moderated by Paul DiPerna

Marshall Poe Interview

Paul DiPerna:

Marshall.. I first came across your work when I read "The Hive" in The Atlantic last year. What led to your interest in Wikipedia, and wikis in general?

Marshall Poe:

Well, that's an interesting story. I learned about Wikipedia like most people, that is, I just stumbled on it. I think I was Googling myself (I admit it) one day in 2002 when I noticed a page returned from Wikipedia had my name on it. Weird, I thought. I clicked through and it was (of all things) the "Sigismund von Herberstein" article. SvH was a 16th century Austrian diplomat made famous by his seminal description of Russia Rerum moscoviticarum. I wrote a book on early descriptions of Russia, and it was cited in the article. All this, I thought, was very strange. SvH is obscure. My book was even more obscure. Yet someone took the time to write an entry on SvH and, more remarkably, to read my book and cite it. The entry was, I should add, quite good. At that time, I didn't really know what WP was, so I began to look into it and discovered how it worked. Of course, being an academic, I was amazed that it worked at all, and realized that, historically speaking, it was virtually unprecedented. Never before had this sort of open collaboration been attempted and if the entry on SvH was any measure, I never had it worked so well.

All this made me really curious about Wikipedia, but I didn't have much to do with it until several years later when I was working at The Atlantic. I had moved from the editorial side to business development, and we were looking hard at ways the Internet was changing the production and dissemination of serious information. The top people at The Atlantic are really forward looking and they were very interested in the new media. So I did some research and wrote a long memo on what would come to be called "user-generated content." Wikipedia was one of the sites I profiled.

It was really as a result of that work that I encountered the amazing story of the founding of Wikipedia, that is, how Larry Sanger and Jimmy Wales, having failed to launch a rather conventional encyclopedia (Nupedia), then opened editing to everyone and created Wikipedia. Wikipedia, it seemed to me, was a terrific example of how insight and serendipity operate together in many successful online projects. An online encyclopedia is a good idea for a whole host of reasons. This Sanger and Wales knew. But they didn't know how to produce it. They didn't really "plan" Wikipedia; they discovered it. I think the same can be said of eBay, Craigslist, Napster and many other Web 2.0 projects. The fact is that we just didn't know that people could collaborate productively the way they have in the past decade. The propensity for mass collaboration online just wasn't apparent to anyone (though Richard Stallman and Eric Raymond might object to this statement).

Paul DiPerna:

We probably should back up a bit so that people know how eclectic your professional life has been..

How does your graduate school experience at Cal-Berkeley help, and influence your work today?

Marshall Poe:

Eclectic is the right word. I have varied interests and I've been lucky enough to be able to pursue them in a variety of settings with a whole bunch of really smart people. That's a true blessing. In high school I was a jock with geeky interests. I majored, so to say, in basketball (I still play), with a minor in all the other sports. I went to college to play basketball, but there I met a professor - Dan Kaiser - who became my mentor and changed everything. He turned me on to the life of the mind. Absolutely amazing person. Meeting him was about the luckiest thing that ever happened to me. He was a Russian historian, so I became a Russian historian.

I wouldn 't say Berkeley really had much to with my work today. I was a graduate student and spent nearly all my time in the library. I saw my first online catalogue there, and I got my first PC (one of the original Macs), but the things I was into really didn 't touch on the Internet or collaboration. In fact, there was no Internet as far as I knew and historians are among the least collaborative folk on the planet - they almost never work in teams. For me, it was books, books and more books at Berkeley. And a lot of basketball (they have great pick up games).

Paul DiPerna:

What position do you like to play? Any favorite players.. past or present?

Marshall Poe:

I'm not as mobile as I used to be, so my game has become binary - I either play post (I'm 6'4") or I play on the parameter (I'm a decent outside shooter). I always tell the people I play with-- If you see me dribbling the ball, we're in trouble. As to players, I really liked George Gervin, James Worthy, and Larry Bird. I don't really follow basketball today, so I've no particular favorite players. I hear this kid Kobe something or other is good.

Paul DiPerna:

I understand you spent much of your time in the 90s, and shorter stint from '00 to '02, teaching topics in Russian Studies at Harvard.

What was it like to be in Cambridge at that time?

What do the experiences there mean to you?

Marshall Poe:

Well, things didn't change much when I got to Harvard. Historians spend a lot of time in libraries, and Harvard has one of the best Widener. I lived in that place. It really wasn't until the early 1990s that I became interested in issues of scholarly communications, and those led to interests in the Internet. I loved listservs, and was an active participant on the first Russian history list (H-Russia). I understood that they had the potential to open up scholarly discourse. I also saw that the Internet could be used as a low-cost publication device for "cold" materials such as the ones I (and nearly all scholars) produce. "Cold" here means articles and books that are of interest to only a small number of people and therefore can't be economically produced in print.

Early Russian history is a small, scattered field, and the materials we produced were read by only a tiny audience so are hard to publish. So the listserv and the web seemed to me to be perfect ways to: (a) bring us together as a field and (b) to disseminate our materials to those who needed them at very low cost.

With this in mind, I founded the early Slavic studies list (ESSL at the time, now H-EarlySlavic) and began to post materials I produced on the web (e.g., source material). Both of these initiatives gained attention in the late 1990s, and I began to write on the subject of scholarly communications and the Internet. That's how I came to write "Note to Self: Invent New Publishing Model." This article and another in the Chronicle of Higher Education (both are online) caused a stir because I was, in effect, proposing that the current model for scholarly communications-centered on the department, conference, university press, and library-- were in need of reform.

I became more interested in issues of scholarly communication and collaboration than I was in Russian history per se (though I was still writing the latter). After all, there was a revolution going on, and I wanted to be part of it. I didn't want to be studying history while history was being made. I also wanted to write for larger audiences. So, when I got an offer to come work at The Atlantic, I jumped.

Paul DiPerna:

Based on your experience and intimacy with inner-workings at The Atlantic, how do you see the Web changing mass media and journalism?

Marshall Poe:

This is an important question, perhaps the question for providers of serious content. If I had to boil it down, I'd say one word: video. We know that people would usually rather watch than read. Whenever people are given the choice, they will almost always watch. The growth of movies and then TV are proof of this fact. And this preference makes perfect sense.

Think of it this way. Writing/written words is to information as composition/written notation is to music. Composing music is skill and its language is written notation. You have to be trained to compose music and read it (hence we have music schools). But most people - without any skills or training at all - can enjoy performed music. The same is true of writing/written words. Writing is a skill and its language is written words. You really need to be trained to write and read (hence we have grammar schools). But anyone - without any skills or training at all - can enjoy a performed written piece, that is, a play, movie, video, or what have you.

The preference for watching over reading has existed forever (or rather, as long as we've had writing and reading!). Yet until recently, we had to read and didn't have much to watch. The reasons for the popularity of reading were two. First, before film there was no means to permanently capture a performance. Plays are one-off. So if you wanted to revisit a drama (or what have you), you had no choice but to read it and re-read it. Second, even after a means to permanently capture a performance existed (on film), it was not economical to do much capturing. Film is expensive and hard to deliver, so that means that you are not going to film very much and most of what you film is not going to be broadcast very widely. Now, however, we have both a means of capture (digital video) and a means of delivery (the Internet) that are so easy and so low cost that pretty much anyone who puts their mind to it can "perform" information that once could only be efficiently written and read.

Consider this. The read-only web made everyone a publisher. Blogging made everyone a journalist. YouTube will make everyone a filmmaker.

So the question that faces print content providers today is how to migrate their content into video. They know people would rather watch than read, and they know low-cost, high-quality video can be produced. The question is one of execution. In that sense it's a very exciting time to be a "creative" in a publishing, newspaper, or magazine company, because new video genres are being invented as we speak.

Not surprisingly, I suppose, the path is being blazed by people outside the institutional confines of these companies. Take the remarkable video "Amen Break" by the artist Nate Harrison. It's the history of a drum break told in sound and video. Watch it and tell me if it doesn't remind you of an Atlantic or New Yorker article. It's original, informed, a bit quirky, and extremely well executed. It was also done, I imagine, at very low cost. And it is available to anyone with an Internet connection anytime for free. "Amen Break," I think, is a good example of what serious publishers (or perhaps we should say "content providers") are going to be producing in the coming decades.

Paul DiPerna:

USA Today just did a major overhaul of their website. It looks like they've tried to make the site much more interactive for the reader/user by incorporating community features..

What's your impression?

Marshall Poe:

It's good, but I still don't think that print publishers have figured out how to do "news" on the web.

Here's the problem...

Newspapers and magazines are generally consumed where: (a) it's inconvenient to access other media and (b) you need mobility. Since paper/print is silent, self-contained, light, and low cost it's perfect when you are (say) in the loo or on the go. No web-based computer anything can compete with ink/paper in these places - they are the print niche (though portable digital paper might do the trick someday).

"... the Web's primary
distinguishing feature -
the hyperlink."

Trying to put an entire big newspaper or magazine online is like trying to put them on TV: you are taking a genre designed for one medium and plopping it into another with different rules. The folks at USA Today have done a terrific job of putting a newspaper layout onto the web. The problem is that a newspaper layout doesn't take advantage of the Web's primary distinguishing feature - the hyperlink. On the web, you can go anywhere from anywhere else instantly. This ability more or less dictates the proper form for the presentation of "news" on the web'that form is a list of hyperlinked headlines + short descriptions, divided into "sections" and ordered from most important to least important. That's it. For examples of this sort of design see Metafilter or Arts and Letters Daily.

I think print publishers moving to the web are paying too much attention to the fancy front page with charts, pictures, colors, videos and other eye candy. The graphic-heavy front page (newspaper) and cover (magazine) are devices used to entice people to buy newspapers and magazines. That's why they so often have nearly naked ladies on them. Gets attention, sells copy. But it doesn't work on the web because news websites are not sold - they are given away. Moreover, no one is going to be enticed to peruse your newspaper or magazine website because you have Ms. Starlet Du Jour on your front page or cover. If they want to see pictures of Ms. Du Jour on the web, they can go to any number of fan sites and see all they want (and probably more).

Here's the first principle for building a serious, intelligent audience on the web:

People will come to your website if you have something smart and useful not available anywhere else.

Readers do not go to Metafilter or Arts and Letters Daily - both of which are very popular - because of their fancy front pages. They don't have them. They are just hyperlinked lists. Rather, they go to these sites because the material is good. I hate to use a tired clich - , but content really is king on the web. If your newspaper or magazine has better stuff than your competitors, people will come and read it. We see this every day in the winnowing out of blog market. The better blogs find audiences, the worse ones don't. Sure, there are exceptions, but the rule generally holds. Newspapers and magazines have to learn this lesson and stop futzing with their front pages. They should pay reporters and writers to produce good, original content'in text, audio and video - and the audiences will come. In other words, they should work on breaking stories, not on designing websites.

As for the user-gen components to the USA Today site, they are fine. These things are now a cost of doing business, and are a kind of boon for smart companies looking to improve their content. They should be embraced by everyone providing news on the web. And they will be.

Paul DiPerna:

You launched the wiki-based site called MemoryArchive in August 2005, and I see you have about 875 memoirs so far. That's impressive.

What propelled you to embark on such a project?

Marshall Poe:

Well, I had been investigating new media businesses for the Atlantic, one of which was Wikipedia. The more I looked at it, the more I thought there ought to be some way to use this technology in popular and academic history. I considered various options - a topically specific wiki (say on Russian history), a biographical wiki, a bibliographic wiki, and so on. Since I was working on the project with a class of very bright undergraduates at American University in DC, I threw the question to them and, after a lot of discussion, we all agreed that personal memoirs were the best option.

Memoirs are important historical sources, and people love to tell stories. So we started to work on the site. I divided the class up into teams - software, content, design, publicity, etc - just like you would at a magazine. Despite the fact that none of the students had any experience with these issues, they did a marvelous job and we launched the site in November of 2005. I should add that Erik Miller, currently serving on the Wikimedia Foundation Board of Trustees (the folks who govern Wikipedia), was instrumental in the creation of the site. He did most of the technical work and hosts it for free.

And MemoryArchive has been a success. As you say, we are approaching 1000 memoirs and get good traffic everyday. We've had some nice press as well. The site is growing: Erik is currently setting up non-English language versions, so soon there will be German, French, Russian, and other MemoryArchives. Anyone who'd like to help should feel free to contact me.

Paul DiPerna:

Some logistical questions about MemoryArchive..

How many editors do you have?

How is MemoryArchive funded?

I ask the latter question since there's been a lot of discussion on blogs regarding the financial sustainability of wiki and social network sites. (recently, Wikipedia ran a "capital campaign" to support its services)

Marshall Poe:

Well, there are really only three editors per se, me, Erik, and Natasha Assa (our Russian editor). We have, however, a bunch of excellent "Site Stewards" and are looking for more all the time. The Stewards solicit new content, do clean up, and generally spread the good word about the project. As for funding, MemoryArchive is entirely supported by me and Erik, and we don't plan to change that unless the costs of hosting the site become too burdensome.

Paul DiPerna:

What is Data360?

How did you get involved?

Marshall Poe:

Data360 is a serious data aggregation and presentation site. Its aim is to be the dashboard for a democratic society. Data360 contains serious historical and contemporary data of every kind - on the economy, military, politics, society, culture and so on. It's what you need to know, data-wise, to be an informed citizen. The site also has a big user-generated component insofar as it allows users to contribute serious data to the site and share them with others.

Data360 was started by Tom Paper, a broad-minded consultant in SF who saw the need to quickly and telegraphically present the "big picture" to his clients and, by extension, his fellow citizens. Tom believes, correctly I think, that a central repository for important, serious data will raise what we might call the "collective intelligence" of both business and the nation.

That's the point of Data360 - to make us all better decision makers. I met Tom through a friend. Once I heard his vision for Data360 I was sold and I've been contributing to the project ever since. I plan to continue to do so as the site grows.

Paul DiPerna:

You've told me that "Web 2.0" is a term that you like to avoid when possible.

It seems to me that use of the term "Web 2.0" creates a lot of confusion now, even among Web watchers, and is no longer very useful as an operating concept. I've read a few people pushing for a Web 3.0. (good grief)

Why do you avoid using the term?

Marshall Poe:

I just don't want to become involved in a terminological dispute. People argue endlessly about what Web 2.0 is and isn't, but the core of the concept is captured by an older, unfashionable tag that I prefer the "Read/Write Web. " The web is a medium that allows you to consume and produce content. That's pretty much it, and it distinguishes the web from one-way media like print, radio and TV. "Web 2.0" is opaque.

The "Read/Write Web" is telegraphic you can read it, and you can write to it. That's about as theoretical as I get with these things.

Paul DiPerna:

You are currently writing a new book to be published later this year, or early next year. The working title is "Everyone Knows Everything".

Can you give us a trailer of sorts as to what topics you plan to cover?

Marshall Poe:

Sure. "Everyone Knows Everything" is a history and analysis of mass collaboration on the Internet. By "mass collaboration" I mean Internet projects that allow many people to work together to accomplish something they could not accomplish in the real world, at least very easily. Examples include: finding ancestors (FamilySearch), playing mass games (Worlds of Warcraft), selling stuff (eBay, Craigslist), finding romance (Match), trading files (BitTorrent), making friends (MySpace), exchanging news and opinions (Metafilter), and catalogue knowledge (Wikipedia).

The book begins by describing the physics of what I call "WikiWorld" that is, the virtual space in which all these websites operate. The basic thesis is that we don't really do anything different in WikiWorld, we simply do what we always did more efficiently and at a distance.

"WikiWorld
is the
largest of all..."

In the second chapter I discuss the reasons humans cooperate with non-kin at all (turns out we are very unusual in this way) and lay out the evolution of cooperative institutions over the past 50,000 years. These include the family, tribe, city, state, religion, and science, in that order. I highlight some interesting and little noticed patterns in social evolution, for example, that each successive cooperative institution is demographically larger than its predecessor and in some sense includes its predecessor. WikiWorld is the largest of all (being virtually universal) and includes all that came before it.

The remaining chapters of the book explore how these cooperative institutions have been migrated into Wikiworld, so we have "The Families of WikiWorld," "The Tribes of WikiWorld," "The Cities of WikiWorld," and so on.

Paul DiPerna:

Who is publishing your new book?

Marshall Poe:

Broadway Books, an imprint of Random House. A big shout out to my agent, the magical Bob Mecoy and my editor, the mystical Gerry Howard.

Paul DiPerna:

What is the writing process like for you? Do you have any particular routine?

Or do you approach projects and pieces differently from one to the next?

Marshall Poe:

I begin by looking for the big picture, the basic structure or plot upon which all the parts will hang. In other words, I need to find a thesis and a story. Without a thesis and a story, I can't do anything. Finding them involves nothing more than reading broadly, talking to a lot of people, and thinking, thinking, thinking. I'm also a bit of a contrarian, so I try to find theses and stories that go against the received wisdom. Pretty much all my work has the structure - You thought you knew X, but actually the truth is closer to Y. - The Greeks called this rhetorical play "anagnorisis," the moment of discovery. Now when use this form, you have to be careful to not oversell you story. This is why I shy away from expose as a rhetorical form, because most things that claim to be exposes are actually just ginned up correctives. Anytime you read "And now the TRUTH! " or "This CHANGES everything! " you can be pretty sure you are being sold a false bill of goods. The truth is subtle and nothing ever changes everything. But sometimes things are not as they seem, and that's what I try to get at, usually by putting things in a big frame and telling a compelling story.

"Web 2.0" or the "Read/Write Web" is a good example. We know what it is and it doesn't change everything. In fact, what it is is pretty obvious - that is, a better form of inexpensive interactive communication - and what it does is not new - that is, it lets us do what we already did, just a bit better. What's really interesting to me is that we didn't see it coming. The story of Wikipedia (or eBay, Napster, MySpace, YouTube, etc.) is one of discovery, not invention. In the unfolding of these sites, we found out that the human propensity to collaborate was much greater than we suspected. We had misread human nature.

Now that's a story.

It's not everyday that you discover in a huge natural experiment that your picture of human nature has always been a little off.

Paul DiPerna:

You have a lot of very interesting projects going on right now. Do you have any other big plans this year, whether it's professionally related or for recreation?

Marshall Poe:

Actually, yes. I want to start to make non-fiction videos. Short, high-quality, compelling Internet-distributed videos on topical themes. Think of "An Inconvenient Truth" in 5 minutes or an Atlantic article in video. The basic aim is to flesh out a new non-fiction genre for a new medium, namely digital video, a bit like Montaigne and Bacon invented the "essay" for a new medium, namely print, some 400 years ago.

I'm working with Tom Paper at Data360 on this and talking to people at The Atlantic about it. If anyone who reads this is interested in participating, tell them to give me a call.

March 13, 2007

Comments (2)

home | interviews index | Join the email list | RSS for interviews | Paul's email
Blau Exchange, est. 2006 | Blau Exchange, All Rights Reserved 2006-2008
site design by gralmy