Meine Merkliste Geteilte Merkliste PDF oder EPUB erstellen

Next Wikipedia, take a right | Open Source |

Open Source Was ist Open Source? Markus Beckedahl Sandra Braman Vera Franz Rishab Aiyer Ghosh The Big World of the Many Small Volker Grassmuck Claudio Prado Into the Great Wide Open FOSS, die Firma und der Markt Open Source und Entwicklung Atul Chitnis Ronaldo Lemos Fernanda Weiden "Microsoft loves piracy" Interview Larry Sanger Buch Freie Software Glossar Literatur Links Redaktion

Next Wikipedia, take a right Wikipedia, Citizendium, and the politics of knowledge: An interview with Larry Sanger

Larry Sanger

/ 12 Minuten zu lesen

Sometimes, it's better to ask the experts, says Larry Sanger. The co-founder of Wikipedia wants to start a fork of the free online encyclopedia Wikipedia – with expert control. In his interview, he explains the why's and how's.

Lawrence M. Sanger (© Larry Sanger)

Google, MySpace, YouTube: There is a band of brothers behind many success stories of the net, and as with many great stories, fraternity often ends in fraternal strife. In the case of the free online encyclopedia Wikipedia, Cain and Abel bear the names of Jimmy Wales and Larry Sanger. Today, Jimmy Wales is Chairman Emeritus of the WikiMedia Foundation, which operates Wikipedia. Until 2002, Larry Sanger was editor-in-chief of Wikipedia, and is one of its most outspoken critics today.

In 2000, Sanger, then a Ph.D. student in philosophy, was taken onboard the "Nupedia" project by Jimmy Wales. Nupedia started 1999 and was financed by a Californian dotcom enterprise named Bomis. Together with Wales, Sanger developed the basic concepts of this first free online encyclopedia, set up a board of scientists which developed a seven step review process in the following months, and recruited interested contributors – in the end, more than 2.000, which later became the cadre of the Wikipedia community.

Mainly because of the review process, Nupedia dragged along – at the end of 2000, the project had used up 250.000 US-$, and about 20 articles online. Wales and Sanger looked for alternatives and decided to start a radically simplified version parallel to Nupedia, based on the then-infant "wiki" technology.

In January, 2001, Wikipedia went online and literally exploded. At the same time, Nupedia silted up silently. In the end of 2001, Nupedia had about 25 articles online (150 more were in different stages of the review process), whereas Wikipedia already counted 18.000. Sanger left his official position at Wikipedia and Nupedia in 2002, since Bomis couldn't continue to pay him.

The rest is history. Nowadays, Wikipedia amounts to over six million articles in 250 languages, of which 500.000 are found in the German version. About 27.000 active "Wikipedians" world wide contribute regularly to the different language versions. Under the roof of the WikiMedia foundation, Wikipedia has grown numerous offsprings; using the wiki technology, they create free dictionaries (Wiktionary), books (WikiBooks), collections of quotations (WikiQuotes), news sites (Wikinews), multimedia collections (WikiCommons), and a university (Wikiversity). The separate project "Campaigns Wikia" wants to help social and political movements to organize themselves via wikis. And even an own parody is nursed with equal diligence and care: The Externer Link: Uncyclopedia now counts over 20.000 articles.

The enormous success of Wikipedia did not remain without critique. While universities and schools complain that students increasingly replace their own mental efforts with copy-and-paste from Wikipedia, traditional encyclopedias (such as the Encyclopaedia Britannica or the German Brockhaus) saw themselves threatened with extinction, and stressed the lacking quality and reliability of Wikipedia.

But then, "Nature", maybe the most renowned science journal of all, published the results of a Externer Link: study in December 2005, according to which the accuracy of scientific articles in Wikipedia came close to that of the Encyclopaedia Britannica. Britannica (in a Externer Link: long counterstatement) and Nature (in an official response and its Externer Link: editorial) exchanged a series of blows, without resolution.

Sanger himself joind the choir of criticism in December, 2004, with an entry at the populat technology website Externer Link: "Kuro5hin". Already, he stated that if Wikipedia could not "jettison its anti-elitism", that is, give due authority to designated experts in editing articles in their field of expertise, a Wikipedia "fork" would become necessary. He repeated this idea in Externer Link: several Externer Link: other online articles, and finally put it into practice: In Sepetmber, 2006, at the Externer Link: "Wizards of OS 4" in Berlin, he announced the start of the Externer Link: "Citizendium".

Mr. Sanger, here on "Wizards of OS 4" you have announced the launch of the project "Citizendium". What is it?

The "Citizendium" is "The Citizen's Compendium". In time, it will hopefully become worthy of the name "encyclopedia". To begin with, we'll be calling it an "experimental workspace", and it will be a fork of the Wikipedia, that means, a copy of all articles in Wikipedia, that allows to edit them in a new wiki.

As people make edits to the articles of the Citizendium, those articles are not refreshed from the Wikipedia, whereas the other articles, which are yet untouched, will be regularly refreshed with the newest versions from Wikipedia. So it is a progressive fork, a gradual fork in that way.

You are the co-founder of Wikipedia, which is seen as one of the great success stories of the "Web 2.0" and collaborative peer production on the internet. What is your reason to start a fork?

On Externer Link:, I have an essay that lists a number of points – I could just read the bullet points: First, the community does not enforce its own rules effectively or consistently. Secondly, widespread anonymity attracts people who merely want to cause trouble – in other words, the troll problem. Thirdly, many complain that the leaders of the Wikipedia community have become insular, wherefore it is increasingly difficult for people who are not already part of the community to get fully on board. And lastly, this arguably dysfunctional community is extremely off-putting to some of the potentially most valuable contributors, namely, academics.

What will be the main differences to Wikipedia?

The most important differences between my proposed project – because right now, the wiki doesn't even exist yet – and Wikipedia are: First, there will be a new role in the system, that of "editors", who will be credentialed experts who have a few rights in the system that ordinary authors do not have. These experts will still be asked to work right alongside the other authors, not to control their work in a top-down fashion. But what they will be able to do is to make decisions about content disputes and then articulate those decisions on a "discussion page". They will also be able to designate articles as approved or not. All of this is going to be done in the same radically uncontrolled, bottom-up way that Wikipedia itself has done.

For example, editors will designate themselves as "editors", there will be no committee required to approve someone as an editor. Instead, if someone has the required credentials, they will simply go to the website to their user page, state them, link to evidence online and then declare themselves to be an editor, for which there is not technical support needed outside the software that already runs Wikipedia. The change in that regard is a social change, not a technical one. It really doesn't change the core operation of the Wiki – what makes Wikipedia work – at all.

The second big difference is the adoption of a community charta, and the explicit expectation that people who are involved in the community support the charta. They become "good citizens" in a new online polity. So there will be a fair and open process whereby project rules will be enforced. There shouldn't be too many rules, because the proliferation of rules is confusing to people and they start ignoring rules. And in general, people work best without any rules. But what rules there are, need to be taken seriously.

One of the problems that I had with Wikipedia is that it's rules in many cases aren't taken seriously. When I say this, I know that a lot of people are going to think that I am for "order" and "control" and "rules" and so forth – that isn't the case. I am the co-founder of Wikipedia, think about that. I am for a lightly controlled chaos, to a certain extent. That's what makes Wikipedia and open collaboration work.

The reason why I insist that people take the accepted rules of a project seriously is that these rules define the project. If you want to engage in the project of building a neutral encyclopaedia, for instance, then you have to take neutrality seriously. Otherwise, you are not engaging in the project you agreed to help out in the first place.

What are the criteria to become an "editor"?

The specific criteria have yet to be determined. Perhaps, one way to generate a list of criteria is to ask: What sort of minimal requirements does a hiring committee for a university or research institute have for getting someone on a tenure track job? That would be for academic-type fields.

But because there are subjects in an encyclopedia whose main experts are not academics, obviously, the criteria for those kinds of fields would be different. It might involve publishing a certain number of works of a certain type, giving presentations, organizing workshops, things like that. It will be something that is objective, something people will be able to look at evidence for.

The reason to keep these criteria objective is that we don't want this process politicized, because it becomes politicized so quickly and easily. I know many people who say that this is just "credentialism". Well, if you want to invite people to serve as experts in a system, and you want to avoid those essentially political disputes involved in hiring and firing experts, the only way forward is to rely on the same sort of real-world credentials that hiring committees rely on and just get rid of the committees.

But isn't the setting-up of criteria for experts already a political decision? Don't you already enter the political by deciding in advance who is allowed to be an editor?

If someone were to say that, I would ask them what they mean. In other words, if you were to make the claim that having a Ph.D. isn't required to be an editor in the field of philosophy, I would ask: Why do you think that so? The burden of proof is on those who want to maintain that society's means of credentialing experts in all kinds of different fields – not just the Ph.D. – is political.

But still: Where to draw the line? Should you have a PhD? Should you have tenureship? Isn't that decision something political?

Not in the relevant sense. It isn't political in the sense that the process of deciding who is an editor and who's not does not depend on the acceptance of the hiring committee's particular ideologies or theoretical approaches and so forth. The field of possible views will be as wide as is consistent with the having of those credentials, which is very wide indeed in the scientific and academic communities. Even if there are huge majorities in outlook, there is an enormous long tail in diverse outlooks in many fields, and we are going to be open to that diversity. That's the point.

What about defining the standards for areas of knowledge where there is not yet a set of formalized credentials in the academic system?

It will be very interesting to see how that will work out, and obviously, that's something where we will need input from people in the relevant fields. Engineering is a good example. Engineering professors may not be the best people to make judgements about readable encyclopaedia articles on technical topics.

Then there are all kinds of gaming and hobby topics where there isn't anyone who writes for professional journals, because there are no professional journals. Hardly anyone studies these subjects in universities, and the most knowledgeable people might be some who don't have any kind of academic credentials at all.

To make this process relatively objective and least receptable to politicization, it will be important to come up with some other kind of credentials, now using this term in a very, very loose sense. To give another example: Some of the leading experts in Irish traditional music have never written books or academic articles about the subject, they don't have any degrees in music, and yet they are really the sources of our knowledge about this music.

Will the definition of these criteria be an open process?

Yes, that's absolutely correct. And I would encourage people to realize that in general, that is going to be the case with the Citizendium. The whole idea is that the development of policies is going to be open. But we are going to settle on some very definite policies.

And who is going to settle these policies?

My idea now – and I am very much open to debate –, is that there are going to be some public mailing lists – actually, there are already half a dozen mailing lists –, and anyone who is interested can contribute to the discussion on those mailing lists. Those discussion will be crucial to determine policies. I take it as my job as editor-in-chief of Citizendium to articulate the consensus.

In addition to that, we will set up an advisory board of some of the leading thinkers on relevant topics – open source software development, open access publishing –, and some subject area experts who are technically inclined. After quite a bit of debate and hopefully, some face-to-face meetings, we will have a constitutional convention. We will actually sit down, after many issues have been debated, put on finishing touches, and ratify a community charter.

This is something which I think has never been done in quite this way. But that's the way the meatspace world works, and that is considered only just and right in meatspace. So why shouldn't it be the just and right way online?

That would be the major question of constitutional experts to your project: More important than the rules themselves is the process that determines how new rules will be set up and who will decide in the end. So will this process of formulating a constitution itself be democratic and open, as the UN has tried with the so-called "multi-stakeholder approach" at the World Summit for the Information Society, for instance?

Generally speaking, I totally agree with that openness, and as anyone who is familiar with the history of Wikipedia knows, I followed this advice and expectation – perhaps to my fault – in the development of Wikipedia. But the process has to get started somehow, right? It's important to begin with some definite concept around which many people can rally, so that a self-selecting community can form itself. You can't just say: "I am going to do a fork of Wikipedia", and then let everyone who is attracted to that single idea try to hash out as to what the fork should look like – that's simply to vague.

That's why I have articulated a number of other policies that essentially define the project. Then I ask the people who like this set, or most of this set of policies: "Come help me, taking this as a starting point, work out something that the largest number of people who generally like this proposal can agree to." There has to be some articulation of the parameters of the project in advance, or you will never get a community started in the first place.

What should drive people to change from Wikipedia to the Citizendium, or to join the Citizendium while they didn't join Wikipedia?

Again, I would refer to the main differences between the projects, because those would explain the motivations for people to join one project rather than the other. I think that most people would prefer to work side-by-side with begnin, non-controlling experts on articles about topics that they care about. Everything get's a little better if you have a senior on your side, gently giving advice about the process. That is, I think, much more attractive to everyone than just working in a free-for-all, where people who have made it their life purpose to study a subject are treated the same as everyone else, even when they are writing about that subject.

Another reason that people might be attracted to the Citizendium is that the Wikipedia community has become very difficult for some to work in. It requires a certain kind of person to really go head to head with the administrators and regular contributors to Wikipedia, who can be quite difficult in various ways.

The hope is that if we require people to log in with their real names and agree to a common charter – a social contract –, and if there are people who are respected as enforcers of this agreement, the result is going to be a socially more pleasant place to work. That's the hope. And obviously, we need to work together on actually defining the parameters of what will make this community pleasant and productive.

I don't think that it is pleasant at all to have administrators breathing down your neck all the time. That is not what I propose. In fact, it seems to me that that happens on Wikipedia quite a bit. The "constables" – I am explicitly proposing a new name for that role – will be instructed to act as hands-off as possible. But once they get involved, they have the authority to act swiftly.

Interview: Sebastian Deterding

Weitere Inhalte


Nächste Wikipedia rechts abbiegen

Man sollte doch jemand fragen, der sich auskennt, meint Larry Sanger. Der Wikipedia-Mitgründer hat einen Ableger der freien Enzyklopädie gestartet – mit Expertenkontrolle. Warum und wie,…



Der Name 'Wikipedia' ist ein Kunstwort, das sich aus dem Verweis auf die Wiki-Technologie und aus dem englischen Wort für Enzyklopädie 'encyclopaedia' zusammensetzt.



Auf den ersten Blick wirkt die Statistik wie ein Beleg für den riesigen Erfolg des Projektes Wikipedia. In allen Sprachversionen steigt die Zahl der Artikel von Jahr zu Jahr. Allein die…



Das Wikipedia-Prinzip „Keine Theoriefindung“ setzt auf die Sammlung, Darstellung und Aufarbeitung von bereits gesichertem Wissen. Damit steht es im Kontrast zur Wissenschaft, die die Generierung…


Wenn sich Wissen im Plural trifft

Möglichst viel vom Wissen der Welt soll in nur einer Datenbank zusammenfließen. Was aber wenn indigenes Wissen grundsätzlich anders funktioniert. Wenn Wikipedia in seiner Ordnung des Wissens sogar…


Konsens in zwei Sprachversionen

Unter den über 270 Sprachversionen sind die englische und die deutschsprachige Wikipedia die größten Angebote. Auf den ersten Blick erscheinen sie nahezu identisch, vertreten sie doch die gleiche…

Lawrence M. Sanger was born 1968 and holds a Ph.D. in philosophy. He was co-founder of Wikipedia and editor-in-chief of its predecessor Nupedia. Up to the start of his Wikipedia fork "Citizendium", he worked for the environmental "Encyclopedia of Earth", which is part of the larger online encyclopedia "Digital Universe".