David Weinberger on Everything is Miscellaneous and the Wonderful World of Digital Information
Jun 18 2007

Author David Weinberger, a fellow at Harvard's Berkman Institute for Internet and Society, talks with EconTalk host Russ Roberts about the ideas in his latest book, Everything is Miscellaneous: The Power of the New Digital Disorder. Topics include the differences between how we organize and think about physical and digital information, the power of the internet to let us consume information in unique and customized ways and the implications for retailing, politics and education.

Cesar Hidalgo on Why Information Grows
Cesar Hidalgo of MIT and the author of Why Information Grows talks with EconTalk host Russ Roberts about the growth of knowledge and know-how in the modern economy. Hidalgo emphasizes the importance of networks among innovators and creators and the...
Cass Sunstein on Infotopia, Information and Decision-Making
Cass Sunstein of the University of Chicago talks about the ideas in his latest book, Infotopia: How Many Minds Produce Knowledge. What are the best ways to get the information needed to make wise decisions when that information is spread...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Brian H
Jun 18 2007 at 3:13pm

So why bother categorizing or sorting or clumping data and definitions at all?

Well, if everything remains misc., then it’s unusable. The pattern-recognition which enables knowledge/understanding/prediction is also a trimming and communications tool. If every effort to communicate or think about the world has to be preceded with an exhaustive catalog or layout of what associations and meanings are about to be assumed for every term used, then it will never happen at all.

That agreement-base of definitions is the stable base upon which the most complex and transient or hypothetical of postulates is built and within which it is tested against reality. So there will always be efforts to freeze these meanings, at least for a while; otherwise talk and investigation becomes impossible.

Stephen A
Jul 5 2007 at 9:59pm

While I can see the argument that the author is making about fuzzy boundaries and the potential of the internet to organize information, I think that an important economic element was neglected in the podcast. Transactions costs and the need for signaling drive a great deal of decision making. In the podcast Russ you lament the lack of interdisciplinary teaching. When George Mason hires its next assistant professor I imagine that an applicant with a more traditional (narrow) economics training will be at advantage over an “everything is miscellaneous” interdisciplinary scholar. I think that this need for signaling will likely preserve many of the arbitrary distinctions that exist in the world for many years to come.

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Books:


      • "Internet Encyclopedias Go Head to Head," by Jim GilesNature, Vol: 438, Dec. 15, 2005, pp. 900-901.

Web Pages:

Podcasts and Blogs:



Podcast Episode Highlights
0:36Intro. How we organize information and our thoughts. Staples laboratory store tries to behave like a website, not like common stores which put most popular items in back of store to force you to walk through. But there are limitations. There are lots of places you'd want to put CDs, near CD players, near software. Customer looking for one thing might want another thing, but there is only a limited amount of shelf space, plus stocking problems. Cables. Staples is also trying to put up more information, but it's still not as much as on the web. Putting the info all there would obscure the items on the shelves. So, even a store that is trying to be a good actor has physical limitations. Aren't other stores like Walgreen's also trying to be more customer friendly? Paco Underhill book: varies by genre. Signs in store as you walk in is also limited. Shelving can't go all the way up because you want ceiling signs to be visible; average height of human doesn't work well if you are in a wheelchair.
7:01Digital world has a lot of advantages by comparison. Three types of order. The way we organize physical things themselves--books on shelves, photos in albums. Second order: we separate the information about the stuff, the metadata--library card catalog, scanning the card catalog or even retyping them was the analog. Greatly reduce the amount of information available. Which things go on card? Maybe can organize in a few different ways--by author, by title. Third order: everything digital. Everything online. Limitations go away, new principles. Information as leaves on a tree, reshuffle cards in card catalog, quotes, anecdotes, stories. Aside: Melvil Dewey, founder of Dewey decimal system, was also interested in phonetic spelling and metric system, and considered changing his last name to Dui, a phonetic spelling. Actually did change his name for a while. "Close to being an insane rationalist." Functional, but 19th century set of beliefs. Dewey decimal system proposed a single "right" way to organize things; but it's still just one way of organizing them. Not necessarily best for all people. Mixed up with a romance, deep set of beliefs that there is order to the universe. Universe has a whole bunch of attributes, and we can sort and organize based on our interests, needs, and culture. Not all categorizations are correct: if category is things you can hammer nails with, a hammer belongs but not a marshmallow; but if it's things that belong in a house, then hammer and marshmallow both fit. Internet is practical about these things; no one centralized real way of organizing. Organizing photos example. Internet is not susceptible to the mistake of thinking that there is only one way of organizing. It's miscellaneous: not simply that it's disorganized but disorganized with so much metadata about the information that it's possible to pull together clusters and clumps on the fly that suit our needs.
16:48Will that change the way we look at the physical world? Yes. Retailing is one way. Bookstores, though, also encourage browsing. World doesn't suddenly become all weblike. But when you go to a store and are online, you might start thinking that you could alternatively get online. Home Depot has people and you are expected to ask where to find things, but they have given up and positioned signs at ends of aisles. Change in notion of knowledge itself. Pluto example: We grew up thinking there are nine planets, but it's hard to define what a planet is. We've only pointed at planets and named them previously, but there has never been a definition. But if you define it by big enough to include the nine we've got, you end up with nine hundred. Neil deGrasse Tyson: Analogous question is: How many mountains are there? No one asks for a definition of mountains or to come up with a small number of them so that children can name them. IAU (Astronomical Union) had some committees to try to find a definition, came up with two criteria: object that rounds itself, clears the space around it. Those two criteria get eight of the nine named planet, but nothing scientifically interesting about those criteria, though they pass a test of sounding scientific. (Is Pluto not round or does it not clear the space?) Alternative criteria: objects that are rocky and have water; objects that have atmospheres--those are scientifically interesting criteria. Useful to science to be able to slice and dice criteria. If you want to mine the asteroids, copper is interesting. We've previously insisted on coming up with small number of organizing criteria, but now we allow more. We like to put things in boxes, but things can go in more than one box. Things can cluster in varying ways.
26:44Is every economic transaction unique? Yes and no. Yes, because every good's quality, individual and situation is unique. But no, because transactions take place in a market and compete with other transactions. Art rather than science. "Market" is a loose term, transactions linked because they compete with each other. No one answer. Student reaction is something that then there's nothing. But by understanding the metadata you can make useful generalizations. Hamlet example. Play by Shakespeare. There is a play. But if you are librarian and look closely, there are three editions from Shakespeare's time, and hundreds of editions that can be tracked and sometimes differ in significant ways. Do we conclude there's no such thing as Hamlet? No, of course not. The answer is "Whatever is useful." Rosencrantz & Guildenstern Are Dead is deeply related to Hamlet. It would be in the cloud of Hamlet books. In a library browsing Shakespeare books you wouldn't stumble on it but on the web you might. Not just a matter of browsing, but also what happens to knowledge. The notion that things have sharp definitions, from Aristotle, has involved most of our intellectual life. Sometimes you absolutely have to do that. Boat-vehicle example. What is the most "real" part of the object? Nevertheless, that gets in the way of thought. The web is making things complex, all about sticking together ideas, infinitely complex, infinitely browseable.
34:12Is this change a cause or an effect? Young people are claimed to have short attention spans; but Russ also finds the pleasure of leaping from link to link to be very high. Do we like that because the web is available and has changed our thinking patterns, or did we develop the web because we like doing it? Russ: This is the third podcast invoking Aristotle. [N.B. Actually, it's the fifth.--Econlib Ed.] Podzinger (now Everyzing). "Social Knowing," chapter in book: making it easier to share like stuff, Wikipedia. Information/facts vs. knowledge: how does Wikipedia serve those two? Third distinction: understanding. Our culture has focused on knowledge; we are the knowing creatures in philosophical tradition. But knowledge is only one important thing. Understanding is why we want to know things. Post-modernism. Facts: What year was Churchill born? Useful, but they are commodities. Facts are like nails. Daily newspaper coverage has also been commoditized. You don't care where you get it. Wikipedia has a box with just the facts--what year were you born? Box metaphor. But Wikipedia is not just facts, not just an almanac. Also about knowledge, knowledge we have settled on, knowledge born of argument. Maybe Wikipedia is commoditizing knowledge, if you want to know what we've agreed upon, you can go to Wikipedia. What will Wikipedia be like in ten years? New articles will have slowed down, most articles should be pretty stable at that point. Google also does this. What's left is understanding. No plural for the term "knowledge," but understanding is multiple. There will never be a single understanding of the world. Human history. Italian grad student: If you are in a bar and you disagree about a fact you'll argue till one of you stalks out. But online you settle it or just disagree and go your own way. And sometimes on the web without a crowd around people actually concede things. Blogs, Wikipedia--"fallibility becomes a marker for trust." Ability to err, opposite of how authority has worked historically. Authority has trouble admitting error, matter of shame. Wikipedia puts up a notice saying that articles are questioned or contentious, retains history--even spreadsheets. Gottfried Leibniz. Paradoxical effect is that Wikipedia becomes more credible, it's "on our side" because it is helping us get knowledge. Can you cite Wikipedia in a research paper? It's not an authority, not as reliable; but study done in Nature comparing Wikipedia and Encyclopedia Britannica found reliability about the same. Response of the two organizations was very different: Wikipedia just changed the articles; Britannica wrote a detailed defense. Nature study is provocative but itself not definitive. We used to go to Encyclopedia for facts, now we go to Google for that. But you wouldn't want to go to an encyclopedia for understanding. Understanding is elusive. Consider cause of Civil War. People are passionate about that; in Wikipedia you can get the full flavor. Over time, existing contents get shorter in printed encyclopedias like the Britannica because you have to cut something. Britannica has "see also" but doesn't have the same notion of a topic as Wikipedia where every few words are linked.
54:49In chapter "Social Knowing": students today behave differently, more socially, emailing each other, IM'ing each other, studying together. Will exams of future be administered to groups? Generational shift, but increased emphasis on testing measurement is anti-social learning. Grad school, book clubs, social element. Today kids IM each other and discuss their homework, but turn it in and get graded individually. Dan Pink podcast, interdisciplinary learning hasn't developed quickly. Web is naturally interdisciplinary. Home schooling in theory would allows this, though social aspect may be difficult. Want to make children curious about the world. Taking history even in good schools turns high school students off. Same for economics. Exams testing large classes hampers the appeal of interesting subjects. Web is more interesting than traditional media. On behalf of disciplines: nobody thinks the world divides into categories and that's that, but there are distinctions that should be maintained. If you are studying science there is a methodology you need to know that differentiates it from, say, economics. It's not that everything is chaos, but that we are able to find what the ways of organizing and clustering are that make sense without being stuck with decisions made by experts for us that are meant to apply in all cases. Specialization is costly but unbelievably productive. "Blogging is the way in which we get to complicate the world again." We want both to simplify and to make things more complex. Sopranos example: romance that we can all watch the same thing and share same national experience; but blogosphere allows people to specialize.
1:06:10Effect of internet and miscellaneous information on politics. Weinberger self-describes as liberal in the book. Does increase in bottom-up stuff increase the likelihood of smaller government? Traditional liberalism of the modern kind is focused on top-down, paternalism. Perhaps both conservatives and liberals see utility in embracing distributed processes. Homeland security will always have a lot of top-down command and control structures, but somebody could come along and suggest an alert system (making this up) using the internet that is actually useful. Not left/right politically. Hanson podcast, could perhaps bet on likelihoods, dispersed information. Intellipedia. Information silo is built into the structure, need to know, prosecuted for sharing information; but security risk may be balanced by value of sharing of information. Weinberger, Senior Internet Advisor to Howard Dean. Will internet play a larger role in the upcoming election process? In previous election, it was considered crazy for Dean to have a blog. Now it would be crazy not to have one. May not be a new killer app, but it is remarkable how quickly the effect of the internet has grown, not only in emails and twitters and IMs but also in the rhetoric of candidates. Men take ties off, more informal.