Wikipedia is the most widely read online reference on the planet, it's the standard of truth that most people are implicitly using when they type a search term into Google or Yahoo. On Wikipedia, truth is received truth: the consensus view of a subject.'"

Food for Thought

[B]ut there is a problem with appealing to the authority of other people's written words: many publications don't do any fact checking at all, and many of those that do simply call up the subject of the article and ask if the writer got the facts wrong or right.

Which raises an interesting question that no one seems to be asking: What if the problem is not Wikipedia at all? What if Wikipedia is a symptom of a much larger problem in our culture? What if the solution isn't to berate Wikipedia for that which they cannot fix, but rather to ensure the foundations upon which the system is based are fixed?

Failures in authority are of far greater reach than just Wikipedia. That's why academia seeks to correct itself on a regular basis. But the rigid standards of academia (standards which have weakened over time) are not applied to all fields that Wikipedia reports on. Using the case of Jaron Lanier, how is an impartial observer supposed to distinguish between a failure in authoritative reporting vs. an attempt to rewrite history for personal benefit? The only way to prove one over the other is to find evidence. In the case of Wikipedia, it must find another authortative party to dispute the original because doing detective work is beyond what is reasonable for an encyclopedia.

Re:Food for Thought

That is if you are an empiricist there is no other way to provide proof other than evidence. I never took a philosophy class so I'm not sure of the term but I think there are far more people who subscribe to proof through consensus which would be wikipedia's methodology. An abundance of rigor tends to make alot of people shut down or at least slow mental processing down to where they are non-functional(admittedly probably by choice). I don't know that converting everyone to empiricism is actually a rational goal.

Re:Food for Thought

Argumentum ad populum [wikipedia.org] ... yes that's a wikipedia link; did I just blow your mind a little?

Re:Food for Thought

To add to my point, the Nintendo DSi announcement is a perfect example. Take a gander at the Slashdot story:

http://games.slashdot.org/games/08/10/02/2116202.shtml [slashdot.org]

"Nintendo finally came out with a solution to the Wii's lack of storage capacity -- a 2GB SD card from which users can execute games"

Sounds pretty cool, eh? Expect that it's wrong. Nintendo announced a solution to DOWNLOAD games to the SD Card. At no point did they confirm an executable solution. (In fact, they seemed intent on steering away from such an announcement.)

But Slashdot's reporting was not the worst. The worst was GameSpot [gamespot.com], a site that SHOULD by all rights be authoritative. Yet here they are putting words into Reggie's mouth:

9:23] "Iwata is addressing the problem of Wii storage," he says. "Soon you will be able to download and store virtual console and WiiWare titles directly on your SD card, and play them off your SD card. This will make the Wii download experience much easier."

I emailed a more reputable editor who was at the event and confirmed for a fact that those words were never spoken. Yet many, many people quoted GameSpot's poor journalism as proof positive that Nintendo announced a solution to execute games off of SD Cards.

What is a site like Wikipedia supposed to do?

Thankfully, this is a case where a mountain of solid reporting existed to counteract the poor reporting. So Wikipedia reports the correct information. But what if this was more obscure information? How would Wikipedia know who to trust? How would they be able to check again bad reporting?

Answer: They can't. Reporters must be help accountable for the factual nature of their statements. (In the case of GameSpot, that means they should have issued a retraction.) If they cannot maintain a reasonable level of journalistic standards, the industry as a whole should start advertising them as an unreliable source.

Re:Food for Thought

Your specific example effectively demonstra a serious limitation for Wikipedia. One of wikipedia's flaws is not that it has such limitations, but that it doesn't recognize them. Wikipedia is not well suited for the task of distilling rumors and such to predict the future. That doesn't mean that wikipedia is worthless or somehow broken, it just means that it shouldn't have entries that try and guess.

Like you said, when operating at its fullest potential, Wikipedia is really an aggregation of well supported facts. I think everyone would be better served if Wikipedia as a general philosophy would remove topics where those sorts of facts can't be found. Keeping them around only serves to draw into question the usefulness of the entire site.

I think the more useful solution is multiple wiki's, each geared towards a more specific category of knowledge, and having the appropriate level of requirements for an entry to be considered valid. A wiki about future trends of the video game industry is not a bad thing, but it has, inherent in its subject matter, a huge amount of uncertainty. The very idea that such information would be compiled in the same collection and through the same process as something as straight-forward as descriptions of historical medieval weaponry is sort of silly. (Of course, it's also half the fun of wikipedia, following the strange paths that you can end up taking by clicking interesting links between entries.)

More subject specific wiki's do exist, and more are popping up every day, but they're all currently stuck in the shadow of Wikipedia. Hopefully as people become more savvy about finding information online, they'll start to look for more focused sources.

Re:Food for Thought

I think the more useful solution is multiple wiki's, each geared towards a more specific category of knowledge, and having the appropriate level of requirements for an entry to be considered valid.

That was exactly my thinking when I launched DocForge [docforge.com]. Topics for programmers need to have a lot more information than a source like wikipedia can provide. We use some wikipedia articles as a starting point and expand from there. Sometimes opinions, especially pro / con arguments, are preferred for some articles because they provide much more insight than a flat reference. Plus we can collect subjective things like tips [docforge.com] that don't belong on wikipedia.

I think this route of categorical wikis is very useful. But unfortunately, you're correct in that most will remain in the shadow of wikipedia for quite some time.

Indeed

Perhaps one of the best things to come from the Internet (for me, at least), is a high level of professional skepticism. I love Slashdot, I read it near religiously, but I know better. The truth for any Slashdot posting is usually found in the comments, or in some misreported part of the article. I know how to look at the comments, deal with conflicting statements, and find the real answer. Sure beats having a single source newspaper.

Ob simpsons quote

Don't you worry about Wikipedia we'll change it when we get home. We'll change a lot of things.

We can only hope...

That slashdot isn't considered some other publication.

Simson Garfinkel

Simson Garfinkel? You mean that singing duo?

Truth...

Wikipedia: Where consistent opinions are correct opinions.

Re:Truth...

Wikipedia: Where persistent opinions are correct opinions.

There, fixed that for you.

A Reasonable Aggregate of Truth

I think we shouldn't look at Wikipedia as being absolute truth, or not truth, but "a reasonable aggregate of truth." I know that's why I look to Wikipedia when I'm curious about something: not as a source of final truth on a subject, but a starting point. Wikipedia does a great job at collecting relevant information and presenting it in an easy to read fashion, but it should only be used as one tool in research.

As the article author suggests, Wikipedia, when compared to magazine articles or books, is still only the best opinions of other humans. True, magazine articles and books typically have more fact-checking involved - because the author has a reputation to protect - but it's still opinion - just like Wikipedia. The only way a reader can assess ultimate truth is to view Wikipedia in comparison to as many other publications as possible - online or offline. This is the scholastic method and should be the method for every Wikipedia reader. I know this isn't always the case, but this isn't always the case for your average book reader or magazine reader either: they read an opinion that jives with them, and it becomes truth - no different than a Wiki entry.

Re:A Reasonable Aggregate of Truth

Perhaps we should just consider Wikipedia a reasonable aggregation of information. Some true, some false.

Wikipedia takes the truth very seriously

http://en.wikipedia.org/wiki/WP:TRUTH [wikipedia.org]

And of course

there is the danger of the self-refferential wiki-loop, where an unverified statement on wikipedia gets used in a reputable newspaper, which is then used to 'verify' the original statement.

The Register loves this sort of thing: http://www.theregister.co.uk/2008/10/17/wikipedia_and_the_mirror/ [theregister.co.uk] is a minor example, but who knows what else has been elevated to truth by circular reasoning? (smart alec answers to *that* question are welcome :))

philosophically groundless criticism

the crticism offered in the story summary is accurate, but pointless. the idea would be to find some sort of impossibly noble source of information for which the criticism leveled at wikipedia does not also apply. since all sources of media suffer from the same sort of suspect appeal to authority or questionable fact checking, then the criticism leveled against wikipedia is not valid in the sense that it makes wikipedia any different from any other media source you can find

all media is suspect, anywhere. you go through life with a good bullshit meter, or you don't go through life at all. there is no such thing, nor will there ever be, a perfectly verifiable and 100% trustworthy media, anywhere on this planet. media is a human endeavour, and as such, it is as flawed as we are. it is not a question of purposeful intent or partisan manipulation, it is a question of the unattainability of true impartiality

it is impossible for you to discover a media source that does not also suffer from the same criticism leveled at wikipedia. so continue using wikipedia, with a healthy functioning bullshit meter, teh same bullshit meter you should have on when reading any other media soruce. the criticism is useless

learn to accept the fundamental limitations of media in your world, and stop expecting the impossible out of media. it is biased, and always will be

Multiple sources

but there is a problem ... many publications don't do any fact checking at all

That's why multiple sources are the best. Whenever sources disagree, the more reliable [wikipedia.org] sources are trusted over less reliable sources.

Verifiability is really an appeal to authority--not the authority of truth, but the authority of other publications. Any other publication, really.

That's just not true. Many talk pages are filled with disputes over "my source X is more reliable than your source Y because ...". That's ultimately a very healthy discussion. And WP:RS [wikipedia.org] does say that some sources aren't reliable enough to be worth including at all.

Calling Captain Kirk...

So, if somebody creates a Wikipedia article called "Everything on Wikipedia is a Lie ", will it start arguing with itself, then explode?

Notability is King

There's also the accepted rule that "Celebrity equals Existance." Don't believe me? Try and write a highly detailed wiki entry about a webcomic that has been consistently updating for years but won no awards, or a music band who has been steadily working on the independant scene but went largely unnoticed by the major labels. Your hard work is sure to be rewarded by a "lack of notability" deletion notice. Does this mean that I don't exist until I get the cover page of People magazine? Wikipedia seems to think so...

Help change Wikipedia for the better (seriously)

There is a real attempt at changing some of Wikipedia's guidelines going on at http://en.wikipedia.org/wiki/Wikipedia:Notability/RFC:compromise [wikipedia.org]

Please have a look, and please chime in. Please strike a blow AGAINST deletionism.

The problems with wikipedia:

Wikipedia is not authoritative.

1. Wikipedia is not a primary source.
2. Wikipedia is not a reliable source.

   Wikipedia's content is generated by pseudo-anonymous individuals who incorrectly assert the public Internet is a reliable source of information. The public Internet is not a reliable source of information, therefore wikipedia is not a reliable source.

3. Wikipedia is not an objective source.

   Wikipedia's editors break the rules governing their behavior and the behavior of others if it will benefit them. As a result, wikipedia advances the subjective views and beliefs of its editors.

4. Wikipedia is not a representative source.

   Contributing factors to this delusion include the competing concepts "notability" and "neutrality", as advanced by wikipedia. Lacking from that discussion, of course, is the question: notable or neutral, to who? Rather than host disputed versions of articles, representing the majority opinion and any significant minority opinions, wikipedia prefers a version advancing assertions, but not facts, which are easily disputed by any minority.

And I frankly despise the appearance of wikipedia in search results, or having some article on wikipedia quoted in a discussion online, as if it provides information of value, in lieu of the reliable primary sources wikipedia references, as if wikipedia itself is the source of that information, and not merely a link farm with some content wrapped around it.

But then, I make a living because of the difference between assertions and facts, and I'm apt to notice such things. Wikipedia is long on assertions, and short on facts.

Verifiability, not truth

Wikipedia hasn't redefined truth. It is very explicit that doesn't claim to report truth, only verifiability. Read http://en.wikipedia.org/wiki/Wikipedia:Verifiability [wikipedia.org] :

The threshold for inclusion in Wikipedia is verifiability, not truth.

If readers sometimes look to it for truth, well, they're misusing it.

Always read the discussion pages.

A couple of times I've had someone "correct" me pointing to Wikipedia, where the article that they're pointing to is one I'd contributed to. Sometimes the article has become self-contradictory under the influence of multiple editors, other times the article is being more actively edited by someone who he happens to agree with. Either way, I "know" at least as much about the subject as Wikipedia does.

You really can't tell what a Wikipedia entry really means without reading the discussion page. In fact, that's often more informative than the article itself.

Actually, it doesn't work like this

Wikipedia doesn't say that A is true because reference X says so. Wikipedia says that reference X tells us that A is true. There is a fundamental difference:
In the first (incorrect) version, Wikipedia cites X and adds something to this, specifically that X is trustworthy and makes correct statements about A.
In the second, correct version, Wikipedia doesn't claim that A is true or false. It just claims that X claims that A is true. Wikipedia doesn't add anything, it simply accumulates facts and let the reader choose whether A is true or not, and whether X is trustworthy or not.

Nothing is true just because you can verify that someone else thinks it is true. That idea is stupid and so is this story.