Wired editor Chris Anderson offers a spirited defense of internet “systems” like Wikipedia, Google, and the blogosphere. Criticism of these systems, he argues, stems largely from our incapacity to comprehend their “alien logic.” Built on the mathematical laws of probability, they “are statistically optimized to excel over time and large numbers.” They sacrifice “perfection at the microscale for optimization at the macroscale.” Our “mammalian minds,” by contrast, are engineered not to apprehend the wonders of the vast, probabilistically determined whole but to focus on the quality of the individual pieces. We’re prisoners of the microscale: “We want to know whether an encyclopedia entry is right or wrong. We want to know that there’s a wise hand (ideally human) guiding Google’s results. We want to trust what we read.”
Google in particular, Chris writes, “seems both omniscient and inscrutable. It makes connections that you or I might not, because they emerge naturally from math on a scale we can’t comprehend. Google is arguably the first company to be born with the alien intelligence of the Web’s large-N statistics hard-wired into its DNA. That’s why it’s so successful, and so seemingly unstoppable.”
Maybe it’s just the Christmas season, but all this talk of omniscience and inscrutability and the insufficiency of our mammalian brains brings to mind the classic explanation for why God’s ways remain mysterious to mere mortals: “Man’s finite mind is incapable of comprehending the infinite mind of God.” Chris presents the web’s alien intelligence as something of a secular godhead, a higher power beyond human understanding. Noting that “the weave of statistical mechanics” is “the only logic that such really large systems understand,” he concludes on a prayerful note: “Perhaps someday we will, too.” In the meantime, we must have faith.
I confess: I’m an unbeliever. My mammalian mind remains mired in the earthly muck of doubt. It’s not that I think Chris is wrong about the workings of “probabilistic systems.” I’m sure he’s right. Where I have a problem is in his implicit trust that the optimization of the system, the achievement of the mathematical perfection of the macroscale, is something to be desired. To people, “optimization” is a neutral term. The optimization of a complex mathematical, or economic, system may make things better for us, or it may make things worse. It may improve society, or degrade it. We may not be able to apprehend the ends, but that doesn’t mean the ends are going to be good.
In a comment on Chris’s post, a fellow named Brock takes issue with the idea that Wikipedia is a probabilistic system. The value of Wikipedia, he says, lies not in the whole but in the individual entries, and the quality of those entries is determined not by statistics but by the work of individuals: “Wikipedia is wrong when a single person is wrong.” Chris counters that, even with Wikipedia, the whole matters: “The main point I was making about Wikipedia was not that any single entry is probabilistic, but that the *entire encylopedia* is probabilistic. Your odds of getting a substantive, up-to-date and accurate entry for any given subject are excellent on Wikipedia, even if every individual entry isn’t excellent.” He then provides a hypothetical illustration:
To put it another way, the quality range in Britannica goes from, say, 5 to 9, with an average of 7. Wikipedia goes from 0 to 10, with an average of, say, 5. But given that Wikipedia has ten times as many entries as Britannica, your chances of finding a reasonable entry on the topic you’re looking for are actually higher on Wikipedia. That doesn’t mean that any given entry will be better, only that the overall value of Wikipedia is higher than Britannica when you consider it from this statistical perspective.
OK, but what are the broader consequences? Might not this statistical optimization of “value” at the macroscale be a recipe for mediocrity at the microscale – the scale, it’s worth remembering, that defines our own individual lives and the culture that surrounds us? By providing a free, easily and universally accessible information source at an average quality level of 5, will Wikipedia slowly erode the economic incentives to produce an alternative source with a quality level of 9 or 8 or 7? Will blogging do the same for the dissemination of news? Does Google-surfing, in the end, make us smarter or dumber, broader or narrower? Can we really put our trust in an alien logic’s ability to create a world to our liking? Do we want to be optimized?
Over a virtual Bethlehem rises a virtual star, and in the manger we find Kevin Kelly’s Machine, conjuring thoughts beyond our ken. Is it Our Savior or a mathematically perfected Rough Beast?
A quick comment I posted over there:
“Umm, all you seem to be saying is that these system are built to be mostly right, most of the time, and we strange weird primitives don’t “GET IT” when we are bothered that they’re notably wrong many times.
That’s a comprehensible view – but not necessarily an easily defensible view!”
Much of the article is torturing terminology in the name of the War On Sense (“emergent” is the new “fractal”).
Fascinating observation on applying a parallel of Gresham’s Law on encyclopedia quality, and by extension to other areas where the “macroscale” web competes with pay-for-use sources such as news outlets.
Herein lies the biggest risk of the web, the devauling of the currency of authoritative accuracy. Equal voting rights make sense in the political realm, but what about in the information arena? Are four bloggers more accurate than, say, the NY Times?
Journalist and gadfly AJ Liebling said that freedom of the press belongs to those who own one. What would he have made of a world where most everyone in the first world owns a press? What will we make from this world?
And the entry on Gresham’s Law is itself an example of the value and non-value of Wikipedia. It’s a detailed economist’s view of the nuts and bolts – nanoscale, so to speak – of the theory with little observation of how it is applied in the real world other than to metal coinage.
The argument that Wikipedia has value as a whole ignores the fact that knowledge only has value when it is operationalized. We do not operationalize the knowledge base of Wikipedia as whole. We can only operationalize the parts. Thus, the quality of each individual entry is the only measure of quality that is relevant in the context of people actually putting that information to use.
The quality of the whole only matters when one is initially selecting a knowledge source. If we know that the overall quality of Britannica is, say, 7 then we should prefer it to Wikipedia with an overall quality of, say, 5, UNLESS Britannica (or other source of equal quality) contains no information on the subject of interest. This puts Wikipedia in the position of being a knowledge base of last resort, not a preferred source.
The increasingly tortuous attempts to endow Wikipedia and other exercises in groupthink with supra-human value seem to be desparately trying to gloss over the main fact: Wikipedia is an attractive resource because it is free and does not require any effort to use.
Regarding “Journalist and gadfly AJ Liebling said that freedom of the press belongs to those who own one. What would he have made of a world where most everyone in the first world owns a press?”
I think he’d recognize that the basic problem still applies where some people metaphorically own a huge offset multistory machine, and others have a mimeograph. He’d say something like “Freedom of speech belongs to those who can get heard”.
Which connects to: “Are four bloggers more accurate than, say, the NY Times?”
It depend on *which* four bloggers.
Four experts on the topic are likely to be more accurate than the NY Times.
Four slavering partisan hacks are likely to be less accurate than the NY Times.
The problem is that evangelists talk about the experts on the topic, as a theory, but it’s too often the slavering partisan hacks who get heard overall, in practice.
I take issue with the “average quality” metric not being weighted by article popularity. You shouldn’t care about the percentage of bad articles in the corpus; what you care about is the chance that an article you are interested in is bad. Since you don’t pick research topics completely at random from a list of articles in Wikipedia, the chances are pretty high that any specific topic you cared enough to look up is a topic that somebody else cared enough to put on a watch list and keep current.
Unless your only goal in selecting topics was to find weak areas in Wikipedia, I suppose. In that case there’d actually be an inverse relation.
i like your skepticism. very healthy. much needed.
ubi dubium ibi libertas
Regarding: “the chances are pretty high that any specific topic you cared enough to look up is a topic that somebody else cared enough to put on a watch list and keep current.”
How high is pretty high? Does *cared* mean *correct*?
The above is a kind of argument by assertion. Underneath it all is a grain of truth, that popular articles likely won’t have unpopular errors (as opposed to popular errors). OK, that’s nice. But is it really some sort of incomprehensible mathematical phenomena that a rich guy with a lot of PR can get a bunch of volunteers to produce something that pleases people well enough to be appealing, even if it’s not really very good by objective absolute standards? That’s almost the definition of mass-marketing.
By providing a free, easily and universally accessible information source at an average quality level of 5, will Wikipedia slowly erode the economic incentives to produce an alternative source with a quality level of 9 or 8 or 7?
Surely if the average is 5, there’re likely to be articles of 8,9 and 10, along with articles of 0,1 and 2.
I’m starting to think that the idea that wikipedia will be all the reference we need (or want) and disincentivate the production of anything of higher-quality, is implausible.
For example, anyone who’s an expert on topic A, and thinks the wikipedia article is inadequate, only needs put up a better resource and add a link to it at the bottom of the existing page as “further reading” or “an expert article which disputes this”.
Sure, not everyone will click-through. But as wikipedia’s popularity goes up, it will the portal which will drive traffic to these secondary expert sites.
“…anyone who’s an expert on topic A, and thinks the wikipedia article is inadequate, only needs put up a better resource and add a link to it at the bottom of the existing page as “further reading” or “an expert article which disputes this”.
This implies that people who truly are experts have the inclination to bother with it. Why would an actual expert, who has invested money and time in developing said expertise, be willing to (a) give it away for free and (b) subject her-/himself to being “corrected” by anyone who disagrees, regardless of that person’s knowledge (or lack thereof)?
Wikipedia is destined to be the amateur model of a knowledge base. Professionals work on material that they get paid for, and professional research is subjected to expert peer and editorial reviews. Amateurs who contribute to Wikipedia may be very passionate about their subjects, but they are not trained in research methods, nor are their entries subject to critical review by qualified peers. Amateur research is as valuable in the context of knowledge production as amateur baseball is in the context of the World Series.
Probability, Superstition and Ideology
Nick Carr makes the humanist case against Chris Anderson’s defense of probabilistic systems like Google and Wikipedia, taking issue with Anderson’s argument that qualitative criticisms of these systems fail to recognize the virtues of sacrificing “perf…
“By providing a free, easily and universally accessible information source at an average quality level of 5, will Wikipedia slowly erode the economic incentives to produce an alternative source with a quality level of 9 or 8 or 7?”
–Question: If it does erode these incentives, couldn’t that mean that society is better off with a free 5 than a paid 8? Surely, Gutenberg’s bible had fewer pretty illuminations than did the hand-copied books of the time? Not to be hyperbolic, but the point is that for most people, “good enough” is just that, especially when we’re comparing free vs. expensive.
That, I think, is a point that’s largely missed, because most of the people who debate Wikipedia are information professionals–the kind who have the resources and knowledge to get access to a large number of information sources. For someone who’s not an elite, who can’t pay for Brittanica or use the intricacies of Google (News, Local, Maps, Groups, etcetera), Wikipedia is an invaluable resource. And remember, it’s the wiki structure that allows it to be free-as-in-beer–there’s no way people could have gone out and said “I’m collecting monetary donations to build an encyclopedia”, at least not on the scale required.
Two points: First off, there are economic DISincentives to producing a corpus of average quality 9. Specifically, people generally don’t think in terms of “X is better than Y” so much as “Y is inadequate, X is adequate”. What constitutes “adequate” varies from person to person but I suspect it forms a bell curve. Increasing quality tends to increase production cost disproportionately, so you wind up with a situation where going from 8 to 9 costs twice as much and will only increase your audience by a small percentage. The only time I can think of where people really stand up and take notice of a “better” product is when it’s just so massively better (10x better perhaps) that it is immediately self-evident that it’s better. Even then, if the cost is substantially higher than the mediocre solution, most people still won’t be interested.
The original article is a bit breezy, and so doesn’t cover *why* these emergent systems tend to be a good thing. Specifically, the question of whether the ends will be good or bad in a statistical system depend entirely on what the driving force is. If the driving force is the demands of the consumer, then you’ll tend to achieve, on average, something that is adequate to your audience. This begs the question of how do you know that the demands of the consumer are really the driving force behind changes in the system, and that’s a question I don’t know how to answer…
But is it really some sort of incomprehensible mathematical phenomena that a rich guy with a lot of PR can get a bunch of volunteers to produce something that pleases people well enough to be appealing, even if it’s not really very good by objective absolute standards? That’s almost the definition of mass-marketing.
What’s not being talked about are wikipedia’s other restrictions : neutral point of view, no original research etc. ( http://en.wikipedia.org/wiki/Wikipedia:No_original_research )
These are meta-criteria designed to prevent certain kinds of abuses of wikipedia. And to empower the collective to resist it.
For example, I may need to be an expert to recognise whether a certain physics theory is right or wrong. But I don’t need to be an expert to recognise that it’s “original research” in the sense of not being backed-up by a cited external document.
Equally, in an ideal situation, whether or not I agree with a particular opinion, I can recognise that it is an opinion and as a responsible wikipedia citizen, delete or rephrase it.
In the case of the rich guy with an army of PR people, application of the wikipedia rules should at least ensure that whatever he wants to put on wikipedia is phrased as “some people think that X, (citation)” Rather than merely “X”
And of course, “some people think that X” is certainly an accurate piece of information.
Naturally, the mere existence of rules can’t force compliance.
But what follows if you start worrying about that?
That there should be nowhere on the web where speech isn’t policed by technology enabled experts? (Or by governments and lawmakers?)
Critics of Wikipedia sometimes act as though Wikipedia started as a giant, powerful information resource, which imposed these crazy anarchic practices on us.
But, of course, it started as a place for these crazy anarchic practices, and happened to grow into a giant, powerful information resource.
That’s the striking result. Which seems to demand a response. Either we celebrate that it seems like giving people freedom can produce something that’s not at all bad. And often surprisingly good.
Or we can go “conservative” and decide that the result is so dangerous that it turns out people shouldn’t be given freedom to speak out and write in public. And that Wikipedia as it is, should somehow be closed down or otherwise changed into a moderated site?
There is a third response : to try to compete with wikipedia with an alternative, gated by experts.
But I’m not sure what other “solutions” people could be asking for?
So we’re saying that the experts want the right to be recognised as experts, and wield control over everyone’s access to knowledge; but without the responsibility of going where the people are, and engaging with them?
The point is, the experts could behave like that because of the institutions such as universities, academic publishers etc. Now the institutions are in flux : there now are institutions like wikipedia which follow a different logic.
So the question arises. If you don’t like the situation, what do you do? It seems to me that either the experts need to find ways to engage with the new institutions like wikipedia, or they have to try some kind of institutional engineering, which is going to come down to either a) building a rival institution and persuading people that it’s better (eg. http://www.digitaluniverse.net/ ) or b) trying to get the existing institution banned or changed.
I don’t see a) having been that succesful, although I think every expert and academic who’s worried about wikipedia should throw their support in with Digital Universe. b) I’m very sceptical of.
We’ve heard this record before : real programmers get paid, they don’t write crappy amateurish free software; real journalists don’t write blogs.
There’s no divine connection between the institution of the market and the expert. The greatest philosophers wrote and thought for love, not money. Some of the greatest scientists were gentlemen amateurs. Many experts do need ways to invest in their expertise, and there’s a tough time coming. But I believe it’s not insurmountable. And some experts will find ways to solve the problem; they will find ways to support themselves while engaged with a public who are often better informed because of the free, mediocre information available on wikipedia.
“Equally, in an ideal situation, whether or not I agree with a particular opinion, I can recognise that it is an opinion and as a responsible wikipedia citizen, delete or rephrase it.
And therein lies the rub. On what criteria is one able to determine, for a certainty, the difference between opinion and fact? Our culture, at present, has a general inability to do so: instead of granting credibility to fact (i.e. statements backed up by valid evidence), all statement are instead treated as opinions, with no one opinion being considered of higher value than any other. This is the kind of background that enables some people to argue that intelligent design ought to be taught in public schools because evolution is “just” a theory.
In an environment where everything is treated as opinion, anyone with a wiki account can delete anything as being “just” another opinion.
Of course you’re a non-believer, you’re relatively old. See, it never fails, older people inherently do not understand the web. It’s OK though, these people will retire and go away soon enough, then we can truly make real progress.
Biff: I hate to tell you this, but Ray Kurzweil says we’re going to live forever. So you’re hosed. -Nick
I’m so sick of seeing the phrases “Long Tail,” “probabilistic system,” etc. Those ideas are merely mathematical models of the world, not the world itself. Google’s search methodologies are designed around probabilistic models, yes, but the fact remains the web itself is NOT probabilistic – after all, it is something created by humans, using human ideas and motivations. And that isn’t alien at all. The alienness arises precisely because Google’s model does not match reality, it only approximates it.
Wikipedia may offer an additional incentive for creating 8-to-9-quality articles: it’s licenced under the GFDL, which means you can adapt the articles relevant to your field into something: correct them, rephrase them, etc., and sell the final document for profit (instead of having to write something from scratch). I wrote a roundup about this awhile ago.
Now what really bothers me today is that google.com or google.ca simply won’t work!! I’m a network consultant and haven’t been able to find the problem for the last 48 hours. You may find this an irrelevant post here, but it’s not when you really think about it…why would only Google not work? It could be my ISP or it could be my router (not my computers since no machine on the internal network can reach Google), but whatever the reason, someone, somewhere, is trying to prevent Google to work and is SUCCEEDING.
Someone very powerful…spooooky???
p.s.: it’s unacceptable that our email address is mandatory here. At least, let us type it in the form of email [at] domainname…
Nice answer.
As Chris put it, “just one blog may be wrong – what you need is to read more blogs”.
:-)
If you limited your commentary to “Wikipedia vs traditionally authoritative sources” then I could read your posting without worrying about how many conceptual omissions I can think of.
Approaching the subject of “web 2.0”, you are probably aware of the fact that it is a philosophical ideal, a marketing term, and probably an historical signpost similar to such non-existent terms as “the printed word 2.0” and “the spoken word 2.0”. Criticising, giving credit or assigning morality to “web 2.0” seems ambitious.
What is the moral value of the “Semantic Web”?. Now that’s a “faceted” subject.
Wikipedia is so inaccurate that its writers are referred to as wikiling writers. Another problem that Jay’s article overlooks is that wikiling administrators actually ban writers who post correct info that is disliked by the administrators. That skews all “accuracy” discussions. http://rexcurry.net/wikipedialies.html
Recently, someone who posts to Wikipedia wised up and improved the “Roman salute” article somewhat so that it recognizes and repeats some of Dr. Rex Curry’s discoveries. http://rexcurry.net/wikipedia-lies.html
Wikiling writers cover up new discoveries by Professor Curry that the salute of the horrid National Socialist German Workers’ Party originated from the USA’s Pledge of Allegiance. http://rexcurry.net/book1a1contents-pledge.html
Wikiling writers cover up Dr. Curry’s discovery that although the swastika was an ancient symbol, it was also used sometimes by German National Socialists to represent “S” letters for their “socialism.” Curry changed the way that people view the symbol of the horrid National Socialist German Workers’ Party. Hitler altered his own signature to use the same stylized “S” letter for “socialist” and similar alphabetic symbolism still shows on Volkswagens. http://rexcurry.net/book1a1contents-swastika.html
Maybe it’s just my training and experience as a journalist, but I look at any reference resource — from Wikipedia to the CIA Factbook to any encyclopedia — as merely a source of leads, or information that requires corroboration.
I’ve written more about viewing Wikipedia as a source of leads, rather than definitive information, here: “Wikipedia: Good for Leads.”
I’ve found errors (or at least questionable statements) in many established reference resources. Believe me, conventional encyclopedias CAN be wrong, sometimes in significant ways.
That’s why, if the information is important to me, I find ways to consult primary sources or get independent corroboration. It takes effort, skill, and time, but it’s not really hard. If the information isn’t terribly important, I cite and link to the source.
Also, since some resources (including wikipedia) are subject to continual revision, I always use Furl to save a copy of the page I’m citing as it appeared at the time of the citation.
– Amy Gahran
Contentious
The Right Conversation