Net Media Perspective
Citizendium and the Writer’s Voice
It’s time to look at Citizendium—the project begun by Larry Sanger in late 2006 that’s a little similar to Wikipedia but also a lot different. I wrote about it in November 2006 (C&I 6:13) and briefly in March 2007 (C&I 7:3). It hasn’t gone away. In some respects the project is doing well.
Unfortunately, given the ferocity with which even the mildest questioning of Wikipedia seems to be met these days, I need to provide disclaimers:
Ø If I say good things about Citizendium, that is not an attack on Wikipedia—unless you believe Wikipedia should be the only wiki-based knowledge base on the web.
Ø If I criticize Citizendium and Wikipedia—and I’m going to, at least in some respects—that does not mean I think either one is useless. It most certainly doesn’t mean I think wikis are useless! My primary source of income these days comes from editing PALINET Leadership Network, which uses the same MediaWiki software as both Wikipedia and Citizendium.
Ø I do not regard “If you’re unhappy, go fix it” as a reasonable response to criticisms of Wikipedia. I always thought that response was a glib way to dismiss criticism and it’s an even more ludicrous response these days. Do I believe the Library 2.0 article is truly “neutral”? I do not. Do I have the energy to attempt to fix it and maintain those fixes in ways that satisfy the Wikicrats? Absolutely not. Nor would I suggest to others that they step into the morass of Wikipedia politics without looking first.
Ø I use Wikipedia. There are many areas where it’s good enough for my purpose, including areas where I see no reason to doubt what’s there. Of course, the more I know about a subject, the less likely I am to take the Wikipedia article at face value (e.g., some months back, the Blu-ray article was so far from neutrality it could almost have been written by Toshiba).
Ø I have no editorial or financial stake in Citizendium. I think it’s an interesting project, one that librarians should pay attention to. I’m not a Citizendium contributor and haven’t made much use of it. I have no horse in this race.
Since most of you probably haven’t paid attention to Citizendium and my early notes were buried near the end of a long Perspective on Wikipedia, I’ll repeat excerpts of those notes (from C&I 6:13).
Larry Sanger helped found Wikipedia. There’s no dispute about that. He’s expressed his unhappiness about Wikipedia’s lack of regard for expertise.
And he’s trying to do something about it. Go to www.citizendium.org; you’ll find the papers discussed below (as opposed to the Many 2 many posts), an FAQ, and if you’re interested that’s the place to watch the project unfold—assuming it does unfold.
Toward a new compendium of knowledge (longer version) is a 12-page essay with elements of manifesto, but mostly hope and design. Sanger’s intrigued by the idea that “Tens of millions of intellectuals can work together, if they so choose” (emphasis in the original). “Whenever I think about this now, I literally quiver with excitement.” He makes an odd prediction:
In the next year, by the end of 2007, every major university, library, museum, archive, professional organization, government, and corporation will be asking themselves with increasing urgency: how, using what systems and methods, can we pool the entire world’s intellectual resources to create the ideal information resource? What worldwide projects and organizations should we join or help to create?
I find that prediction so improbable that—much as I’m intrigued by Sanger’s idea—I can’t take him seriously on that point. Many professional organizations and corporations are essentially incapable of “asking themselves” questions like that, quite apart from governments and other agencies.
Sanger discusses Wikipedia as “an early prototype” of “how [open source hacker] principles should be applied to reference, scholarly, and educational content.” He considers himself a fan of Wikipedia—and wants “to help launch something better, if that’s possible.” He notes a few historical details—including his claim that Nupedia’s history has been told badly. He cites four “serious and endemic problems” with Wikipedia: ineffective and inconsistent rule enforcement, anonymity serving as a troll magnet, insular leadership, and his claim that “this arguably dysfunctional community is extremely off-putting to some of the most potentially valuable contributors, namely, academics.” He finds it likely that Wikipedia “will never escape its amateurism”—indeed, that it’s committed to amateurism. “In an encyclopedia, there’s something wrong with that.”
His solution? Citizendium, a fork of Wikipedia with a messy name that means Citizens’ Compendium. The fork would be “progressive.” It would start by importing all of Wikipedia (legal given the GNU Free Documentation License and also, he thinks, “morally permitted”). Then people—experts, he hopes—will start changing Citizendium articles and adding new ones. When refresh sweeps are done to pick up new and modified Wikipedia articles, such articles will only be picked up if there haven’t been changes in the Citizendium version.
He plans three main changes in the editorial process: Inviting experts to serve as editors, requiring that all contributors use their real names and follow a charter, and reversing some of the “feature creep” in Wikipedia. He offers more details for the proposed editorial system and asserts there will not be top-down bureaucratic structures. Insisting on real-name participation and expecting people to follow a brief charter should help avoid trolls, and there will be “constables” to eject “the project’s inevitable, tiresome trolls” based on a clear set of rules…
Two other differences are interesting. He insists on a “zero tolerance policy” toward copyright and libel abuses—and anticipates “much more courteous treatment” for living subjects, including (maybe) the ability to “request removal of biographies about themselves—if they are not politicians or other prominent public persons—or even to have a crucial editorial role in the articles about themselves.” Are we not all experts about ourselves? Finally: Citizendium will be called an experimental workspace and compendium. It will require a vote of the project’s governing body/bodies to call it an encyclopedia. “It’s a wiki that aspires to be as good as a real encyclopedia.”…
Clay Shirky posted “Larry Sanger, Citizendium, and the problem of expertise” at Many 2 many on September 18, 2006. Shirky’s not much for subtlety: He asserts that Sanger’s opinions are based on three beliefs, then states “All three beliefs are false.” Shirky says experts don’t exist independent of institutions—so much for Albert Einstein, independent scholars, and thousands of others who would generally be considered experts. “You cannot have expertise without institutional overhead.” Later, he says “experts are real,” which seems contradictory. There’s more here, but I was most struck by that odd assertion.
Two days later, Sanger responded (in a guest post, posted in full by Shirky). He accuses Shirky of building a “straw Sanger” by psychologizing about him and showing an “annoying tendency to characterize my assumptions uncharitably and without evidence.” He questions Shirky’s certainty that Citizendium will fail (repeated several times in different ways), “but clearly he badly wants it to fail.” There’s a lot more here (the response is longer than the original post), including a side note that Shirky has his facts about Nupedia wrong. While I find Sanger’s style overwrought at times, in this particular exchange I believe Sanger gets the best of it. Shirky does indeed spout all sorts of certainties and assumptions for which he appears to have no evidence. But I’m hardly an unbiased observer; given Shirky’s overweening insistence that crowds of amateurs are superior to supposed experts (his whole “folksonomy rules!” schtick), I’m surprised to see him claiming to define expertise—except, I suppose, to knock it down.
There’s a second Sanger essay at the Citizendium site, about half the length of the first: “How open collaboration works: an introduction for scholars.” It’s interesting and clarifies that Sanger thinks of scholars—academics—as his prime source of experts. I disagree with Shirky that expertise requires institution, but that assumption may be closer to Sanger’s beliefs based on this paper. Or maybe not: it’s early in the project to read too much into it. The essay explains open source software and why it matters—and how the vision behind open source software can be extended to other forms of collectively owned work. He sees a breakdown between open source software projects (for which, he says, there’s often a small set of “senior developers”) and Wikipedia, which lacks “senior content developers.” “Clearly, the job of applying the OSS model to encyclopedias is unfinished.”
I think it’s time that the editors of the world—meaning academics, scientists, and others whose work essentially involves editing—got involved, not necessarily in Wikipedia, but in similar, suitably altered projects. I want to encourage you scholars, who make it your life’s work to know and teach stuff, to become students of the wonders and beauties of OSS development, and think about how it can be applied to the development of content.
I wonder how many scientists consider themselves editors—or whether this passage indicates Sanger could use an editor. (Yes, I know, I could also use an editor.) That leads into a brief discussion of Citizendium and notes on “promoters of OSS and open content” who say “these projects won’t, or even can’t work.” I won’t quote the stirring paragraph that follows, but here are Sanger’s final two paragraphs:
These well-meaning but wrongheaded promoters of OSS and open content seem to think that open collaboration is a method reserved exclusively to amateurs, students, the “general public,” and so forth.
Let’s prove them wrong.
What’s interesting is how unwilling some folks are to give Sanger that chance…
The only real mention of Citizendium in these pages after that came in another mostly-Wikipedia roundup in March 2007. It’s brief enough to quote in full:
Citizendium hasn’t disappeared; in fact, it’s now opening registration. But there’s been a big change in the project. Based on comments from early contributors, Larry Sanger concluded that forking Wikipedia in its entirety resulted in too much mediocre material, which contributors found tiresome to edit.
So the fork is being “unforked.” All Wikipedia articles in Citizendium that haven’t yet been edited by Citizendium contributors will be (or have been) deleted. That makes the site much smaller but should substantially improve the average quality of entries—if Sanger’s basic theses are correct. It’s far too early to tell whether that’s true.
I believe that decision was crucial. Without it, the relatively few edited articles in Citizendium would probably always be hard to spot in the mass of Wikipedia articles. Sanger’s bold prediction for the end of 2007 was, of course, wrong: There really aren’t millions of experts out there anxious to contribute serious amounts of time to the collaborative project, and we certainly don’t have calls from every major institution to pool all the world’s intellectual resources. Nor, in my opinion, should we.
Let’s look at some developments over the past year at Citizendium—noting that www.citizendium.org will still get you there, but in the U.S. you’ll be redirected to en.citizendium.org, the home page of the wiki. Most notes come from the Citizendium blog (blog.citizendium. org) and Citizendium itself, some from various reactions. I’m using “CZ” as an abbreviation (as Citizendium does), if only to save space and typing.
One oddity strikes me at the home page: Something happened to the “not yet an encyclopedia” disclaimer—unless the governing board took what I’d consider to be a premature vote. The home page doesn’t quite say “this is an encyclopedia,” but it does say, “A wiki encyclopedia project” and “We are creating the world’s most trusted encyclopedia and knowledge base.” Most naïve readers, including me, would say, “Oh, this is an encyclopedia, but it’s not finished yet.” (There is, of course, a “beta” tag.)
About the time CZ officially launched, this essay appeared in a March 21, 2007 post. Extensive excerpts:
How is the Citizendium similar to Wikipedia? In quite a few ways. In enough ways that you might make you wonder why we’ve started another project. Consider:
1. We aim to create a giant free general encyclopedia.
2. We’re managed by a nonprofit.
3. We use MediaWiki software.
4. We use wiki methods of strong collaboration. We don’t sign articles or even have lead authors; we strongly encourage everybody to “be bold” and mix it up.
5. No credentials are needed to participate (as an author).
6. We still rely on “soft security” to a great extent. We mostly trust people and solve what few behavioral problems we’ve seen as they arise.
7. We are committed to a neutral, unbiased presentation of information.
8. We have similar naming conventions, and some other similar conventions.
9. Quite a few of our articles came from Wikipedia.
10. The community and project has been organized by the same person who organized Wikipedia.
Quite similar, it seems. But…
How do we differ? Let us count the ways.
1. We’ve got editors. They are experts in their fields…
2. And we respect them for their expertise. We do not dismiss their expertise as the mere accumulation of meaningless “credentials.”…
3. We have a method for approving articles. While Wikipedia has a “featured article” system, we have expert-approved articles…
4. Our community and contributors are very different.
* We have no vandalism. Excluding the short period in which we permitted self-registration, we have had zero vandalism–none.
* We use our own names and identities. Not only do we require people to sign in, we require them to use names that they attest are their own real names and to fill out a publicly-readable biography…
* We expect professional behavior and have very low tolerance for disruption. Our Constabulary has some pretty firm rules which require professionalism…
* Our Citizens are bound by a social contract…
* We don’t use “userboxes.” User pages are biographies, not vanity pages.
* We don’t use zillions of acronyms… Using a lot of acronyms for every small point of policy creates a sort of in-group cant that makes the community only more insular.
5. Our community managers (called “constables” not “administrators”) are different.
* Our constables are not high school students. They are required to have a bachelor’s degree and to be at least 25 years old.
* Unlike Wikipedia administrators, constables do not make editorial decisions… Constables oversee behavior and adherence to basic policies; editors oversee content.
* Unlike Wikipedia administrators, constables are held to a conflict of interest policy…
6. Policy decisions are increasingly made by representatives, not “consensus.”…
7. The Citizendium editor-in-chief is a limited-term position; he is not “dictator for life.”…
8. To be confirmed: Our license disallows unauthorized commercial use. We are using the Creative Commons Attribution-Noncommercial (CC-by-nc) license for our original articles…
9. Contributors share their copyright with us. Contributors give to the Citizendium Foundation a nonexclusive right to relicense their work…
10. Our article policies differ.
* We are aiming to create introductory narratives, not collections of data. We are encouraging our contributors to create coherent, readable, extended narratives that actually do the job of introducing a topic to people who need an introduction to the topic…
* We use an older version of the neutrality policy… We use the old-fashioned English words “neutral” and “biased.”…
* We take defamation seriously…
* Since we’ve got expert editors on board, we can take a more sensible approach to citing sources. The editors we have on board create the sort of sources that Wikipedia cites…
* We talk about maintainability (or feasibility), not notability…
* We don’t overuse templates…
* We will never have nearly as many articles about porn stars and sexual fetishes. We aim to be family-friendly.
11. We don’t have as many articles. Yet. Give us a little time.
What to say about all this? Several things:
Ø The first similarity and last dissimilarity may be unfortunate. Maybe CZ shouldn’t strive to be giant. Maybe it should strive to be better in a much smaller number of articles, knowing that Wikipedia isn’t going away.
Ø Requiring real names seems entirely sensible. Yes, anonymity and pseudonymity have places in political and other discussions—but I fail to see any justification within an encyclopedic fact-based project.
Ø While I applaud the call for coherent, readable, extended narratives, I believe there’s a conflict between that call and the later decision that articles need five or more authors before they can be signed. I believe the most coherent and readable narratives will always be the product of a single mind and voice, possibly tweaked by good editing. That’s the focus of my grump at the end of this essay.
One link to this post from another blog dealt with the licensing issue—and a comment on that post was interesting, if only because it seems to represent a common and, I believe, unfortunate assumption. Asserting that CZ won’t knock Wikipedia off “its top spot,” the comment concludes: “There can only be one. Who will it be?” Why must there only be one? When did monopoly become not only desirable but also essential? Since its inception, I’ve looked at CZ as a promising project that could complement and possibly strengthen Wikipedia—not as a “Wikipedia-beater.”
A few days later (March 28, 2007), the blog had a post “We ain’t elitist” responding to charges that CZ is elitist because it has an explicit role for experts. Larry Sanger offers a hierarchy of elitism in content-production organizations (excerpted):
1. Very exclusive: the only participants permitted are not just experts, but distinguished experts. Example: Stanford Encyclopedia of Philosophy. One of the finest free reference works available online, just by the way. Would it be appropriate to accuse them of being “elitist”?
2. Exclusive: experts/professionals only. Examples: most big newspapers and magazines; some academic journals.
3. Expert-focused, but semi-expert-welcoming: while experts are most actively recruited, honored, and empowered, the system is also open to people who have a solid but nonexpert understanding of the relevant material…
4. Open, but making a special role for experts: no expertise is needed to participate, but experts are invited to fill a special role in the system. Example: Citizendium.
5. Radically epistemically egalitarian: everyone may participate, and no roles are made for experts; everyone is on an equal footing when it comes to making judgment about what is allegedly good, true, and beautiful…Example…most Web 2.0 projects.
This is a strong and (I believe) true statement: Paying some attention to experts is a far cry from elitism—unless you believe there are no experts. He concludes: “To accuse us of elitism is merely to expose the limitations of your world.”
A March 26, 2007 post links to this ten-page essay. It’s an interesting read, perhaps more so a year later. Already, CZ had enough activity to overcome some doubts. On the other hand, Sanger expected “the Google effect” to make a huge difference once the site went live—and, realistically, moving from 1,100 “CZ Live” articles in March 2007 to 5,500 “CZ Live” articles in March 2008 is, while impressive, not the sort of growth I think Sanger had in mind.
Those are quibbles. Sanger discusses what he regards as a sizable and growing latent demand for something like CZ. He also summarizes ten objections to CZ and attempts to answer each of them. Some points:
Ø Sanger asserts that CZ can become “more useful and more reliable than Wikipedia” while having fewer articles—but still guesses that, in the long run, more people will want to contribute to a CZ-style encyclopedia than to Wikipedia, “just because our system is likely to be more civil and pleasant and actually focused on the work of creating a credible encyclopedia.”
Ø He (reasonably) objects to comparisons between CZ’s very first articles and the best of Wikipedia’s current articles—”You should have seen Wikipedia after its first few months!”
To an objection that professionals either need money or credit that counts toward tenure, he responds that there are clearly motivated professionals—and that it’s quite possible CZ will eventually have only a fraction of its articles “approved” (which requires signoff by an expert).
Ø Is anonymity a reason for Wikipedia’s rapid growth? Sanger doesn’t believe that; neither do I. He says it’s “virality” (word of mouth), an ugly neologism but perhaps the right concept.
Ø There is, of course, the “credentialist” objection, which Sanger calls “uniquely Wikipedian…and little better than wishful thinking.”
Ø Then there are Shirky’s objections and related objections regarding relations between experts and authors—and Sanger basically responds that it seems to be working out. That’s not surprising: Why would experts sign up for CZ if they didn’t want to deal with authors?
Larry Sanger seems to get on Nicholas Carr’s nerves. Sanger wrote an essay in Edge; it’s long and in a Sangerian style I can only tolerate for brief periods. Carr wrote a long and surprisingly nasty rant at Rough type (the current version—he modified it after first posting it—is entitled “Stabbing Polonius” and dated April 26, 2007), doing a good job of sneering at CZ and Sanger himself, throwing in a bit of scatology for good measure.
I don’t think Carr’s rant is worth summarizing in total. It includes a statement that entirely confounds me: “What normal people want from an encyclopedia is not truth but accuracy.” Nor will I summarize Sanger’s April 27 responding rant. I think Sanger’s strongest point is that Carr seems not to care much about what CZ’s actually doing—he’d rather belittle Sanger’s writing.
In May, it was Jimmy Wales’ turn. Assignment Zero, a “pro-am collaboration” between Wired and NewAssignment.net, explored “crowdsourcing” as its first topic, including a discussion of CZ. In that discussion, we get Wales’ assertion that Sanger does not deserve cofounder credit for Wikipedia, credit he had earlier found appropriate. It’s an interesting article with odd notes—e.g., claiming that CZ’s model challenges “online culture at large” and that unapproved CZ articles are “generally inferior to what’s available at Wikipedia,” which may be true but seems irrelevant to the goal of having experts, um, approve the best articles. It is, in essence, a meaningless comparison. That and some other details make the piece come off as, if not a hit piece, certainly slanted. Sanger replied to Wales’ new anti-Sanger comments at more length in the CZ blog (May 11, 2007), labeling some of Wales’ comments libelous.
One interesting piece on CZ appeared within Wikipedia—this article by Mike Johnson, posted July 30, 2007 in the Wikipedia Signpost. (It was an invited piece. Johnson is a “casual Wikipedian” and member of CZ’s executive committee.) Excerpts:
If I had to summarize Citizendium into a sentence, it’d be this: Wikipedia was concerned with making a working online encyclopedia; Citizendium is concerned with making a community that, if it works, will make a really good online encyclopedia…
Why should Wikipedians care about Citizendium?
I’d offer three reasons.
1. The first is the most obvious. We’re an alternative to Wikipedia, and we have a lot of good and interesting things going on. You might consider editing here. We welcome Wikipedians, and a lot of you may appreciate how we do things and what problems we don’t suffer from (e.g., vandalism).
2. The second follows the saying, ‘let a thousand flowers bloom’: it becomes easier to understand and improve your wiki once your sample size rises above [one].
3. The third may not be intuitive, but I think it’s very real: I suggest that Wikipedians should be deeply invested in Citizendium’s success since having a viable competitor is invaluable for the long-term health of any organization. I won’t make the full argument here, but it could be that if 10% of Wikipedians left and joined Citizendium, it’d be better for Wikipedia in the long run. It’s just a thought—but do think about the value of having a strong competitor.
Notes make it clear that Johnson really does admire Wikipedia, that he believes CZ’s process lends itself better to “fleshing out often-disjoint content into a lucid encyclopedia,” and this:
My personal theory is there’ll be a quality differential between Wikipedia and Citizendium depending on the type of content: articles on inherently ambiguous topics, such as history and society, articles on controversial topics, and articles which are introductions to a topic may benefit the most from Citizendium’s collaborative model of explicitly empowering expertise (e.g., those are the sorts of articles I think of as benefiting the most from a “guiding hand” and “lucid expert narration”, or on the flip-side, being hurt the most by edit wars, over-compromising, and cranks).
A December 15 CZ blog post notes that Google’s announcement of Knol—an ad-supported project with signed articles, potentially many competing signed articles on a given topic—has resulted in a lot of press for CZ. On one hand, I’m sympathetic to the Knol concept, for reasons discussed later in this piece. On the other, it’s a little confusing. Knol seems a bit like a searchable collection of signed individual web pages with comments and ratings. Its edge over other web pages would seem to be that Google’s search algorithms could favor Knol articles and, as a result, yield more ad revenue for Google (and those Knol authors who choose to have ads). The Google announcement includes this paragraph:
Once testing is completed, participation in knols will be completely open, and we cannot expect that all of them will be of high quality. Our job in Search Quality will be to rank the knols appropriately when they appear in Google search results. We are quite experienced with ranking web pages, and we feel confident that we will be up to the challenge. We are very excited by the potential to substantially increase the dissemination of knowledge.
Does that mean being part of Knol will increase a page’s ranking? Does it suggest Google’s algorithms actually reward objective quality as opposed to link popularity? I can’t tell.
When Sanger was interviewed about Knol, he identified three problems:
First, quality. It looks to me as if Knol is a high-level attempt to do what many others have done. Countless websites already exist that invite signed essays and information (remember h2g2.com?) and other content for public rating. Time will tell, but Knol will probably resemble other such websites, and have a huge amount of mediocre content, with a little excellent content mixed in. The concept does not sound like a model that would attract many genuine experts. I say that because the notion that anyone may write a “knol” and be compared and ranked by “the crowd”—not by expert peers—is apt to attract relatively little notice from experts who are very careful about where they publish…
Second, lack of buy-in from the free culture crowd. Many of the sort of people who contribute knowledge to projects like Wikipedia and the Citizendium are likely to be very skeptical of a giant corporation organizing such a project, particularly with Google Ads appearing on the articles. It does not appear to be in the spirit of the free culture movement. Still, it is good that Google has decided to make ads optional.
Third, lack of collaboration… The quality and depth of encyclopedia articles written collaboratively by a huge global community, especially under expert guidance, will eventually beat out anything produced by individuals, regardless of their ability.
I’m less convinced regarding the third point. There’s a huge difference between collaborative creation of a multi-article whole and collaboration on each article. I’ve seen no convincing evidence that collaborative articles are inherently better than those written by single authors, and there’s reason to believe that, in certain respects, they’re likely to be worse. That’s a basic tenet on which I fundamentally disagree with Sanger, Wales and, I guess, the whole “wisdom of crowds” crowd. That doesn’t automatically make me wrong; nor does it make me likely to be right.
CZ announced a landmark on January 22, 2008: Five thousand “live” articles. That’s not much compared to Wikipedia’s two million plus—and it might make sense for Sanger to say less about the potential for CZ to wind up with more articles than Wikipedia, a future that may be both unlikely and (perhaps) unattractive. Another measure is interesting: CZ articles as of early February totaled more than five million words. Sanger feels obliged to say, “More than Wikipedia did in its first year,” but that’s irrelevant. Five million words of good, edited, trustworthy copy is a lot. That’s fifty typical 300-page nonfiction books. It’s almost three times as much as the entire run of Cites & Insights.
More important is that CZ’s growth is accelerating. If that is accompanied by high quality, CZ can become a serious force before too long. “Millions of articles” in a few years? Maybe, maybe not. The equivalent of a quality printed encyclopedia? Quite possibly.
A two thousand word essay within CZ, “Why Citizendium?” lays out the case for CZ as it stands. It’s an odd combination of attack on Wikipedia (and specifically Jimmy Wales) and manifesto for a brighter day—and as longtime readers know, I’m rarely fond of manifestos.
“What is the point of the Citizendium,” you might ask, “when Wikipedia is so huge and of reasonably good quality? Is there really a need for it?”
To put it forcefully: there is a better way for humanity to come together to make an encyclopedia. So we make this appeal to you. If we can do better than Wikipedia—or more positively, if we can pioneer a truly effective way to gather knowledge—then shouldn’t we?...
[T]o make our case, we don’t have to say that Wikipedia is broken. While different Citizens have different views about Wikipedia’s merits, we agree on one thing: we, humanity, can do better. But why think that the Citizendium, in particular, can do better?
Why think the Citizendium can “catch up”?
[Notes on growth and acceleration].. Even if we merely continue to triple our rate of growth every year, we will have millions of articles ourselves after some more years.
In other words, we look to the long term—just as Wikipedia’s founders did in its first years. And the long-term outlook is positive indeed. In five to ten years, we can expect similar growth, similar numbers of active contributors, and a similar traffic ranking. So we need not worry that Wikipedia will “always be larger.”
We can do better
We do not think that Wikipedia is “good enough.” We think humanity can do better: Wikipedia is full of serious problems. Many of the articles are written amateurishly. Too often they are mere disconnected grab-bags of factoids, not made coherent by any sort of narrative. In some fields and some topics, there are groups who “squat” on articles and insist on making them reflect their own specific biases. There is no credible mechanism to approve versions of articles. Vandalism, once a minor annoyance, has become a major headache—made possible because the community allows anonymous contribution. Many experts have been driven away because know-nothings insist on ruining their articles. Wikipedia co-founder Jimmy Wales acts as a law unto himself, not subject to a written constitution, with no official position, but wielding considerable authority in the community. Wales and other Wikipedia leaders have either been directly involved in, or have not adequately responded to, a whole string of very public scandals… The people with the most influence in the community are the ones who have the most time on their hands—not necessarily the most knowledgeable—and who manipulate Wikipedia’s eminently gameable system.
But even if you disagree with much of this indictment, you might still agree that we can do better.
Real names are better
By requiring real names, we give both our articles and our community a kind of real-world credibility that Wikipedia’s articles and community lack: if you look at our recent changes page, you will see nothing but real names…. The Citizendium has virtually no vandalism and very little abuse of any kind.
To this, you might say that real names also exclude too many people, so that the Citizendium will grow too slowly. But this is puzzling to say, considering that many thousands of people have signed up to the Citizendium under their own real names. A community that asks its members to use their real names is more pleasant, polite, and productive than one that allows abusive people to disrupt the community under the cloak of anonymity...
A modest role for experts is better
We too permit very open contribution; the general public make[s] up the bulk of our contributors, as “authors.” We agree that broad-based contribution is necessary to achieve critical mass as well as the broadest spectrum of interests and knowledge.
But we believe that it is merely good sense to make a special role for experts within the system. A project devoted to knowledge ought to give special inducements to people who make it their life’s work to know things. We believe—and we think our work so far bears this out—that a project gently guided by experts will in time be more credible, and of higher quality, than a project making no special role for experts. So we allow our expert editors to approve articles (creating stable versions, with a “draft” version that can be easily edited). Editors may also take the lead, when necessary, in articulating sensible, well-informed solutions to content disputes—disputes that sometimes go on interminably on Wikipedia…
The potential of the Citizendium is stunning
…the most important reason to get behind the Citizendium is not a comparative point at all: it is that a fully-developed Citizendium would be stunning. Not only would it have millions of articles, but it would have, at least, hundreds of thousands of expert-approved articles, all available for free, all being instantly updatable with the latest research and events, and all wonderfully well written. Imagine enormous quantities of content combined with the highest quality and exhaustiveness of scope, all achievable only by radical collaboration. Imagine, as well, a whole raft of supplementary reference materials.
The world has never seen anything like this. But we can create it. Our best chance to do so is by throwing our support behind the Citizendium.
Some personal motivations to support the Citizendium
But what about you—why should you get involved?
It’s mainly because it is fun and rewarding to share your knowledge with the world. Your contributions to the Citizendium are less likely to be degraded by poor edits later on: others will move your contributions forward, not backward. In time (we can’t say when—but eventually), the article you contribute to will be approved by an expert editor, and so represented to the world as containing a credible, reliable introduction to your topic. And all for free. We are accomplishing something truly worthwhile.
Many people, especially academics, are concerned that in a strongly collaborative project like this, they cannot get the individual credit they need. Well, you can already point people to the article history, where your real name will appear, crediting you with the specific edits you make. Also, we will soon probably start a pilot project that will allow people to be credited with their contributions on a “byline,” under certain circumstances…
Furthermore, academics and other experts can submit what we call “Signed Articles,” presenting their own personal, but hopefully objective take on an aspect of an article already in the Citizendium. We add “Signed Articles” to a “subpage” of the main article—one of many different types of subpages a main article has…
Fun, rewarding, and worthwhile—what more could you want?
As of this writing, there are more than 6,100 live articles in CZ. “Live” means either that it’s original to CZ or that it’s copied (typically from Wikipedia) and there have been “at least three significant changes in three different places.” That 6,100 includes 778 developed articles (articles that are complete or nearly so), 56 approved articles (it’s a relatively slow process), and a bunch in various other categories.
As wikis go, the statistics aren’t impressive—but they may be misleading. The current pageview total is some 1.135 million over some 11,000 content pages. There are slightly more than 7,200 registered users. There’s definitely significant activity: more than fifty new pages were added within the last two days (as I write this), and the most recent 50 changes go back less than an hour (in a midafternoon check).
These are early days. I have no idea whether it’s conceivable or likely that Citizendium could overcome Wikipedia’s role as the quick-and-dirty place to check things out. I’m not sure that’s even a reasonable goal. Could Citizendium become known as a better place to go for well-written articles on a smaller number of topics—a better encyclopedia? Possibly, given time and energy.
I am unclear as to why so many seem determined not to let that happen—why so many commentators seem to want CZ destroyed in its early stages. I have heard nobody—nobody—suggest people should stop using Wikipedia and start using Citizendium instead, as things stand now.
I don’t remember. Was there horror when Google appeared by all those who knew that Alta Vista was all we needed—good enough, and we already knew it? Did columnists write attack pieces saying Google would never be of any use? If so, I must have missed them. What’s different now?
Wikipedia is admirable in many ways and flawed in some. Citizendium seems to be off to a good start and has admirable goals. But it shares one flaw with Wikipedia that causes me to hope these projects don’t totally undermine traditional encyclopedias—or, worse, lead people to believe these resources are always “good enough” to understand a subject.
That flaw is the belief that the “group mind” is always better than any expert’s mind—that collaborative writing is always better than a single writer’s voice. The extent of that belief at Citizendium shows in the proposal to allow signed articles: An article can’t be signed unless at least five people have contributed to it. Once you have five people, there’s a good chance an article will be well on its way to the disembodied group prose that characterizes most Wikipedia articles. I think that’s a shame.
Here’s what I said in March 2007:
When Jimmy Wales says college students shouldn’t cite Wikipedia in research papers because they shouldn’t cite any encyclopedia, I agree. When Jimmy Wales says, “One aspect of Jaron Lanier’s criticism had to do with the passionate, unique, individual voice he prefers, rather than this sort of bland, royal-we voice of Wikipedia. To that I’d say ‘yes, we plead guilty quite happily.’ We’re an encyclopedia,” I disagree. Lanier struck me as calling for voice—not necessarily “passionate” but coherent, turning sets of facts into stories. There is nothing about an encyclopedia that precludes coherent, well-written entries representing single voices with personality; groupthink and bland speech are not prerequisites for encyclopedia entries. (Remember the “scholars’ edition” of the Britannica?)
I looked up an article in a traditional encyclopedia (albeit one in DVD form): Encarta 2007. The article, “Pre-Columbian art and architecture,” is long, segmented, and interesting; it’s written in a clear voice that tells a story. It’s also signed, in this case by Robert J. Loscher of the Art Institute of Chicago, an expert in the field. So are many articles in many encyclopedias. Wales’ defense is simply nonsense.
One frequently cited issue, the uncertainty as to whether stuff in Wikipedia has any basis in fact, is to some extent being dealt with as articles show ever more footnotes. Unfortunately, that process seems to have two negative side effects: It makes the articles harder to read (when there are superscript numbers every sentence or two), and it may be making articles even less coherent and “voiced.”
I think those issues deserve more discussion.
I decided to read one of the Approved articles at Citizendium and compare it—first with the equivalent at Wikipedia and then with the equivalent at Encarta 2007. The example does not prove my assertion, because the example turns out to be primarily one scholar’s essay at Citizendium, and a good one at that.
The example was Andrew Carnegie. CZ’s article is roughly 7,000 words long—more than a typical encyclopedia would devote to most biographies, even those of major industrialists and philanthropists. It reads well, with good narrative flow and coherent organization. I felt as though I knew something of the man when I finished. That’s not surprising. Fundamentally, this is Richard Jensen’s article, with some edits and contributions from others. The article has voice—and deserves a byline, in my opinion.
Wikipedia’s article is longer (about 8,000 words) but it’s a mess. Sentences and paragraphs are choppy. The organization isn’t coherent. While it’s not flooded with footnotes, that turns out to be a problem: The article had its “Good Article” status removed because it doesn’t meet Wikipedia’s cookie-cutter approach to verifiability: There are dozens, maybe hundreds of facts, and there’s not a number next to each one. The discussion page includes notes that the article is a mess—and given that Carnegie is a controversial figure, it’s almost bound to be a bit difficult.
Neither article paints Carnegie as a hero, but Wikipedia barely paints him at all: Lots of incidents, no real sense of who he was.
The Encarta article is, well, terse: A biographical sketch that’s much less satisfactory than either online choice. It’s also an unsigned article, clearly not one of Encarta’s prizes.
The anonymous collaborative approach used by Wikipedia has its virtues. Those virtues do not include clear narrative flow or coherent narrative voice. The whole methodology pushes against the kind of voice that turns facts into stories, stories that communicate meaning.
I firmly believe that individuals are the best storytellers, certainly aided by good editors. Maybe not always, but as a rule. I also believe that the best way to communicate nonfiction—concepts, not just facts—is through stories.
Could any collaborative effort produce science writing with the clarity and coherence of Isaac Asimov’s nonfiction works? I doubt it. Would Churchill’s great works have been better vetted by large anonymous teams of interested parties? I’m certain not.
As an editor—now as a livelihood, and for nine years with the LITA Newsletter—I’ve always believed my role was to clarify and revise as needed while retaining as much of the writer’s voice as possible. I believe I succeeded back then and am succeeding now. Yes, there have been cases where I’ve thought, “I could write that better”—but that’s not my job.
Experts aren’t always effective writers. Effective writers frequently aren’t experts at anything except writing. Collaboration between expert and writer frequently works brilliantly—but that’s not the same as group writing.
I’ve seen the claim that group efforts always improve writing. I don’t buy it. I’d hate to see groupthink viewed as the goal of nonfiction writing in general. I’d hate to see the neutered (not neutral) prose of Wikipedia become the norm. If I’m just looking up a quick fact, maybe—but that suggests that Wikipedia articles should be a few hundred words long, or maybe just a paragraph. If I want to understand something, I’d like to hear a voice.
Cites & Insights is sponsored by YBP Library Services, http://www.ybp.com.
Opinions herein may not represent those of PALINET or YBP Library Services.
Comments should be sent to email@example.com. Cites & Insights: Crawford at Large is copyright © 2008 by Walt Crawford: Some rights reserved.
All original material in this work is licensed under the Creative Commons Attribution-NonCommercial License. To view a copy of this license, visit http://creativecommons. org/licenses/by-nc/1.0 or send a letter to Creative Commons, 559 Nathan Abbott Way, Stanford, California 94305, USA.