Marginal Revolution: How Google changes information markets

December
S	M	T	W	Th	F	St
	1	2	3	4	5	6
7	8	9	10	11	12	13
14	15	16	17	18	19	20
21	22	23	24	25	26	27
28	29	30	31

« What were the most blogged about books in 2005? | Main | Has there been a free trade breakthrough? »

How Google changes information markets

The underappreciated David Zetland loves Google but sees two negative externalities:

One externality is a reduction in community bonds between individuals who consume more commodified knowledge and less common knowledge. The second externality is a reduction in the creation of new knowledge when Google delivers others’ presentations too quickly—removing the Aha! moment when composing a known concept generates a completely new one. To not be evil, Google should act to offset these externalities. I suggest solutions to build community bonds and increase innovation.

Here is the paper. I too love Google, but I see the negative externalities differently. My main worry is that we learn more about topics with readily imagined keywords, and we neglect hard-to-define abstract concepts. We are headed toward a culture of facts and small bites of information, and may neglect the sort of judgmental wisdom found in, say, John Stuart Mill. I call this kind of knowledge "Factor X." In relative terms, Factor X is becoming more difficult to disseminate; whether it is declining in absolute terms is harder to say.

David's philosophy of travel offers some Factor X. An excerpt:

I wasn't sick very often. Someone asked me what I learned in five years. I learned that parents love their children. The rest of my experience was about coping with me, what I wanted, what I couldn't get and why that hardly mattered.

If you wish to offer an alternative account of the knowledge biases in Google, comments are open.

Posted by Tyler Cowen on December 18, 2005 at 07:48 AM in Web/Tech | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/t/trackback/3576/3860067

Listed below are links to weblogs that reference How Google changes information markets:

» Blog Post Mile 5 from PSoTD
By the way, just a quick aside. With the past few days of political news for Bushmerica, OF COURSE CHENEY WENT TO IRAQ. It was either that or terrorist warnings/attacks/arrests. And now, some damn good reading, if... [Read More]

Tracked on Dec 18, 2005 9:03:32 AM

Comments

On subjects with a variety of viewpoints it promotes the current consensus, which in turn has a self-reinforcing effect.

Posted by: Chris Stiles at Dec 18, 2005 9:19:05 AM

Google is a tool like any other. Using Google to search for Mill-like wisdom is comparable with a misuse of that tool (e.g. treating a toothache with a sledgehammer).

If each time we want to find out something we mechanically go to Google then yes, we're in trouble. The best medicine for this condition is a good understanding of what Google provides, i.e. ranking based on popularity (no. of links) not expertise. In other words, my view is that Google is not "guilty" of anything, but its our use of Google that might be problematic.

Posted by: Gabriel Mihalache at Dec 18, 2005 9:40:24 AM

I would like to see the equivalent of a search thesaurus. For all I know it may already exist in some search tool. Begin a search with one or several keywords and have the tool suggest others that could be selectively added to the search to expand or refine it. It would be nice if this expansion was part of the normal search process, could be toggled on or off and be adjustable as to scope.

Posted by: jim in auston at Dec 18, 2005 11:35:43 AM

I agree with Tyler -- Google and the Net generally tend to seduce us into confusing information with knowledge and knowledge with wisdom. Easier access to all the information out there is great, but it's still just shadows on the wall of the cave. I see no prospect of any institutional changes that would correct that. I'm sure those of you who are in the education business have already seen a decline in students' interest in anything other than what can be Googled. Perhaps some of you have developed a strategy to deal with it, in which case I would like to hear about it.

Posted by: Kent Guida at Dec 18, 2005 11:47:49 AM

We're already losing spelling, and I fear we will lose the poetry of a well-turned phrase once Google brings us effective automatic translation. Machine translation will be great for social scientists, but, poetry and clever wordplay will still be lost in translation.

This might be just another form of "Factor X."

Posted by: DK at Dec 18, 2005 2:33:25 PM

I ran across this passage (in a book, not on the Internet) from Alexander Pope's "An Essay on Criticism" last week. I think it sums up the problems with Google, Wikipedia et al. quite nicely:

A little learning is a dangerous thing;
Drink deep, or taste not the Pieran spring.
There shallow draughts intoxicate the brain,
And drinking largely sobers us again.

Posted by: Jule at Dec 18, 2005 5:15:59 PM

Thanks to Tyler for the post and to others for their comments. I have a little feedback for a few:

1) Chris Stiles is less right than Gabriel Mihalache. In the paper, I argue against the hegemony of some ideas over others, an idea that is popular with conspiracy theorists but without foundation.

2) jim in auston take a look at http://www.kartoo.com/, http://www.aeiwi.com/ or http://www.grokker.com/ (the last is marketed for internal company use)

3) In the paper, I suggest that the only solution for dumbing down (re: Kent Guida's comment) will be the positive lifetime effect on young people of guidance from their teachers and parents. I suggested that Google initiate some carrots for innovation. (Their summer camp for coders is an excellent example). The sad thing, as KG would recognize, is that students do not only cut and paste their "knowledge" into essays, but download and hand-in entire essays as their own work. That problem is the same as the "problem" with porn - it will destroy you if you are not able to handle it.

4) DK will note, of course, that Google (and the Internet) can bring out new poetic voices (as does wikipedia, by offering pages in many languages). So that machine translation (eg, Altavista babelfish) will merely be a tool for abuse. See also (1), above.

Posted by: David Zetland at Dec 18, 2005 8:12:26 PM

DZ, I didn't say that _poetry_ would suffer, but that the well-turned phrase would. Completely different!

This is closely related to what Tyler said about Factor X, but it is more about aesthetics than wisdom. Both will suffer in a "culture of facts and small bites of information." And getting those facts is clearly not an abuse -- it is a great benefit, but a benefit with the side effect that many well-intentioned users will not realize the aesthetic or qualitative factors they are missing.

One example would be if you google the works of Al-Sistani or another Islamic scholar. Even the human-translated ones might seem clumsy or stilted in English. But, to be an Islamic scholar, one must spend 20+ years studying the most poetical and lyrical forms of Arabic possible, and then pass tests that most would-be mullahs fail. I don't speak Arabic, but I would wager that the stiltedness is an artifact of translation, and that in the original Arabic, the leading mullahs are mesmerizing speakers whose political influence is in part due to their eloquence, an eloquence completely invisible to non-Arabic-speaking Westerners.

Note I'm not trying to make any political points here, just to illustrate the misperceptions that could result from proper, non-abusive Googling. History is replete with both good and evil leaders known for wit or rhetorical skill.

Posted by: DK at Dec 18, 2005 9:58:00 PM

Sorry for the shameless self-publicity, I have some work in progress (with Guillermo Caruana and Vicente Cuñat) that tries to formally model a related externality.

The starting point is that goods are multi-dimensional and we can find out about some characteristics without learning everythings. As a consquence, there are indirect externalities in the way different people search for goods (since our choices of what to look for and what to base our decisions on affects what is provided). This puts us into the world of the second best where funny things can happen (in particular the one we highlight is that a fall in the cost of acquiring certain kinds of information can reduce welfare).

We motivated the paper primarily with a discussion of business school rankings but also have a small discussion on how the search on the internet may have made it easier to gain information on hard definable characteristics while the relative costs of obtaining information on the kind on intangible, stimulating, joyful ones (riffing off Scitovsky's beautiful book "The Joyless Economy") has not changed much. It is conceivable that as a result society is worse off.

This sort of analysis needs qualification (which we do not yet have in the paper) insofar as there are distributional consequences that may need careful thought - again using Scitovsky's language it is certainly conceivable that as a society we are first more concerned about satisfying needs and those who are currently unable to do so.

For the curious, our paper (which still needs some considerable re-writing, and at the very least a better title) is available at http://pages.stern.nyu.edu/~hbar-isa/hbigcvccheap.pdf

Posted by: hb at Dec 19, 2005 9:23:42 AM

Getting back to the "search thesaurus" question ... there's been information retrieval research on methodologies that make the use of a thesaurus irrelevant, or at least orthogonal. For example document clustering would tend to group documents on the basis of terms used, and the terms in related documents would tend to be similar but not identical. A search would then make a vector that would (hopefully) point in the direction of some cluster of documents. There are many other techniques. However, for answering a specific question, most experimental retrieval technologies perform no better than a Google-ish keyword search.

But one thing that things like document clustering could conceivably accomplish might be something akin to knowledge, which I agree is being shunted aside in favor of information with current retrieval technologies.

Posted by: Melinda at Dec 19, 2005 9:56:23 AM

"...knowledge, which I agree is being shunted aside in favor of information with current retrieval technologies."

How do we get knowledge? Does it not come from analysis of information? Don't shoot the messenger. Retrieval technologies may make it easier for the lazy to be lazy, but they don't change the fundamental methods for gaining understanding. Would paper-based card catalogs and indexes to periodicals be better than Google?

Posted by: Unknown at Dec 19, 2005 7:43:06 PM

"Would paper-based card catalogs and indexes to periodicals be better than Google?"

No (no free-text indexing), but another search technology might. When information professionals were first talking seriously about the move from physical libraries and books as physical objects to online texts, there was concern about the loss of the serendipitous discoveries that come with browsing the stacks. Similar concern, I think.

Posted by: Melinda at Dec 20, 2005 6:31:35 PM

To Melinda and Unknown: In the paper I state that Google is overwhelmingly useful for research. I also note that a small negative effect is the reduction in innovation that occurs when brain cells are not engaged in generating presentations (b/c they are downloaded) and the far more dangerous problem that they may never be able to generate presentations b/c of the onslaught of "useful" material from the web removes that problem. (I borrowed this theme from Cowen's book "Creative Destruction".) This second effect is part of the "dumbing-down" effect that some see as no big deal, since video games are teaching us useful skills. It is already present when students download papers to hand in.

Posted by: David Zetland at Dec 21, 2005 5:38:03 AM

Our team at netpowersoft studied what a loan seeker want from the internet.
Mean what type of firm is most suitable for him.
Which will provide you the human edited information in most relavent and organised way.
We have gone though companies who deals with loan and mortgage stuff.

Have a nice time..
:)
Home Equity Loan
http://loanskey.com

Posted by: Netpowersoft at Dec 24, 2005 9:10:41 AM

amateurs dronken in bed ^^^ smeris dronken ^^^ cameriera strip in ufficio ^^^ sublime fighette ^^^ etrange femme merde ^^^ jolie vingt ^^^ le plus froid dix sept ^^^ papa trente quatre ^^^ strappare amatoriali merda ^^^ allievo fighetta fotti ^^^ vaghalsig skolflicka kon ^^^ stamning servitris karlek ^^^ cuttier fighetta fotti ^^^ lovely infermiera gruppo ^^^ kjoligere penere tittehull ^^^ kjoligst skjonn kysk ^^^ kjoligst sot sonn ^^^ anselig fingre ^^^ rihno mora avnanismos ^^^ rihno souideza avnanismos ^^^ froid mere mpeg ^^^ mere videos ^^^