From printed to wikified encyclopedias

Citation:
Geser Hans: From printed to "wikified" encyclopedias. Sociological aspects of an incipient revolution
In: Sociology in Switzerland: Towards Cybersociety and Vireal Social Relations. Online Publikationen.
Zuerich, June 2007 (Release 2.5). http://socio.ch/intcom/t_hgeser16.htm

The success of the Wikipedia is spectacular insofar as it contradicts most current common sense assumptions as well as almost all conventional theories about human motivation and social organization.
Even the founders – who originally aimed at a conventional elitist project (Nupedia) - were completely surprised by the processes of incessant growth and expansion they have inadvertently kicked off in January 2001 when the English WP was started.

Ironically, the most central premise of WP (that the "swarm intelligence" [1] constituted by all contributors together surpasses any individual wisdom) is regularly compromised by the very inaccurate collective predictions users make about the further development of their own project. For example, when English WP users were invited to guess when the 1 millionth article will be posted (actually at March 1 2006), more than 220 of them overestimated the time needed considerably, while only 46 were about right (or somewhat below the actual time) .[2]

From a sociological point of view, there are three major reasons for studying the Wikipedia: because illustrates so profoundly how modern human society and culture is to be reshaped and transformed by the Internet and its still unexplored future potentials of global digital communication.

1) On the macrosociological and macrocultural level, studying the WP provides insights how the shift from the printing era to the digital age goes along with revolutionary new ways to produce, organize, distribute human knowledge, and how such transformations affect the worldwide interrelationships between different institutions, collectivities, nations and cultural regions. In particular, we see that milleniaaged patterns of monocentric eliteguided cognitive systems give way to more open, dynamic and polycentric knowledge cultures, and that knowledge may become more independent from money and power as well as from the sphere of formalized education and academic credentials. Thus, the question arises how these changes affect the complexity, scope and dynamic adaptation of human knowledge, its relationship to political, economic and academic-scientific spheres, its coherence and acceptance thru different cultures and demographic strata, and its characteristics on substantive and epistemological levels.

2) In a mesosociological perspective, the WP (like Linux) contradicts the established (Weberian) wisdom that complex cooperative performances and products can only be realized in centralized and formalized settings of bureaucratic organization. Instead, we see that some of the most complex of these productions can evidently also take place in informal, decentralized open source networks continuously activated by unpaid volunteers who deliver their contributions according to their own preferences and judgments; without overall blueprint planning, formal role assignments and hierarchical controls.

After six years of unimpeded evolution, the Wikipedia has grown into a complex organization from which many insights about the preconditions, functional prerequisites, consequences and limits of open source network communities can be gained

3) On the microsociological plane, finally, the Wikipedia begins to change the basic ways how individuals (or small groups) search, select, retrieve and apply knowledge on their workplace, in school, as medical patients or voting citizens or in any kind of private situation. Particularly impressive are the new potentials to access relevant knowledge without cost and delay under almost any circumstances and role conditions (even while on the move); and the capacity to enact self-guided learning processes by navigating through hypertext structures in a personalized fashion. In addition, we see former passive "readers" to be transformed into hybrid "users" who switch flexibly between receptive and contributive roles, and to use encyclopedias in a prosaic instrumental fashion - not to be compared with the intimidating status display effects emanating from by thirty-something exclusive leather backed tomes on polished Mahogany shelves.

Unquestionably, the breathtaking complexity and dynamics of the WP makes it extremely difficult to reach a sound, scientifically founded judgment of the project s a whole - particularly on its characteristics. Thus, there may never be comparative study that include more than some few out of the 250 WP's worldwide - simply because no research team is acquainted with so many languages. And many studies on the accuracy or comprehensiveness of the reported information may be curtailed because any findings become quickly obsolete as a result of incessantly ongoing updating and modifications.

As a consequence, the following passages have the more modest aim to carve out some insights on a more general level (by comparing printed and wikified encyclopedias), and to illustrate the arguments with anecdotal evidence from only three major Wikipedias: the English, German and French.

2. The new "asymmetric competition" between open source networks and conventional bureaucratic organizations

When the Internet became popular some ten years ago, many pundits predicted that like all the preceding conventional media (press, radio and TV), it would soon become commercialized and dominated by professional groupings and large bureaucratic organizations. However, the subsequent developments provide little support for such upbeat economic expectations.

On the one hand, the "Dot.com Crisis" of 99/00 has illustrated that many business models imported from the era of top-down (or: one-to-many) communication were illsuited in the new Net environment where everybody had the same technical means for creating, transforming, storing, copying and transmitting information. And secondly, the decentralized and interactive features of the new medium have recently come to dominate in unprecedented spectacular ways: particularly in the rising prominence and significance of and usercreated content.

First of all, it is striking that the biggest and most successful players on the Web are those that rely on "bionic software" (You Mon Tsang) [3] by aggregating and analyzing the information generated constantly by millions of users. [4] By exploiting the "Long Tail" (Anderson 2004) of small and irregular users, they gain more knowledge and produce more useful services than conventional enterprises that typically focus on a much smaller number of "essential clients". [5]

On a most elementary level, Bit Torrent exemplifies this basic principle of Web 2.0: those services improve with increasing number of users, because everybody contributes his own computing and storage capacities of his personal computer:

Similarly, Google and similar search engines base their algorithms of page ranking on the surf activities of users and on the hyperlinks set by all webpage producers (Barnett 2005); Amazon derives its attractivity from methods of collaborative filtering from which users get recommendations about what they shall buy next.; and in the case of "del.icio.us" [6], thousands of users coproduce a search engine based on the public exchange of bookmarks and the "tagging" of visited sites.

In all of these cases, the Web as a platform gives rise to new manifestations of "emergence": in the sense that qualitatively new "molar" products arise out of the combination of a very high number of (sometimes extremely tiny) "molecular" contributions.

Secondly, we can observe the rise of the “Blogosphere” as a new non-commercial and non-professional arena of interactive public discourse: challenging the traditional monopoly of the monological “mainstream media” to steer the public agenda setting and to shape public opinion.

And thirdly, we are most fascinated by the rise of peer-to-peer networks that successfully compete with big corporations (or interorganizational systems) in generating goods and services of the highest complexity mankind has ever produced. Thus, P2P file sharing networks are easily capable of substituting the conventional music industry in distributing songs on a worldwide basis; by pooling their excess computational capacities, 4.5 million PC users are able to constitute the most powerful supercomputer on earth (SETI@home) for searching signs of extraterrestrial civilizations; thousands of networked software developers are able to compete with Microsoft’s in producing GNU/Linux, an operating system comparable to Windows; and innumerable of unauthorized collaborators pool their knowledge to create the Wikipedia: an encyclopedia that matches or even surpasses the Encyclopaedia Britannica or the German “Brockhaus” in at least some crucial ways.

As seen most succinctly in the rivalry between Microsoft and Linux [7], the emergence of open source communities has given rise to an "asymmetric competition between social organizations that produce very similar products, but with completely different (even antagonistic) cooperative structures:

Such “open content communities” (Reagle 2004) or networks of “commonsbased peer production” (Benkler 2006) are characterized by at least twelve common characteristics that set them in a sharp contrast to conventional bureaucratic organizations:

Of course, these twelve features are functionally interrelated. For instance, the lack of payment is caused by the fact that products cannot be commercialized, and it has the consequence that no selective recruitment practices, hierarchical controls and rigid working duties can be implemented.

While we all know that voluntary activities are quite competitive with formalized organizations in many modest small productions (like cutting hair or cooking a meal), we are astonished to realize that nowadays, they seem to challenge bureaucracies also in the realm of the most complex products of goal-directed human cooperation: e. g. computer operating systems and encyclopedic works.

It is no doubt that the Wikipedia aspires to be an “encylopedia” in the precise sense of this highly traditional term, because in his article “What the Wikipedia is not” [8], founder Jimmy Wales takes great care to deny explicitly that it is something else: a dictionary, a news feed, a collection of essays, an instruction manual, a repository of links, a directory or (horribile dictu) a vehicle for propaganda, self-promotion and advertising. Of course, this insistent explicitness is highly necessary because thousands of contributors all over the planet have to be precisely instructed so that they are enabled to behave conformingly and to recognize and correct deviant entries.

It is advisable to have a short look on the glorious tradition within which the Wikipedia insists to place itself – despite the cogent insight that by passing from paper to the Internet, literally everything is different than it was ever before.
In its extensive article on the concept, the Encyclopaedia Britannica defines encyclopedias as “summaries of extant scholarship in forms comprehensible to their readers.”

As such they have existed since 2000 years in very different size and format and as products of highly variable (individual and collective) modes of compilation. The term “Enkyklios Paideia” originally implies that all human knowledge can be represented in a closed circle and mirrors an ordered cosmos that can be explicated in a consensual and definitive way because human knowledge is thought to basically stable and the world to which it refers not subjected to fundamental change.

Typically, encyclopedias are representatives of cultural epochs that (aim to) contain its most authoritative and respected knowledge in condensed form. As a consequence, the most comprehensive editions also tend to be written in the language in which most contemporary knowledge is produced (Latin in the middle ages and English today).

Encyclopedias typically address themselves to “interested educated laymen” who are consulting them in matters where they are not experts, but about which they are able and motivated to gain reliable basic knowledge. As a consequence, they have to strike the right mean between a high-level educated language and a simple, widely readable style. This also implies that the best contributors of articles are often not the most eminent scholars (because they are often too much absorbed by current research), but many second-rank experts (e. g. teachers, writers or officials) who have professional reasons to acquaint themselves intimately with a particular topic or who have accumulated their knowledge in the course of their occupational experience or institutional career. They induce such contributors to do their best to clad their specialized elitist knowledge into a more popular form, so that rather broad strata of “interested readers” are able to understand the texts without consulting auxiliary sources (like dictionaries etc.).

In many conventional encyclopedias, there was no guarantee that each article stems from an expert in the corresponding field. In Diderots Encyclopédie, for instance, about three quarters of all articles are said to have been provided by a single collaborator (Chevalier Louis de Jaucourt).

Until the Renaissance, most encyclopedias didn’t address to a general public, but to specific elite socialized within a specific circles or formal formalized settings (e. g. clerics). The printing press then has given rise to a much less circumscribed, anonymous public: consisting of expanding bourgeois strata, academics and “intellectuals” who have acquired knowledge by selfeducation or on other informal ways. In the course of modernization, encyclopedias have changed due to the rapid expansion and fragmentation of existing knowledge on the one hand and the growing divergences of different knowledge spheres on the other.

Thus, earlier cosmological architectures have given way to neutral alphabetical orderings of articles, because societal consensus about ontologies and the priority of different knowledge spheres has evaporated. And tight connections to educational systems and powerful cultural elites have been loosened, because knowledge became increasingly distributed broadly among various population segments (even rather marginal and politically dissident groups). In addition, ambitions of authoritative knowledge codification were abandoned; many more recent encyclopedias can be rather understood as a reaction to rapidly increasing flows of new publications: by satisfying the need for shorter digests that provide easily accessible information and orientation (Yeo 2001).

A major step in this evolution was the appearance of Diderots “Encyclopédie” (between 1751 and 1772) which was promoted and realized by pre-Revolutionary intellectuals who maintained a critical distance toward classical authors, and even an overt hostility toward the reigning political and religious regime (Munzel 2003).

Objectivity was particularly cultivated in highly “eclectic” epochs characterized by highly pluralistic cultural elites; e.g. in the encyclopedias of the old Roman Empire like Pliny’s “historia naturalis”; and in the 19th century). In the 19th century, cultural pluralism gave rise to highly objectivistic compilations (like the German Brockhaus) devoid of all ambitions for synthesizing knowledge or transmitting it in a pedagogical fashion.

While these traditions persist until the present, the 20th century gave rise to huge governmentally sponsored encyclopedias whose mission was to reflect the knowledge culture of a specific nation. However, even highly authoritarian and totalitarian Regimes (e. g. the Soviet Union or Italy under Mussolini) have produced encyclopedias in which most subjects are treated in a rather open-minded, non-ideological way. By compiling existing knowledge from a multitude of sources, encyclopedias seem to be disposed to affirm the autonomy of objectivistic cognitive orientations vis-à-vis the restraining influence of powerful societal actors, reigning ideological fashions and established cultural institutions.

In the following, it shall be demonstrated how the Wikipedia fits into this encompassing history of human endeavours to articulate and transmit the essential canon of knowledge of a specific epoch or culture. On the one hand, it is easy to show that the migration to an interactive online hypertext format is a necessary step in order to realize encyclopedias adapted to the conditions of contemporary societies, because traditional paper editions are not able to keep pace with the amazing manifold, complexity and dynamic change of science and other current cognitive cultures. If the Internet did not yet exist, it would have to be immediately invented in order to secure the continuity of encyclopedic ambitions. On the other hand, it is also evident that by trying to realize old ideas with new technologies, something radically new is emerging for which we don’t yet have adequate terms and definitions.

Within the short history of the Internet, the Wikipedia stands out as one of the most successful non-commercial Web projects at least in quantitative terms: Since its inception in January 2001, it has forked out to more 250 languages comprising more than 5 million articles. [9]

The English version alone has grown to about tenfold the size of the Encyclopaedia Britannica and is currently adding about 50 000 new articles each month. [10] In Sept. 2006, comScore has reported that with 154 000 000 unique visitors per month, Wikipedia.org was occupying rank twelve worldwide (behind giants like Google, Microsoft, Ebay and Amazon). [11] With 56 millions monthly visitors, the English Wikipedia occupies rank 10 of all English websites, while the German version maintains even rank 6 among all German-speaking sites (with 8.9 million visits in June 2006) [12] . The site’s traffic is heavily boosted by Google which places Wikipedia entries regularly at the top of search result lists.

Concerning criteria of quality, judgments are highly controversial and difficult to verify objectively, because the system is so big and volatile that nobody is able to cognize and evaluate it as a whole. While fundamental criticism abounds in many publications [13], websites and discussion fora, one of the leading researchers of open source networks, Yochai Benkler, concludes intuitively that ”most of the commercial, proprietary online encyclopedias are not better than Wikipedia along any clearly observable dimension. (Benkler 2006: 168).

By comparing 42 articles of the WP with 42 analogous entries in the Encyclopedia Britannica, the prestigious magazine "Nature" found that the average number of errors was 2.92 in the EB and 3.86 in the _WP - thus concluding that the level of correctness is rather similar in both publications.

More intransparent are the differences in softer spheres of human knowledge (e. g. in the social sciences, arts and history), because inadequacies can less easily (and less consensually) be assessed.

These spectacular achievements contrast with the very modest costs involved in building up the system and in running the whole enterprise. Founder Jimmy Wales is managing his enterprise with only four full time collaborators and a small yearly budget of about 1.5 million Dollars (provided mainly by small donations between 50 and 100 Dollars bestowed to the “Wikimedia”, the supporting mother corporation).

With such very small financial investments, the WP has reached the most supreme status that can be reached in the Internet:

A rather valid indicator for this growing reference status is the rising frequency of the phrase "according to Wikipedia" on a rising number of WebPages. In May 2005, this expression got already 22 000 hits on Google, in January 2007, this number has already risen to 740 000 in January 2007 [14].In addition, there are very likely innumerable copyings of Wikipedia material without indication of the true source. Such hidings are very common because WP citations are still very much discouraged, especially in academic settings.

Thus, the Wikipedia provides vigorous evidence that some highly optimistic expectations about human online behaviour may under certain conditions be true – despite the fact that they collide fundamentally with traditional common sense assumptions and established theoretical views:

A major shortcoming of the “dead wood era” was that for mere physical reasons, writing on paper does not lend itself to higher levels of collective cooperation. Just because it is cumbersome (or plainly impossible) to circulate paper sheets so that everybody can add his contributions, the notion of individualized authorship is reinforced. Such high "transaction costs" are a major reason why the production and application of knowledge is still characterized by a rather low division of labour. (Teece 1988; Ciffolilli 2003). In fact, most intellectual work has remained on a primitive “handicraft” level: contrasting increasingly with manual labour which got ever more collectivized in the course of industrialization.

Thus, not only "monographies", but even many "encyclopedic" works tend to be mainly anthologies of articles written by individual authors. Cooperation is largely restricted to the intervention of an editor (or editing committee) that may modify or shorten the article of give it back for revision. Collective cooperation is mostly confined to the “molar level”: interrelating articles, streamlining their formats and fitting them into predefined blue print structures,

In very early encyclopedias, even this higher level cooperation was largely absent because they were produced by a single editor (like Cassiodorus, Honorius Inclusus or Vincent of Beauvais) who acted mainly as anthologists: by just selecting and aggregating existing texts. Even many eighteenthcentury encyclopedias were the products of single compilers, such as Chambers' Cyclopedia of 1728 and the first edition of the Encyclopaedia Britannica, issued between 1768 and 1771. By the very end of the century, however, the task of "compiler" (who collects given texts) had metamorphosed into that of "editor" (who commands, directs, selects and modifies incoming contributions).

By contrast, digital media in general and the computer networks in particular provide many alternatives for more sophisticated forms of cooperation: ranging between completely open collaborations where everybody can participate (Wikipedia) to closed circles which restrict access by various means of digital control. For the first time in history, collaboration on the very micro level is also technically supported: by software tools of “collaborative writing” that enable groups of any size to work collectively on the same article or sentence and to influence even the most tiny details of spelling, grammar and punctuation.

On the most general level, it can be said that the Internet lends itself equally to two basically divergent modes of collective verbal communication:

On the one hand, it supports discursive communication [16] by enabling users to express themselves personally, like in an oral discussion. Such exchanges result in “threads” consisting of all the posted messages filed in the order they have come in during time, without any mechanisms available to synthesize or systematize what has been written [15]. As seen in newsgroups, web fora, chats, blogs and other forms of online discourse, threaded online communication makes the achievement of consensus usually more difficult than face-to-face discussions because even in very large groups, everybody can articulate his personal opinion at any time, without referring to what has been communicated by others. Consequently, online discussion groups are more functional when an increase in complexity is sought (e. g. brain storming) than when the reduction of complexity is the goal (as in decision-directed deliberations) (Kerr/Hiltz 1982: 99f.; Gallupe and McKeen1990; Geser 2002). In addition, the diachronic structure makes it very hard for readers to harvest the yield of the discussion: especially in the case of newcomers who need much time to sift through all the accumulated materials.

On the other hand, online communication supports as well synthetic document mode communications , where the individual contributions become bricks or mortar of a larger collective production: regardless of the time of posting and the identity of its originators.

“Wikis” (invented by Ward Cunnigham 1995) can roughly be compared to naked concrete walls that can be painted by everybody:

However, while paints on walls tend to stick irreversibly (or to leave traces when removed), entries in wikis can easily be erased by everybody who does not agree. Thus, an evolution of memes is started where the "fittest for survival" are those with which most participants do not disagree.

However, technical reversibility is not enough: there has to be an incessant intensive activity from the part of users to correct any abusive entries within the shortest possible time. In the case of conventional printed encyclopedias, every maintenance activity can cease at them moment they are delivered and distributed. Wikipedias, however, remain only functional as long as very large number of editing users remain watchful and active. Otherwise, vandalizations –even if produced by very tiny user fractions – would remain uncorrected, so that the whole system would be continuously degraded.

While thread communication boosts subjective selfexpression and individualization, wikis support processes of supraindividual community-building and objectification.

While the thread mode is functional for facilitating communication processes, the document mode gives priority to their results: like in he case oft most conventional written texts whose final form provides no information about the antecedent processes that have led to their creation.

Thread mode communication is based on the “Heraclitean” (or Hegelian) premise that true knowledge emerges within a dialectic discourse between diverging communicators and remains open to dynamic change. Document mode productions rely on the “Platonic” assumption that true knowledge takes the supraindividual form of objectified “ideas” or “theories” whose eternal truth can finally be ascertained beyond all interpersonal disputes.

In contrast to the “dialectical” blogosphere, Wikipedian philosophy relies on the premise that true knowledge is produced in a continuous cumulative process of aggregating and synthesizing information, not in a process of dialectical discourse.

Consistent with this epistemology, Wikipedia participants are advised to in “constructive cooperation rather than adversarial strifes:

In some cases, threadmode productions are subsequently transformed into documents in order to systematize and simply the information and to ease its diffusion to additional participants: e. g. in the case of FAQ pages which inform newcomers shortly about the goals, values and norms that have been elaborated in the preceding discussions. In a similar fashion, the Wikipedia combines the two modes by paralleling each article page with a discussion page where dissensual aspects concerning the articles (e. g. conflictual views about scope or terminology) can be fought out. But the relationship is highly asymmetric, because the discussions are just an auxiliary tool for improving the quality of the article, while the article is not symmetrically seen as an input for fuelling the discussion.

In a functional perspective, Wikis can be considered the informational analogue to the assembly line in the industrial era. Like the latter, they provide the technological basis for aggregating an infinite number of modest individual performances into a highly complex end product that stands out as an object completely dissociated from all its individual co-producers. [18]

In no other sphere of text production, the shift from individual to collective authorship has been so fundamental than in the Wikipedia, where typical articles may well be the product of several hundred edits executed by a many dozens of collaborators.

The success of the Wikipedia even depends highly on this “fine grained modularity”, because extensive participation can only be generated when even users with very modest skills, very little time and rather low work motivations see the opportunity to make valuable contributions (Benkler 2006: 100). This also has the effect that articles are mostly “endogenous creations” shaped by the cumulative influences of the different collaborators, so that exogenous dependencies (on earlier encyclopedias) are less pronounced than in the printing age where new encyclopedias were often very much influenced by their historical predecessors. [19]

While Wikis support convergent collectivist cooperation, they nevertheless do not provide intrinsic mechanisms for synthesis and systematization.

In printed encyclopedias where each entry is usually constructed by a single author, longer articles usually have a highly structured, coherent architecture (e. g. by progressing from more general to more specific aspects). In the Wikipedia, however, articles are usually the product of many independent contributions of piecemeal parts, because nobody is given the responsibility to take care of the article as a whole. Thus IBM researchers have found that most collaborators simply add or cancel specific words, sentences or passages, while very few reorder paragraphs or reorganize the article as a whole Viégas et. al. (2004).

Like many other texts that are digitally created these days, many Wikipedia articles thus tend toward a low level of overall structuring and integration, because they are the product of a lose sequence of copy and past procedures spanning over a wide period of time.

In many cases, the structure of most articles becomes irreversibly fixed at the time of their creation (or soon afterwards), so that the synthetic capacities (or incapacities) of their originators get a decisive weight.

Analysed under various different perspectives, the WP shows a consistently accelerating pattern of growth.
Growth rates were particularly spectacular in 2006 where the total number of active Wikipedians as well as the number of edits, articles, words, images and internal linkings (by all WP's worldwide) has more than doubled between Oct. 2005 and Oct 2006.

In the following, it is demonstrated that the WP unfolds in a sixdimensional space: all dimensions contributing to its quantitative size and ubiquity on the one hand and its qualitative significance on the other.

Since its inception in early 2001, the Wikipedia is a global project rapidly expanding to all major languages, ethnicities and geographic regions. In Dec 2006, the statistics page on “multilingual ranking” lists currently active Wikipedias in not less than 250 (!) languages: among them in dead idioms like Sanskrit and Latin as well as in almost all subnational languages of Europe that have little or no tradition of writing (like Alemannic, Ladino, Piedmontese, Sorbian and Greenlandic. [20]

However, only 176 of these had more than 100 articles, 110 more than 1000, 52 more than 10000 and 12 more than 100000 entries. [21]

As seen from Figure 1, all the larger and middle-sized WP's drawing on more than 200 contributors were founded before the end of 2003, while the smaller versions covering minority languages have been steeply multiplying up to the present time.

While these versions diverge extremely in terms of size and growth (in accordance with the population carrying them), this rapid diffusion and ubiquity is most astonishing, because the question arises what makes the acceptance and active support of WP's is so independent of any specific cultural and societal conditions.

In the course of 2006, the WP has achieved at least a rank among the 30 most frequented websites in most regions of the world. It enjoys a particularly high status in all German-speaking countries (Germany, Switzerland and Austria), where it occupies rank five or six. On the other hand, its popularity is least pronounced in some Eastern Asiatic countries: because of blockages (in the case of China) or strong competition by similar domestic sites (Table 1). [22]

Table 1: Position of Wikipedia in the rank order of all websites visited by the country’s population (based on page views on Dec 15th 2006) [23]

Figure 2 shows that the total number of active collaborators has developed rather moderately in the first four years, and then has multiplied about eightfold (from 38 000 to more than 300 000 from Oct 2004 to the present (Dec. 2006). Evidently, the English WP has been far better able to keep pace with these Worldwide developments than the two European WP’s that have experienced a much smoother (while still perfectly continuous) expansion.

Of course, part of this rise is attributable to the fact that the curve is accumulative: thus including many earlier users who have since discontinued their collaboration.

In fact, only about 25-30% of all these contributors belong the current “labor force” (= individuals who have made at least 6 edits in the current month). Interestingly, this percentage has remained rather stable during the last three years: except in the German case where the percentage of actives has continuously dwindled.

Looking at the expansion of the active user base, it becomes even more evident, that the whole global WP system as well as the English WP is experiencing exponential growth, while the German and French Wikipedia are characterized by a much more moderate expansion (Figure 3).

As the growth data on the Zachte statistics are available for each month, it may be asked whether the evolutionary pattern can be adequately modelled by a not to complicated mathematical equation, from which short-range or even middle-range future predictions may be derived. Given that new contributors are constantly arriving while older ones are leaving, the cumulative historical number of active Wikipedians is likely to rise without limits. Therefore, a polynomial or exponential equation seems more fitting than a logistic curve that approaches an unsurpassable highest value.

By trying out different formulas, it is found that, the rising curve of total collaborators within the last two years can almost perfectly be fitted by a quadratic polynomial (in the German case) or a cubic equation (in the three other cases) (Table 2).

Table 2: Modelling the growth in the total number of contributors with cubic polynomial equations (covering the monthly figures from Oct 2004-Oct 2006).

As to be expected, the core of highly active individuals (with more than 100 edits per month) is much smaller and less subject to expansion. Unsurprisingly, its relative size was most prominent at the incipient stages of the project, and it seems to decline constantly in the course of ongoing expansion (Figure 4). Interestingly, the nucleus of activist Wikipedians is significantly larger in France than in Germany or in the Anglo-Saxon countries.

While the number of worldwide (hyper)activists has not kept pace with the broader base of participants, it has nevertheless about doubled each single year: enlarging considerably the pool from which administrators, arbitrators, bureaucrats and other incumbents of with supervisory and integrative duties can be drawn.

Nevertheless, best curve fittings for the absolute rise on highly active collaborators are achieved when logistic (instead of polynomial) statistical equations are used in which curve approach a maximum values that is not much higher than the present figures (Table 3). Thus, the equations predict that this "ruling elite" of administrators and other activists will only rise very modestly in the coming years: at most by ca. 20 percent.

Table 3: Modelling the growth in the number of highly active contributors (who posted more than 100 edits in the last month) with a logistic equation (covering all monthly figures from 2001 to Oct 2006)

The multilingual proliferation of Wikipedias leads primarily to an expansion on the level of articles, because each collectivity contributes its own particular localities, personalities, cultural productions and historical events. Compared to the skyrocketing trend on the global level, the growth of every single Wikipedias (even in English language) is rather modest (Figure 5).

For checking whether expansion occurs in a linear or in an exponential fashion, it is analyzed how expansion rates change over time. As seen from Figure 6, the creation rates of new articles have increased very much on the world level, while the rate of the other encyclopedias have flattened out: indicating an almost linear pattern of growth.

Very similar to the growth of contributors, the recent increase in articles (2004-2006) also follows a curve that can be adequately modelled by a cubic polynomial equation, in which one of the coefficients (of the first or second degree) is on the zero level (Table 4).

Table 4: Modelling the growth in the total number of articles with cubic polynomial equations (covering the monthly figures from Oct 2004-Oct 2006)

	b0	b1	b2	b3	F-value	R-Square
All WP's	-835287.6	-2230.642	0	+18.994	3148.063	.997
English WP	385332.8	-13371.1	0	+6.054	598.606	.984
German WP	-105175.6	0	+117.683	+1.032	30410.097	1.000
French WP	-164830.6	0	63.438	+695	7082.767	.999

While the global WP system expands mainly by steep rises in the number of articles, the English, German and French versions give more weight to an increasing elaboration of their entries: by submitting them to many edits and enlarging their textual size. The German WP in particular seems to compensate its rather moderate additions of articles by considerable efforts in their internal elaboration: so that the average number of words per entry has increased fourfold (!) between 2002 and 2006 (Figure 7).

No unexpectedly, the German and English WP also excel in the number of edits per article (which seems to have gained much momentum recently in the Anglo-Saxon sphere) (Figure 8).

Figure 9 shows that updating frequencies have generally increased since Oct. 2003: with the exception of the most recent time interval where Germany which has experienced a decline.

While many edits are directed at enlargements, others serve the purpose of correcting errors, eliminating vandalisms or substituting obsolescent by updated information. They are typically made by assiduous "wikiclerks" more dedicated to formal than substantive aspects of encyclopedic work.

Empirically, such aspects of "diligence" can be grasped by relating the number of edits not to the number of articles, but to the volume of words

As seen from Table 10, the intensity of such "maintenance" activities have evidently decreased in the German WP, while they increased sharply (between 03 and 05) in the English version where nowadays, more than six edits (instead of four in the other cases) are on the average occurring monthly for every thousand words.

Considering all these findings. we may conclude that the expanding field of active participants (as seen in Figures 2,3,4) constitutes the basis for more speedy developments on the level of diversification, elaboration and diligence. This is particularly true for the English Wikipedia which currently exceeds the French and German sister WP’s by a fourfold creation rate of new articles (ca 60 000 per month) and by an almost twofold number of additional edits per article (about 2.7 per month).

A fifth evolutionary dimension concerns the degree of internal integration which can be roughly operationalized as the number of interlinkages between the different pages.

As to be expected: there is a monotonic increase in the number of such hyperlinks in all Wikipedias, but despite the exponential increase in collaborators and edits, these increments seem to diminish recently, particularly in 2006. More than in other aspects, pronounced cultural divergences stand out here: with the German and English WP in the forefront, while the French WP lags considerably behind (Figure 11).

Given that the potential number of interlinkages increases exponentially with the rising number of articles, this decelerating growth implies that relative degree of connectedness between the articles is on a sharp decline. [24]

Finally, there is a sixth dimension of growth that refers to the external embedment of the WP within the WWW. This external integration" is also continuously increasing, but (in comparison with the internal interlinking) in a rather modest way.

Until the end of 2002, all Wikipedias have evidently followed an "isolationist" strategy by restricting the number of hyperlinks to other websites to almost a zero level. Since then, the mean number of such links has edged somewhat above one in the global averages as well as in the French and German WP, while approaching two in the Anglo-Saxon edition Figure 12).

These results prove convincingly that the Wikipedia clings to the printing age insofar as it tends to define itself still as a rather selfcontained system offering the whole of human knowledge, rather than as a node within a Net in which for every single entry, a lot of other equivalent (and in some cases much richer) knowledge sources may be found.

6. On the potentials and limits of wikibased open source encyclopedias: some preliminary conclusions after six years of experience

Conventional encyclopedias usually base recruitment on previously achieved status characteristics that are interpreted as valid indicators of expertise: e. g. by inviting only personalities with professoral or doctoral degrees. In other cases, recruitment is expanded to individuals enjoying an informally achieved public reputation (e. g. intellectuals) or with status positions in non-educational institutions (e.g. high ranking politicians or successful entrepreneurs). Of course, such recruitment patterns help to keep the resulting knowledge canon within the limits of institutionally established elitist culture. There was always a strong bias against the inclusion of “indigenous” knowledge originating in folk cultures or esoteric circles. Thus, magic and astrology had no place in European encyclopedias since the 12th century.

In modern societies, such ex ante criteria are of limited value for various reasons

1) high scholarly reputation and status achievement is not primarily based on the possession of existing, but on the production of new knowledge. Thus, highly innovative researchers may have rather limited knowledge about the broader structure and historical development of their specialized topics.

2) Educational status characteristics are always based on past achievements; so that they may not be consistent with present qualifications (e.g. in cases where a scholar is no longer up to date because of illness or advanced age).

3) There may be important spheres of knowledge which are not in the reach of any formally educated scholars, because their acquisition occurs mainly by selfeducation or by accumulating practical experience. This is particularly true for most practical knowledge (e. g. used in the production of goods and services) that has been so prominent in Diderots Encyclopaedia. And it is even more true for any spheres of “subcultural” or “indigenous” knowledge controlled and transmitted informally within special segments of the population: e. g. knowledge about heavy metal music bands or computer games, about sectarian religious belief systems or anthroposophic medical treatments.

4) The knowledge about many subjects has become so complex and multifaceted that it is distributed among many individuals with different specialities and experiences. Thus, no single person will be capable of producing comprehensive articles on “London” “Goethe” or “Renaissance” that treat all important aspects of the topic on the same scholarly level.

In short: widening the pool of potential collaborators, evaluators and correctors is indispensable when knowledge is too complex to be controlled by any “expertocratic elite”. For instance, systems of “peer review” fail when it is not possible to select ex ante the people most capable of evaluating a specific contribution – e g. because the contribution refers to something so new or so specialized that no “reputable experts” exist.

By starting without any advance knowledge about who possesses what kind of knowledge about what topic, the Wikipedia is compatible with a highly complex, intransparent society where new areas of previously inexistent knowledge areas (e. g. about new technological or cultural phenomena) are constantly arising, and where an unknown manifold of knowledgeable individuals have to be taken into account.

According to an internal WP page, an avid interest in Wikipedia has been known to...

The very lack of recruitment procedures has the consequence that the composition of Wikipedia collaborators is rather homogeneous, because the general “digital divides” various population segments are not only reproduced, but even amplified:

By giving everybody the capacity to edit and change existing pages, all Wikis are evidently based on the premise that
1) the majority of participators makes constructive rather than destructive contributions;
2) contributors believe that this premise is true: because otherwise, their motivation would be destroyed.

On the motivational level, this implies that most participators are driven by positive intentions, not by drives of intentional lying or vandalism.

In the case of Wikipedia, this premise is stated in the principle of “good faith”:

On the skill level, this presupposes that any information provided is more likely to stem from knowledgeable than from uninformed individuals, and that more informed individuals have better chances to have their contributions accepted. This again implies that most people have a realistic selfassessment about their own knowledge – so that they accept corrections when they are originating from a more competent side.

In fact, 51% of benevolent and competent participators would be sufficient to set a cumulative process in motion which would lead to a gradual overall improvement of the Wiki because at least in the longer run, “good content” is more likely to be posted and to be maintained, while “bad content” is more likely to be corrected or weeded out. Of course, such low percentages would result in very slow improvement processes: with the consequence that at each point of time, many errors would be uncorrected, and many lessvisited hoax sites or vandalised articles would not be eliminated. Consequently, the assumption to produce highly reliable knowledge bases approaching those of professional encyclopedias is based on the assumption that the percentages of competent and benevolent contributors (as well as the correlations between knowledge level and influential participation) are rather high.

The Wikipedia certainly contributes to a levelling between experts and laymen as knowledge providers, because the names, status positions and qualifications of contributors are not visibly marked. No assertion is accepted as “true” only because it stems from a Wise Old Man who is highly respected because many of his statements have proven to be true in the past.

As knowledge is so much dissociated from personal communicators, it has to be evaluated on the basis of its intrinsic merits, the empirical sources on which it relies as well as its consistency with other facts or theoretical constructions.

"Wikipedia's articles on the British peerage system - clearheaded explanations of dukes, viscounts, and other titles of nobility - are largely the work of a user known as Lord Emsworth. A few of Emsworth's pieces on kings and queens of England have been honored as Wikipedia's Featured Article of the Day. It turns out that Lord Emsworth claims to be a 16-year-old living in South Brunswick, New Jersey. On Wikipedia, nobody has to know you're a sophomore." (Pink 2005).

Since the Renaissance, Western societies have focussed very much on culture as an arena of individual productions (text, pictures, music compositions etc.) neatly attributable to single authors. More than that, every new work should stand out from others by “originality”: showing a singularity of features not realized anywhere in the past and not repeatable in the future. Of course, strategies of individual attribution may be functional for boosting individual motivations: a major reason why they are also widespread in the academic sciences: serving as a driving force for individual careers and reputation

As this "romantic individualism" is inimical to all forms of labour division, it has also undermined for centuries all encyclopedic endeavours, because such projects have to be based on a collectivism that doesn't leave much space for individual selfaggrandizement.

Thus, the question “why does anybody participate” was legitimate during the whole history of encyclopedias, because collaboration in such projects was seldom an attractive way to gain individual rewards. Typically, the articles delivered were not paid and did not contribute much to personal reputation, because the name of authors remained concealed (or were indicated only by initials, like in the newer editions of the Encylcopedia Britannica).

Long before the Wikipedia, therefore, coauthors have been primarily stimulated by other motives than by boosting their personal reputation: e. g. by the extrinsic satisfaction to belong to a specially selected, extremely prestigious scholarly elite, or by the intrinsic satisfaction to co-define the “official” canon of knowledge of a given epoch or society.

For motivational psychology, therefore, the Wikipedia does not pose radically new problems, because by studying any kind of voluntary behavior, the social science have always done wise to assume that “homo sociologicus” (in contrast to (homo oeconomicus) is driven by a multitude of different motivations:

Compared to traditional paper publication projects, the Wikipedia has enhanced capacities to harvest and aggregate such diverse motivational resources, and to channel them efficiently into constructive cooperative endeavours.

1) Many collaborators of moderate and low levels of motivations may be won just because the thresholds to participation and the costs of collaboration are so low. Thus, everybody with a hooked up PC can log in and edit pages anytime at any place on the planet earth. Because no membership role with formalized duties has to be adopted, participants remain free to decide on the modes, ways and intensities of present and future collaborations; and contributions can be so “fine-grained” (e. g. by just adding a single figure or correcting spelling mistakes) that almost anybody can feel self-confident enough to add at least some modest improvements.

Again, it is not unjustified to compare the Wikitechnology with the assembly line: both share the principle of breaking down production processes into small independent modules, so that lower skill levels are sufficient to make valuable contributions:

This also explains why some sister projects like Workbooks have not taken off: because the minimal threshold that had to be reached (e. g. to contribute useful textbook chapters) has been much too high (Benkler 2006: 101).

2) The “law of big numbers” teaches that to the degree you increase the number of participants, the more likely it becomes that among them, you will also have individuals with a very wide variety of (also quite rare) characteristics, skills and motivations.

Thus, very few individuals may find deep satisfaction in correcting the spelling and punctuation of other peoples texts, but in a population of 800 million Internet users, they may still run into thousands; and whoever succeeds in mobilizing them can generate a volume of voluntary work worth millions of Dollars if it would have to be bought on the labor market. Evidently, the success of the Wikipedia is based on such effective filterings. For instance, hundreds of “police constables” are patrolling for overseeing and correcting various cases of abusive behavior; a silent brigade of “janitor-minded” individuals is constantly active to clean up after vandalizations; talented “mediators” feel urged to intervene in order to moderate heated edit wars; and fussy “clerks” with a bureaucratic mentality are highly useful to correct even very tiny errors in biographies or statistical tables.

3) Given its constantly rising status as a primary reference site and its spectacular impact on global knowledge culture, the Wikipedia is attractive for anybody who draws satisfaction from being part of such a big and influential project – even if his contribution is minor and insecure. Thus, a collaborator confesses that “knowing that an article will instantly become a published part of a worldwide reference is an intoxicating enticement.” (Wilson 2006); and another compares himself with somebody working on the most sacred texts of human society:

Some contributors face for the first time the opportunity to make proselytes by displaying the knowledge they have privately accumulated in their hobby activities to a wider public. Thus, a Madison-based software engineer named Sean Lamb has derives personal satisfaction from contributing articles about American railroad history: a very specialized topic not likely to be treated by many others. (Patrick 2005).

On the other hand, however, collaborators may feel demotivated by the perspective that their efforts may be in vain because their contributions are deeply modified or even eliminated completely by subsequent editors.

This levelling implies that even highest reputable professionals find themselves in a fierce competition with colleagues as well as with outsiders who may have gained heir knowledge by extraprofessional (e. g. autodidactic) channels.

Following rational choice theory, we might assume that higher level experts are strongly discouraged to participate, because they gain much less acceptance than when they use more conventional channels [29]; while lay persons are strongly encouraged because for them, wikis may provide the only arena where they can successfully display their knowledge and their ideas. [30]

Printed encyclopedias are highly ambitious endeavours needing high investments of resources and long-term editing commitments. Books in general need much care and effort in order to avoid errors irreversibly fixed on printed paper. In the case of encyclopedias, errors are particularly harmful because they may be copied and diffused throughout society to the degree that the work is used as a reference. Like in the case of telephone books, the reputation of an encyclopedia depends fully on its perfect reliability, and intensive checks and controls are necessary for keeping up to such standards.

Usually, only a few potent societal actors or collectivities are capable of engaging in such a project: e. g. monarchs, rich elite members, foundations or governmental institutions. As a consequence, encyclopedias tend to be born and raised at the very center of societies. Very often, they have been created for the purpose of expressing the tradition and thinking of a societal elite or of symbolizing a national culture. Thus, most of the classic Chinese encyclopedias owe their existence to the patronage of imperial rulers; the emperor Constantine VII of the Eastern Roman Empire (913-959) was responsible for a series of encyclopedias, and king Alfonso X of Spain (1252-1284) sponsored the “Grande e general estoria” (“Great and General History”).

Only rarely we also find more encyclopedias originating more at the periphery of society: like the famous French “encyclopédie” of Denis Diderot that was emerging within the enlightenment movement that opposed the reigning religious institutions and monarchical regime.

Since the 19th century, such independent endeavours have been less frequent for various reasons: e.g. because the volume of relevant knowledge has expanded, the demands for comprehensiveness and reliability have risen, and the costs connected with new printing technologies and distribution procedures have increased.

Especially the 20th century was rich in “governmental encyclopedias” aiming to provide a most impressive picture of national culture and national achievements (e. g. the enciclopedia Italiana, the Soviet “Granat” encyclopedia or the Enciklopedija Jugoslavije (first published 1955–71). This explains why in many cases, not only the most educated scholars, but the most powerful personalities of the respective time (e. g. Lenin and Mussolini) have made significant contributions.

Because of high costs, conventional Encyclopedias have a high expressive value as status symbols. Whenever an EB or a Brockhaus is found on a shelf in a living room or in a private library, a double message is sent out: that the owner has (had) enough money to buy it and enough education to make use of it (while everybody knows and accepts that it is rather rarely consulted).

Seen in this wider historical perspective, digital encyclopedias help to regain the independence from governmental power centers, economic enterprises and other societal institutions: an independence that was quite remarkable between the 16th and 18th century but was later lost in the course of rising nationalism and industrialization.

Labour costs are very low because so much motivation for unpaid voluntary collaboration can be tapped (see 4.2); and capital costs are negligible because like other Web projects, the Wikipedia thrives on hardware and software resources that are already fully available for other reasons: individually owned PC’s already acquired for various private or professional purposes, and excess capacities of networks that have been built for telephone transmission or other commercial purposes.

For several reasons, populations in rich modern societies have high “discretionary resources” (in terms of free time, money or skills) that are disposable for various new purposes because they are not committed to work or family duties (McCarthy/Zald 1977; 1987). There are many potential providers of such resources: e. g. temporarily jobless or partially employed people, students, housewives or retirees. In addition, changes in modern lifestyle contribute to a growing ”decommitment” of resources; many adults live alone or with few or no children, many are rather isolated immigrants far away from relatives and friends, and increasing numbers do not participate in voluntary associations or political parties, so that they spend most of their free time at home (Putnam 2000).

The Internet amplifies these individuals further by providing unlimited possibilities for data transfer and communication and by harnessing them to a large variety of new purposes: e. g. by providing interactive online networks where everybody has a chance to feed in his or her contributions.

A major decline in labour costs is caused by the demise of many more "ritualistic" activities that make the finalization of printed works so cumbersome. For instance, much work is dedicated to "streamlining" and "homogenizing": e. g. by implementing standardized criteria of typing, grammar, orthography, titles, footnotes, bibliographies etc. While such standardizations do not contribute much to readability, they seem nevertheless indispensable for aesthetic or conventional reasons.

In the case of computerized hypertexts, there is much more tolerance for inhomogeneities: maybe because each page is seen in isolation, so that inconsistencies across pages become less visible than by skimming a book As a consequence, much editorial overhead costs can be saved. On the other hand, this implies that it is highly difficult to transform electronic encyclopedias into manuscript ready for publication. Such high prospective costs for "streamlining" were a major reason why the ambitious book edition of the German Wikipedia has recurrently failed. [31]

Among many other consequences, the decline in production and distribution costs implies that conventional limits of growth and accessibility are completely removed.

1) In the printing age, there were always harsh limits on the total size of encyclopedias, and thus indirectly on the volume dedicated to various articles or systematic divisions. (For instance, the total text volume of the EB and Brockhaus has remained on the same level since about 150 years!).

Therefore, editors were always required to exert selection: thus inevitably expressing their own personal preferences about what shall be included and what deserves a shorter or more comprehensive treatment. This selectivity was highest in the one-volume pocket encyclopedias which always tended to be heavily shaped by the personality of a single author.

However, the larger and the more anonymous the public, the more pressing the need to broaden the scope in order to satisfy all the different interests. But as the overall space was limited, this resulted in an ever more atomized knowledge structure with a declining average size of entries [32].

By contrast, digital encyclopedias can expand without such pressures toward atomization: by just adding new and expanding additional articles at the same time.

Because of its unlimited potentials for growth and diversification, the Wikipedia fares far better than printed works in exploiting the "Long Tail" (Anderson 2004): the large number of highly specialized information needs articulated by very many infrequent users.

In fact, the Wikipedia builds its reputation heavily on the whole sum of quite unpopular, rarely consulted articles, while classical encyclopedias base it predominantly on a smaller number of more frequently used entries. [33]

This implies that the Wikipedia is attracting a very large and highly diverse public, similar to Amazon which lives from selling few copies of very many different books. On the other hand, the lack of physical resources makes any kind of filterings and shortenings difficult to legitimate: because they cannot be justified by technical or economic arguments: so that more ideological, philosophical or scientific reasons have to be provided (arguments likely to be quite dissensual over cultures, user groups and historical time).

2) In the age of printing, there was a rigid trade-off between volume and accessibility:

Only very small encyclopedias were cheap enough to get a large distribution and light enough to be carried along. More user-oriented encylopedias including “everything” were not only expensive, but so clumsy that their fate was to remain on rarely visited bookshelves in libraries or other rooms far apart from practical activities and “real” human life.
Digital encylopedias can grow unlimitedly without losing accessibility caused high acquisition costs, clumsiness or other material factors. Soon, they will be fully available for portable handheld devices or in audio form, so that users have all information at their fingertips at the very moments they need them (e. g. within a meeting or while driving a car).

3) Traditional encyclopedias could only be produced by sizable and rather wealthy collectivities; preferably by populations organized in nation states with governmental agencies and large editing houses that could act as initiators and sponsors.

Wikipedias, by contrast, can flower everywhere, because even tiniest groups have enough potential to set such online processes of knowledge accumulation into motion. In fact, there is a global diffusion of Wiki technology, protocols and software, because this knowledge is so formalized and decontextualized that it can be copied, transmitted and implemented everywhere, regardless of any socio-cultural and linguistic conditions. Thus, Wikipedia clones have rapidly sprouted in about 250 languages, even if many of them are still “empty shells” waiting to be brought to life by active users.

Given that all contributions to the Wikipedia belong to the public domain, anybody disagreeing with the current encyclopedia is allowed to initiate a new encylopedia by simply eliminating all the unwelcome entries and retaining all the rest. Such “forking” has given birth to the “Enciclopedia Libre Universal en Español (split already in 2002) [34], and to the foundation of Wikinfo [35] in 2003 (an alternative project rejecting the rigid “neutrality” principles of the Wikipedia).

4) Given their large and long-term need for subsidies, traditional encyclopedias could usually gain only limited autonomy from their mighty economic and political sponsors; and given their linkages to academic elites, they had no choice than to give priority to institutionally established knowledge cultures.

In comparison, the Wikipedia has hitherto remained remarkably independent from economic corporations as well as from governmental agencies and educational institutions. The economic autonomy is dramatically highlighted by the complete lack of advertisements and the very low operational budget that is mainly covered by a multitude of rather small individual donations. While “blind spots” and censuring endeavours certainly exist, they seem to be associated more with idiosyncratic personal sensitivities than with larger-scale institutional interests and strategies. While this allround autonomy is certainly a highly valuable asset, it is also a source of risks because it makes the WP “underdetermined” (Berinstein 2006) and therefore vulnerable to any kind of intrusions, assaults and even “kidnappings” by any highly active particularistic groups.

By encouraging the most knowledgeable individuals of each culture to work on their own encyclopedic project, a multi-domestic and multicultural repository of human knowledge in almost all written languages may come into existence.
In some cases, Wikipedias may even offer a “last secure haven” for a language on the brink of extinction, or they may upgrade indigenous languages (that hitherto have been spoken only orally) to the level of writing (e. g. Alemannic, Sardu, Ladino or “Plattdütsch”). Even artificial survival (or revival) resorts for dead languages may be created: e. g. the rather lively “Vicipedia” in classical Latin that has 1100 entries (in Jan 2007) and is progressing with about 300 updates per day.

Given a median value of about 800 articles. many WP’s are very small – and will remain so – because they represent marginal linguistic communities with only a few hundred (or thousand) speakers.

As they have no potential to cover contemporary knowledge, their function is more expressive than instrumental: making the minority language visible to its population of speakers as well as to a worldwide public, demonstrating that the language has still active promotors who want to give it a place within the sphere of written culture.

In fact, constructing an encyclopedia means: putting a language under a very hard test: stretching its verbal expression capacities to the utmost by conceptualizing and describing an unlimited number of different topics, by importing and assimilating a an ever growing manifold of terms from other languages, and by creating neologisms for keeping pace with new developments and events.

At the same time, a focus of collective identity is created which may be particularly important when the speakers are not living together in the same geographical location, but are dispersed in diaspora where they have no opportunity to use their mother tongue. In some cases, contributors seem to consider the WP as a vehicle for transporting and reinforcing traditional folkways and other elements of traditional culture. [36]

Given the small number of speakers, such minority WP’s have to be built on a very small group of collaborators who tend to shape it strongly according to their subjective preferences and views.

In many cases, they invest their limited energies into creating blueprint structures of many different “stubs” (e. g. about each local community): leaving it to others to insert corresponding content.

How will the relationship between the few "big" and the numerous small WP's develop in the future?
On the one hand, the rise of so manly smaller WP's in practically all existing languages has the effect that the originally dominant WP's retreat into a more modest relative position.
In the case of the English WP, however, this status loss is attenuated (or even neutralized) by the fact that by representing the hegemonic Western knowledge culture in the most dominant of all current languages, it occupies a singular reference position of global reach - not to be compared by other Western WP's that are more exclusively affiliated with "their" national or linguistic culture.

In this respect, the English WP may "inherit" at least partially the supreme status of the Encyclopedia Britannica which is - and always was - able to attract a worldwide public of readers as well as a globally dispersed collectivity of first-rank contributors.

This singular significance may explain why since 2004, its share of collaborators does not diminish in the same way as its percentage in articles and words, but has stabilized on an astonishingly high level: well above 40%. Especially the share of total collaborators has lately again been on the rise, so that at the end of 2006, it has almost regained the level of 2003.

As a consequence of this outstanding hegemonic role, the English WP has a particular responsibility in transmitting accurate, complete and consistent knowledge, because it serves as an authoritative knowledge provider for so many users, including the active contributors of many lower-scale Wikipedias all over the globe. Therefore, it will come more under particular pressure to install far-reaching mechanisms of internal control.

Secondly, it is highly probable that the English WP becomes an arena for any kind of global controversies on scientific, ideological, philosophical or religious levels, including the imminent clash between Western and Islamic culture.

Conventional encyclopedias are produced under the condition that many parameters are rather irreversibly fixed in advance. For instance, the realm of knowledge to be included has to be circumscribed, lists of experts to be called for contribution have to be compiled, and the organization of the whole enterprise (in terms of roles, competences, norms, procedures, deadlines etc.) must be defined.

In many cases, there additional exogenous constraints: the publication has to insert itself into the tradition of antecedent editions of the same product (e. g. Brockhaus or EB), it has to fit into the larger edition program of the publishing house, and expectations of potent sponsors may have to be satisfied.

As a consequence, such printing projects are likely to be overdetermined: far from being flexible for adapting to environmental needs, the personal and organizational parameters may even be in contradiction to the stated mission and the concrete functional needs.

Wiki encyclopedias do not require such antecedent decisions. They may start “from the scrap” as very embryonic projects without any explicit planning and design. Within a process of unplanned incremental growth and “open-ended evolution”, it will be determined ex post what contents are considered, who participates in what way and what kind of organizational procedures and structures may develop as a result of manifold smaller decisions and adaptations. Of course such projects are likely to be “underdetermined” because on the one hand, structures are flexible enough to adapt to task needs an environmental conditions; on the other hand, such conditions are themselves not “given”, but subject to changing collective decisions.

Evidently, Wikipedias are better adapted to highly complex and dynamic societies where the parameters needed for organization building are not known ex ante, because

Apart from its anchoring in the Wikimedia foundation, the Wikipedia’s internal structure is primarily shaped by endogenous forces, and it is highly flexible and selftransformative, because it does not rely on the acknowledgement of externally generated status criteria like educational degrees or professional reputation.
Evidently, it is certainly not an “expertocracy” governed by a reputational elite with formal educational credentials; it doesn’t contain any trace of an “aristocracy” based on outstanding family background and breeding; and it is not a dictatorship stabilized by the repression and sanctioning of dissident voices.

Instead, it starts with “equipotent” participants and status inequalities are generated ex post as a consequence of different levels and qualities of performance.

This means that participants have primarily to relay on their own judgment whether they are knowledgeable enough to contribute to a specific topic, or what they have to learn additionally in order to possess all the relevant information. And when modifying already existing texts, they need enough selfconfidence for being sure that they know better than their predecessors. Especially in the large sphere of more marginal articles rarely read by anybody, it is crucial that only the most knowledgeable individuals feel a motivation for writing, while ignorants have at least enough insight and self-control to keep away.

While this “anarchistic individualism” has proven to be a viable starting point (because it does not predetermine specific organizational structures), there was never the intention to cling to it for fundamentalist ideological reasons. Instead, the overall mission to create a reliable encyclopedia made it necessary to create incrementally a highly “polymorphic” system that combines elements of very diverse regime types and organizational structures. [37]

First, It is still “anarchistic” in the sense that everybody can actively participate without membership duties, without even disclosing his or her personal identity.

Secondly, it is “autocratic” insofar as the founder (Jimmy Wales) has the towering role of a “God King” who can intervene in any possible ways without constitutional controls (Pink 2005). There is even an element of “latent totalitarianism” in the sense that an unlimited centralization of power is easily possible without any constitutional controls. Thus, Jimmy Wales himself and some of his admins sometimes exert the power to “redefine” even the history of Wikipedia: by eradicating earlier text versions so that no hints remain that they have ever existed. Evidently, the leadership of Wales was particularly crucial at the inception of the Wikipedia project where it was important to define the mission (to create an encyclopedia, and nothing else) as well as the most important behavioral norms. (e. g. the principle of neutrality). While the personal (somewhat “charismatic”) authority of Wales may be large accepted currently because he is considered as a “benevolent dictator”, it questionable how this personal authority will ever be substituted when he leaves.

Third, the WP structure is “bureaucratic” insofar as various roles with highly formalized competences and duties have been created. In fact the Wikipedia confirms the regularity that when open social systems want to maintain a higher state of order, they are forced to generate high (and permanently increasing) levels of formalization and highly sophisticated mechanisms of control, because they have to deal with a large variety of problem cases and with very heterogeneous collaborators. An increasing number of “admins” is regularly patrolling the system: with special competencies to block editing, delete articles and revert texts to earlier versions. Many of them do a rather regularized job by getting alerted whenever specific pages are changed. so that they consult in order to "approve" the additions or to revert unwelcome modifications.

Their appointment has itself become a matter or highly formalized nomination and voting rules , so that superior “bureaucrat” and "steward" positions had to be created for taking the decisions or supervising the procedures, and a highest elite of "developers" has emerged that can implement direct changes to the software and database. In addition, mediation bodies and a higher ranking “arbitration committee” have been installed to rule in cases of severe edit wars. And most importantly: an increasing canon of explicit rules, norms and procedures has been formulated for informing all users about the provisions they have to take when creating new or editing existing entries, and for guiding their informal policing activities. However, these bureaucratic mechanisms have only a subsidiary role insofar as they come into play only “after substantial play has been given to selfpolicing by participants, and to informal and quasi-formal community based dispute resolution mechanisms.“ (Benkler 2006: 104).

Fifth, the Wikipedia is “democratic” insofar as “admins” are selected by elective majorities (even if such elections are not based on a representative participation), and as many other decisions are the result of lively open discourse and deliberation (e. g. on the “talk” pages).

Sixth, it is somewhat “plutocratic” to the degree that is depends financially on donors who may have a say over the strategy of the whole enterprise, and that considerable power is exerted by the governing board of the “Wikimedia foundation” – the non-profit frame organization which owns the material assets.

Seventh, it is “technocratic” insofar as specialists determine the development of Wikipedia on the software level (protocols, programs and network technologies).

And eight, finally, it is certainly highly “meritocratic” because only participants with high activity level and excellent performance records have a chance of being appointed to higher roles. [38]

Given the lack of "vested interests" usually pursued by fully paid employees and managers (e. g. for securing employment or maximizing prestige and power), there are good reasons to believe that the evolution of the WP's organizational structure follows "contingency theory" [39] by adapting flexibly to task types and environmental conditions.

A conspicuous characteristic of this organization is certainly the complementary interplay between decentralized structures centralized structures. a constellation not too far from conventional encyclopedic productions.

All printed encylcopedias always rely on a complementary relationship between
- a large number of outside experts who provide the bulk of content
- a much smaller circle of "editors" busy to correct, filter, streamline and synthesize the material.

There is some evidence that a similar labor division between "complexity generating" and "complexity reducing agents has emerged within Wikipedia, without conscious planning.

Thus, Aaron Swartz has found that most of the text volume is provided by rather perpheral users, while registered users and "admins" concentrate mainly on on additions, revisions, abridgments and deletions.

Such empirical findings plainly contradict the position maintained by Jimmy Wales in innumerable speeches: that the Wikipedia is predominantly written by a rather small community of about 3000 regular Wikipedians.

They imply that the growth and diversification of the Wikipedia is primarily dependent on an expanding number of actively contributing users - not on an increased work load carried by more restricted cores.

While it is heavily important to make participation easy and rewardable for such large masses of occasional contributors, it is difficult to do this because these peripheral users have very little say and influence in the whole system:

On the other hand, the "editors" are an important factor in maximizing downloads and readership: e. g. by synthesizing materials, by eliminating technical jargon and by presenting the material in clearly arranged forms.

The future of the Wikipedia will heavily depend on he equilibrium between decentralized contributions and centralizing coordinations. It could easily be stifled if admins are turning to a heavy-handed regime, and it could explode in chaos if these editing services would weaken (e. g. because not enough unpaid volunteers are found for such rather "bureaucratic" tasks).

Another important problem is that these occasional contributors are too little involved in the discussions and modifications made after their postings. Very often, they may not consult "their" pages frequently enough to see the changes and deletions made by other users: so that modifications to the worse may remain uncorrected.

A similar symbiosis is found on the strategic level where centralized leadership is needed in order to direct the efforts of content producers into predetermined channels.

For instance, some more responsible editors create "stub articles" about hitherto neglected topics they think deserve encyclopedic attention. by doing this, they invite experts to channel their work energies on these topics.
In other cases, articles are characterized (with a remark below the title) as insufficient: needing elaboration or a better indication of sources.

In the future, more efforts may be needed to bundle such unsatisfactory pages into topical categories, and to address such bundles to specific groups of scholars: inviting them to contribute their valuable expertise.

It is important to see that internal centralization is fostered to the degree that the WP is confronted with external attacks to which is has to respond quickly and in a decisive fashion. Thus, Jimbo Wales has installed the policy of "office action" in order to avoid conflicts arising from imminent legal action or informal complaints (e. in cases of problematic biographic entries). Whenever a serious complaint is directed at Wikimedia Foundation (the legal person responsible for Wikipedia), Wales or one of his delegates remove the article temporarily, so that harm (e. g. personal slandering or libel) is avoided the justification of the complaint can be examined. [40]

Formal organizations are typically embedded in larger, less formalized structures. Many are components of societal institutions (like economy, polity, military, education) from which they derive their basic values, norms and structural patterns (Powell/DiMaggio 1991), and others are parts of ethnic or religious collectivities or world-wide social movements (Zald/McCarthy 1987: passim).

In the case of the Wikipedia, this larger substrate may well be called a “community”: in the sense of a rather stable collectivity that acts as a breeding ground for common values and behavioral standards, as a group context for interpersonal communication, as a reference group for personal identification and as a supraindividual agency for effective socialization and social control:

In contrast to communities that base their identity on a common history and tradition, on a specific locality or even on a particular founder, Wikipedians anchor their cohesion in the visible output of their cooperative endeavours: the Wikipedia as it flowers and raises in global recognition and reputation.

As an objectified structure, the Wikipedia has a dual quality: on the one hand, it constitutes a centralized focus on which all contributors fix their attention; on the other hand, it constitutes the decentralized environment in which every user easily finds his own "working niche".

While all work independently on their own subprojects, they at the same time feel a sense of togetherness: like masons working on different walls of the same cathedral. In contrast to the cathedral, however, the Wikipedia can become the workplace of thousands or even millions of (simultaneous) contributors without compromising this basic unity and centeredness which is the basis of communal integration:

This integration rather independent of any horizontal interaction, because the vertical reference (of any peripheral member to the common focal center) generates enough sense of unity and social integration. In fact, this vertical (or radial) integration is so potent that users can engage in a multitude of controversies and conflicts without running the risk of desintegration. (The only threat is the exit option of "forking" with its segregative implications).

As seen in many cases, the Internet gives rise to new collectivities by allowing self-recruiting activists and self-constituted groupings to gain worldwide visibility and reputation by pooling their efforts and resources producing widely accepted goods and services.

Thus, the Linux community has become potent enough to challenge Microsoft on the level of operation system software, and the Wikipedia community seems to approach the status of global “cognitive authority” defining the canon of “uncontested human knowledge” as a result of a very extensive and long-term process of cooperative interaction.

Because formal hierarchical control is weak or inexistent, control has to be provided by the contributors in the form of self-guidance on the one hand and mutual peer-control on the other. Thus, the anticipation of being corrected by others subsequently can act as a powerful motivation to avoid errors:

As a consequence, the integration of Wikipedia articles in the WWW is relatively low. (see Figure 12 above).

In some aspects, the Wikipedia is dominated by a “geek adhocracy” [43]: an aggregation of self-recruited activists whose dedication to the project is expressed in a very large amount of editing activity.

“A vigilant army of self-styled Wikipedians defend the site and enforce community policies based on the principle that Wikipedia is an encyclopedia and not a forum for advertisements, slanderous remarks or pictures of your cat. They police the site to try to establish a neutral point of view, warn users against violating copyrights, and call for respect toward the contributions of others. (Wilson 2006).

On the elementary levels of daily activities and interaction, the communal culture is manifested primarily in a particularized language. Thus, Wikipedians "revert" (or even "rerevert" pages when they reinstate an earlier version, they love "Wikignomes" who are dedicated to patient low-profile tasks like correcting grammar mistakes or broken links; and they hate "Wikitrolls" who permanently violate guidelines and engage in various disruptive behaviors.

Among the values of the WP community, a "passion for correctness" stands out that is manifested in many "edit wars" that appear highly ritualistic because they focus on very tiny points like orthography and punctuation. For instance, there was extensive discussion in the “September 11, 2001 attacks” whether a second comma should be inserted (after 2001).

As dissensus and quarrel lingered on for weeks and months, the page was not promoted to the status of a “featured page” in Jan 2005. [45]

Similarly trivial was a fight concerning the entry about scientology where contributors argued for nine months over whether the Scientologist method of childbirth should be called "silent birth" or "quiet birth."

Like most communal collectivities in the RealWorld, the WP community functions as a breeding ground for groupings that arise easily among the members for dealing with specific temporary tasks:

Whenever the edition of an article is dominated by a highly consensual group, it becomes very hard for outsiders to intrude and to make their own contributions (Cormaggio 2006). This illustrates that the principle of openness of the project (implying that every anonymous user can edit all pages) can only be maintained when rather weak community ties among collaborators exist. In fact, community is functional for purposes of integration: e. g. for implementing homogeneous standards of filtering or for fighting vandalism, but it is rather disfunctional for diversification and growth: because such expanding activities demand openness for any new contributions.

Interestingly, the Wikipedia community has reproduced in the digital sphere the same dichotomy between "frontstage" and "backstage" performances that - according to Goffman - is a general characteristic of groups operating before a public (Goffman 1959).

On the one hand, there is the frontstage of serious work relationships: resulting in the articles everybody can see. Here, individuals are under pressure to be behavior in a highly disciplined: conforming to collective norms that strongly forbid the expression of subjective emotions, opinions or the playing out of intimate interpersonal relations. On the other side, there is a high backstage of talk pages, online conference meetings and bilateral exchanges, that allow the playing out of spontaneous personalized activities and the satisfaction of socio-emotional needs.
“Like at a paid job, some people choose to extend the relationships they have within the ‘workplace’ to a context outside the workplace. Metaphorically (and, sometimes, literally!), they stop by the pub with their workmates and have a few beers. They may joke about situations "on the job", they may talk about their personal lives. They may even do back-of-the-napkin brainstorming sessions that fix problems nobody expected. "Beers after work" happens on talk pages, User talk pages, on the mailing lists, in edit summaries, in person-to-person meetups, in private email, in IRC or Jabber chatrooms... the list goes on and on. Whenever Wikipedians drop their business-like demeanor and address each other as human beings, with warmth and personality, there's the smell of beer somewhere in the digital air.” [46]

Given the very long-term production and diffusion processes, conventional encyclopedias had all a strong bias in favour of past knowledge and knowledge about the past. This distance from current knowledge was aggravated by the need to check everything thoroughly (in order to preserve the reputation for reliability), and a strong tendency to rely on extant texts that have appeared already in earlier encyclopaedic editions.

Therefore, they always focused very much on knowledge as a stable, fully consensual canon of immutable facts (e. g. historical events, geographical locations, or just the meaning and spelling of words) or regularities (e. g. mathematical or natural laws), and they abhorred the fields of insecure and volatile knowledge where the state of the art changes weekly as a consequence of ongoing controversies or new scientific publications. Many (like Zedlers gigantic “Grosses vollständiges Universal-Lexicon”) did not include biographies of living persons. Most in harmony with the clumsy printing technology were thus dictionaries, gazetteers and historical encyclopedias: mere compilations of atomized information pieces that had never to be revised. The bias toward stable knowledge had also the effect that many encyclopedias presented themselves as treasuries of highly established elitist culture. Thus, the Brockhaus “Konversationslexikon” had the explicit goal of making bourgeois parvenus fit for successful participation in the more polite aristocratic circles of their time.

Therefore, only encyclopedias of very traditional societies could maintain the conception that they were mirroring the whole of true knowledge: e. g. the middle age encyclopedia “Speculum majus” of Vincent de Beauvais (1244) which aimed to provide a definitive view of “the world how it is and how it should become”.

At least since the Renaissance, encyclopedias have given up such ambitions by acknowledging that in a dynamic society with permanently advancing knowledge, attempts to crystallize out stable knowledge compilations can only have very limited success. The more their focus shifted toward natural sciences and technological branches, the more they had to face a world of constantly changing knowledge – without becoming ever able to react to such changes in any flexible way.

A very clumsy way to keep pace was to add periodic updates in the form of monthly additions (e.g. the “Larousse mensuel illustré” since 1907) or yearly volumes: (e. g. the “Britannica Book of the Year” since 1938). As there was no way to integrate the chronological additions into the alphabetical or systematic structure of the original encyclopedia itself, they were not very helpful for the readers, because an ever growing number of chronological volumes had to be consulted. However, they had a useful function in complementing news media by setting daily events in a broader, encyclopedian perspective, because

Of course, there was also a sharp trade-off between size and updating possibilities. Only one-volume lexica (like the medical "Pschyrembel") could be frequently updated to keep pace with rapidly changing knowledge and terminologies. Paradoxically, updating was most difficult in the case of encylopedias most needing it: large multivolume works that contained detailed information much more subject to change than shorter dictionary entries.

This explains why the intervals between subsequent editions of the Encyclopaedia Britannica have increased between the 18th and the 19th century [48], despite the immensely accelerated production of new knowledge that would have made more frequent updates highly desirable.

Similarly, there was always a very unfortunate trade-off between updating capacities and the degree of interdependence and synthesis of the knowledge presented.

Updating is most easy in the case of highly atomized, fragmented knowledge structures like dictionaries or gazetteers, because each change or addition is only affecting a single entry. The more knowledge is presented in interdependent structures, the more frequently it occurs that a modification of one article has an impact on several other entries. For instance, when the biography of a politician has to be reassessed, this may have implications for other articles like the history of his country or his political party, in which this same person is involved. Similarly the emergence of a new scientific theory may affect many articles where its impact on the interpretation and explanation of different phenomena are discussed.

One of the most fundamental and most problematic innovations of the Wikipedia is to expand the notion of “encyclopedic knowledge” to phenomena of contemporary society and culture, to living persons and to current developments and events – thus entering into competition with journals, magazines and other news sources in the Mainstream Media system as well as in the Net.

In fact, Wikis are highly efficient tools for aggregating information about current events that are experienced by many witnesses from different angles: such as earthquakes, hurricanes and floodings, war battles, city riots, pandemies etc.

In such cases, they can act as platforms for the inductive collection of knowledge by many selfrecruited contributors who may effectively enlarge and enrich (or also relativize or falsify) the information provided by professional journalists or from official sources (Dorroh 2005).

On several occasions, the WP has already proven its status as an authoritative news source because numerous contributors are busy to keep pages tightly up to date with unfolding developments and events:

Similarly, the Israeli-Libanon conflict in summer 2006 gave quickly rise to a corresponding page that experienced more than 10 000 edits within a few days: offering a multifacted and highly balanced account of the unfolding war while keeping pace tightly with all the incoming news. [49]

Likewise, it took only four hours for the "Execution of Saddam Hussein" entry to evolve through 630 edits into a detailed account of the event as well as on the international reactions. Together with all the external hyperlinks, it reached a size more than 1300 words. [50]

Contrasting with the isolated shortterm news reports in the media, such Wikipedia articles often combine timeliness' and historic depth at the same time.

By attracting a large number of contributors, such articles become sites of very timecompressed history construction "from below": by aggregating highly diverse information that cannot yet be integrated in overarching blueprints and concepts because the event - as well the way it is interpreted by the media and the general public - is still under way. The question arises whether such articles arising from current news are later reorganized in the light of subsequent developments and the broader, more distanced interpretations that usually go along with evolving time.

This hybrid role of the WP as a news source and a historical source has several far-reaching implications.

First, this implies that much of its content is focussing on matters widely apart from the canon of classical culture (like that transmitted in institutions of formal education) e. g., computer games, TV series or Heavy Metal music productions. Given the rather low average age of many most prolific collaborators (see 4.1), it is not astonishing that the entry about Augustinus is less comprehensive than that about Britney Spears.

Secondly, many articles are inevitably incomplete, erroneous and controversial, because they refer to subjects still in the realm of change and ongoing public discussion: e. g. recent scientific discoveries, still active writers or singers or unfolding political events. While the Wikipedia certainly derives a major part of its skyrocketing popularity from the fact that it can also be consulted in such current matters, it also suffers from the additional unrealistic expectations associated with fulfilling this widened role.

This problem has been dramatically highlighted at the death of Kenneth Lay (mentioned above), when the Wikipedia was heavily criticized because the true cause of death was only reported with some hours delay. Of course, such denouncements are just revealing to what degree the Wikipedia has already gained the status of a universal news-knowledge provider – not to be compared with a conventional encyclopedia which is given years for collecting, checking and reporting such kind of information.

However, it is remarkable that even the Wikipedia has preserved a conservative bias by disallowing the publication of any “original scientific studies” [51]. In other words: encyclopedic knowledge is still “second hand knowledge” that has already been certified by the academic community: so that very new, not yet certified knowledge has no place. Another ”traditionalizing” effect of Wikipedia stems from the easiness with which already existing online texts can be included by simple “copy and paste”.

Thus, the Wikipedia contains much material from rather antiquated encyclopedias that have been publicized on the WWW because they have become part of the public domain (e. g. the Encyclopaedia Britannica of 1911 and the “Catholic “Encyclopedia” of 1913).

Given its permanent modifications, however, the Wikipedia remains in a state of fluidity that makes it difficult to integrate its contents into larger and more stable cultural productions. For instance, it is difficult to cite a Wikipedia article in any other text (online or offline), because one always has to indicate at what exact time the article has been retrieved – it might have looked different a few minutes earlier and might be modified a moment later.

Conventional encyclopedias are ridden by an unfortunate trade-off between size and usability. Only small packet editions are accessible in a way their knowledge can be retrieved in a manifold of places and situation and become part of many different human activities and social communications.
Larger editions are clumsy to handle, typically stationary in libraries or other rooms where few other activities than mere reading takes place.

As a consequence, many explicit ambitions of paper encyclopedias had to remain utopian: the Brockhaus or Meyer notion that it should support educated human discourse (“Konversationslexikon”) as well as the old Roman perspective that by providing all relevant human knowledge, encyclopedias could help individuals to carry out all their daily tasks on a higher level of competence.

The mere physical problem of handling many heavy volumes is an obstacle for cognitive synthesization. In fact, the increasing the number of articles has inevitably to be paid by a rising fragmentation of knowledge, because even when many crossreferences are included (as in the classical Brockhaus editions), they are not likely to be followed because too much time and effort is needed to switch between different tomes.
Thus, the trend toward dictionary type encyclopedias with many smaller entries since the 19th century (Brockhaus, Meyer, Larousse) has almost eliminated the possibility to transmit more complex knowledge structures that transcend the atomized level of explaining the meaning of particular words, concepts or names.

1. Hundreds of million users can reach it almost anytime and anyplace on the WWW. Given their easy accessibility (irrespective of size and internal fragmentation), digital encyclopedias can fulfil better the function for which their conventional forerunners have already been explicitly conceived: encouraging individuals to enlarge their cognitive world by acquiring at least some basic knowledge about a topic beyond their daily experience and professional expertise.

2. Every single user can access it in a large variety of situational conditions and roles: e. g. for getting immediately specific practical information for solving a current problem or for inserting it in a developing document or mail communication. The chances that such knowledge is actually mobilized when need arises are far greater – and they will increase additionally when Wikipedia knowledge is universally available in mobile handheld devices. Such portable WP editions exist already for the ipod [52] as well as for notebooks and handheld ebook readers [53] and for the Mobile phone (“Wapipiedia”). [54]

3. Given the denselyknit hypertext structure of the Wikipedia, every user has at every moment a multiplicity of possible ways for navigating through the system: thus realizing his specific preferences or optimizing the way new knowledge can be integrated with what he or she already knows. In other words: individuals are better able to transform decontextualized universal encyclopaedic knowledge into contextualized individual knowledge that can be assimilated to particular individual thoughts and activities as well as social communications and cooperations.

This easy integration into microsocial contexts and individual roles provides good preconditions for expanding the sphere of the encyclopedia from factual and theoretical “know-what” knowledge to practical “know-how” knowledge that can be used in everyday life for guiding any kind of human action. For instance, by including advices about how to counter hiccups, how to relieve headache or how to prepare espresso coffee, the Wikipedia revives encyclopedic traditions of the 18th century where a similar weight was laid on such practical knowledge (e. g. the first edition of the Encyclopaedia Britannica 1768-71) – but with far better chances that it will be factually applied.

6.8.2 Combining receptive and participative roles

Until the 16th century, encyclopedists conceived their works for a small, rather neatly circumscribed circle of recipients that shared not only the same language, but also the cultural and religious background of the producers.

To the degree that these people knew each other personally, there could also be high informal feedback – among the recipients as well as between recipients and producers.

In the following period, printing technology was responsible for a drastic segregation between producers and consumers. As editors were increasingly confronted with an anonymous unknown public, they lacked the necessary information for matching their works with the recipient’s preferences.

As a consequence, we see the spread of extremely “producer-guided” encyclopedias that are primarily conceived for expressing the cultural tradition of a national elite or the ideology of a intellectual movement (like the French encyclopédie), not for satisfying any needs of potential readers.

This dissociation was reinforced by the almost complete lack of backchannels: so that editors got no feedback from the reader’s side that would have helped them to adapt better to their wishes

In other words; there was a rigid trade off between expansion of readership and feedback: The wider the distribution (thanks to mass printing), the less it was possible to anticipate the structure and composition of recipients, and the more complete were producers socially isolated from their readers.

When using a conventional encyclopedia, I completely embrace the role of a pure recipient who is confident that the information found is correct. This generalized confidence may of course be based on the high reputation of the encyclopedia as a professional and reliable source, but in addition, it is also made inevitable because as a reader of the encyclopedia, I have no immediate access to alternative information sources and no possibility to communicate with the responsible producers. Under these conditions, of course, reliability is absolutely essential; unreliable encyclopedias are completely useless. As a consequence, printed encyclopedias resemble phone directories or train timetables by aiming at an ideal state of “complete accuracy” - what implies heavy costs because high marginal costs are associated with finding and eliminating the very last remaining error.

Thus, conventional encyclopedias foster the regressive role of an “unconditional believer” who doesn’t take any critical stance. Publishing occurs only at the final point of a very long and complicated editing process that is usually completely hidden from the eyes of outside observers. This implies the premise that readers are only interested in the product, not in the intermediate processes of production.

By contrast, Wikipedias develop in public, so that all participants can not only observe and evaluate all successive stages of development, but also participate in the formation and modification of the rules by which they are guided, and intervene whenever they see a reason. Thus, the categorical dichotomy between “producers” and “readers” gives way to the hybrid role of the “participant” or “user” who can combine both roles in a way completely at his own choosing: by sifting through materials others have written at one moment and by posting his own contribution at another.

As every article is permanently “under construction”, users feel invited to read everything with a critical eye and fundamental provisos: ready to validate any information by additional sources whenever absolute certainty has to be achieved.

As a Wikipedia user, therefore, my role is rather complicated, because I have to combine my stance as a “faithful recipient” with an element of sceptical role-distance: maybe the current content contains errors or has been vandalized ten minutes ago – and in both cases, I may assume the responsibility of not only noticing, but actively eliminating such flaws.

It has often been remarked that in contrast to oral speech, written texts facilitate critical reflexivity because they stand out as objectified artefacts that can be interpreted by anybody at very different occasions from widely different angles. Thus, they give rise to a communicative meta-level where they themselves become the object of oral talk or written commentaries. However, these reflexive capacities could not develop fully in the printing age, because in most cases, readers had no feedback channels available for expressing and communicating their thoughts.

Digital texts on the Internet are disposed to catalyze much higher levels of reflexivity for two reasons:
1) because feedbacks can easily be made by using the same medium that has given rise to the primary text;
2) because feedbacks are themselves in a written form: so that they easily become themselves the objects for further (their order) reactions.

Wikis have a particular capacity to evoke critical reflections, because they make it extremely easy for every user to implement changes and commentaries, while keeping everything that was ever written ready for retrieval.
Thus, the Wikipedia invites me to read articles with the critical eye of a potential corrector because I have the permission to make changes whenever I see the need. In addition, I may feel invited to post my opinion on the appending “talk page” where there is lively discussion about controversial issues (on a terminological, conceptual or substantive level).

On a more general level, the Wikipedia catalyzes reflection because millions of users contribute to a very wide range of opinions and preferences – engendering controversies on the primary level of substantive knowledge as well as on the metalevel of procedural norms.

For instance, highly sophisticated discussions about the neutrality principle are constantly going on: giving rise to a heightened awareness of all the subtle, implicit ways it can be violated (e. g. by using insinuating “weasel words”).

This reflexivity is particularly manifested in the human sciences where there are many scientific concepts that have an intrinsic ideological bias because they have been created and elaborated by people sharing a particular (e. g. political) view.

For instance, this is the case for the term "Right Wing Authoritarianism" that has come under fire by conservatives who claimed that it has an intrinsic leftist bias:

In such cases, it is very helpful that in the article's heading, it is indicated that a discussion about its neutrality has arisen: so that readers get sensitized to these problems of which they otherwise would not be aware.
Such examples illustrate to what degree the Wikipedia has the potential to internalize dissensus and conflict instead of communicating a fictious impression of universal agreement. While conventional encyclopedias support the notion of a canon of "unquestioned truth" (by simply leaving out dissensual views), the Wikipedia is open to reflect any kind of manifest dissensus - thus submitting all truths to a much harsher test of acceptance.

By its mere lack of reliability, the Wikipedia demands mature recipients that are capable of receiving information while at the same time preserving a critical attitude: motivating them to corroborate the information by consulting additional sources.

A critical stance is particularly encouraged in cases where an article is highlighted as being “controversial” (e. g. about Taliban, homosexuality etc.): so that users know in advance that they have to rely on their own judgment, instead of absorbing a nonpartisan, “absolutely neutral” point of view.

Evidently, all these possibilities for personal participation provide ample opportunities to solve tensions and conflicts in smooth, inconspicuous ways. If I disagree with an entry in a conventional printed encyclopedia, I have no alternative than to protest harshly or even sue the editors legally. When the same happens with a Wikipedia entry, I have many other less offensive options: correcting the entry myself or writing to admins that it should be corrected.
Similarly, when articles are of low quality and transport erroneous information, this may not be a reason for denouncing the whole enterprise and for changing to alternative encyclopedias, but just engender the motivation to contribute personally to an improvement.

We all know to what degree all major cultural achievements are factually the products of widespread and enduring collective efforts. Thus, the evolution of law has been promoted by a multitude of infinitesimal contributions like court judgments, legal commentaries or academic opinions; and advanced technologies are the results of manifold improvements enacted by anonymous engineers and technicians.

However, we usually see just the final products, while the production processes remain hidden: inaccessible for analysis as well as for deliberate control. Thus, when Berger/Luckmann follow Husserl and Schütz in characterizing empirical reality as an “intersubjective construction”, they just indicate at the result without unveiling the underlying processes that have lead to it –so that it remains unclear who has participated to what extent, and whether the said processes could have let to alternative results (Berger/Luckmann 1999: passim).

The word “tradition” is usually applied to such past legacies in which we find ourselves embedded like in natural biotopes, unable to know why and how they have come into existence and unable to determine their further development in the future.

Seen under this perspective, the online productions in Peer-to-Peer networks are innovative in the sense that they make cultural production processes explicit and completely visible to all interested eyes. They share with “traditions” the basic feature that the products of collective endeavours degrade individual authors to the modest role of mere “contributors”. But unlike “traditions”, these contributions can be identified, regulated, modified or reverted at will, and the system of rules under which these contributions generated can be explicitly stated and systematized as well as changed by specified authorities and transparent formal procedures.

All encyclopedias must find ways that the information they convey is accepted as "authoritative": in the sense that normal recipients believe that it is reliable and that it represents the most advanced state of knowledge available at the current time.

In the printing age, there was no alternative than to rely on indirect authority of personal credentials: The authority of the encyclopedic knowledge was derived from the high reputation of its contributors: e. g, their academic degrees, Nobel prices, etc.

Of course, this implied a high trust in the formal institutions responsible for distributing such credentials: e.g. in the quality of academic education and certification.

By contrast, the Wikipedia can make itself independent from such derived authority sources because it is able to produce its own primary authority which emerges from collective online interaction. In other words: Wikipedia articles are not trustworthy to the degree they stem from reputated scholars, but to he extent that are the (preliminary) end product of all the preceding edits and discussions to which many collaborators with different perspectives and knowledge background have contributed. Why do these procedures make knowledge authoritative? Because they have been going on in public light and have been stored in a fashion that they can be recapitulated by anybody anytime: at present and in the future.

Thus, the Wikipedia exemplifies Luhmann's hypothesis that in modern societies, traditional legitimation is replaced by "procedural legitimation" ("Legitimation durch Verfahren") (Luhmann 1968). (Another example is the evolution is the political system where modern law derives its authority no longer from tradition or the charisma of a founding leader, but from widely accepted and transparent of law-enacting procedures (e. g. citizen votings or parliamentary decisions)).

From a functional point of view, this provides the basis for flexible self-correction processes that enable the Wikipedia to cope successfully with a wide range of exogenous and endogenous disturbances.

Social systems can be classified according to they way they deal with events that may threaten their essential structures and functioning's. At the one extreme point, there are “resistant systems” that defend themselves against disturbances by preventing their emergence (by suppressive activities) or their intrusion (by boundary controls and filtering). At the other extreme, we find “resilient systems” that allow any disturbances to enter, but then mobilize self-correcting activities in order to eliminate them in due time or to make them compatible with their own structures and goals

Conventional encyclopedias are clearly “resistant systems” that emerge in the context of formalized and centralized organization. By applying highly selective methods of recruitment, bureaucratic rules and permanent supervisory controls, they take care that from the onset, no deviant productions caused by dilettantism or intentional vandalism are generated. Such unbending discipline is all the more important as texts are finally frozen on paper, so that no corrections can be made ex post. Processes of improvement and growth typically take the form of discrete major steps (e. g. "editions"): each of which characterized by a multitude of smaller changes (or even a major change in the work's architecture).

By contrast, Wikis develop continuously over a very large number of minor revisions, so that users may find a slightly modified version at every moment of consultation.

Thus, they remain forever in the unfinished stage of "Perpetual Beta" (Tim O'Reilly): by inviting users of any specific article have to adopt an attitude of "critical acceptance" by synthesizing two contradictory expectations at the same time: that the information offered is basically correct and useful, but still so incomplete and faulty that corrections, improvements and updates are needed (O'Reilly 2005). In the case of commercial goods or services, such a philosophy of "continuous improvement" is difficult to adopt, because customers find themselves at a loss when they try to gather sufficient information about the products' current quality (and corresponding price).

Thus, the Wikipedia exemplifies the resilienttype system that remains permanently vulnerable to all kinds of disturbances, so that the maintenance of order is completely dependent on the selfcorrecting activities that set in after they have intruded. The way it works is by having a large number of people who keep track of recent changes, often through watch lists, which notify the user whenever a page they have marked has been edited.

As all the subsequent versions of an article are stored in the “page history”, it is technically extremely easy for anybody to cancel any recent changes by just restoring an older version. This feature results in a “conservative bias” which is of course functional for fighting vandalists or fierce ideological crusaders, but which may also discourage new contributions (because of the fear that even very laborious contributions are just wiped out).

In a study of the page histories of Wikipedia's English language version, MIT and IBM researchers Viégas, Wattenberg, and Dave have demonstrated that most Wikipedia vandalizations are usually corrected within very short time (a few minutes), so they will escape the notice of most users (Waldman 2004).

The efficiency in dealing with vandalism demonstrates vividly that Wikipedians constitute a tight community – despite the large geographical distances and very weak personal ties among the members:

Resilience implies that at any given moment, the system looks somewhat degraded or even chaotic, because it contains a certain number of (yet) uncorrected errors, Of course, no such deficiencies are tolerable in cases where information has to absolutely reliable because highly consequential actions are based on them (e. g. timetables, price lists, legal codes, telephone directories etc.).

On the other hand, resilience provides flexibility and openness for innovation, because systems remain free to decide which of the intrusions have to be treated as negative disturbances to be eliminated, and which as should be seen as enriching “innovations” that should be kept (or even subject to further elaboration).

Generally, resilience implies that the system’s internal order is permanently dependent on a high level of supervisory and correcting activity exerted by large numbers of participants, and on a rather modest flow of disturbances, so that the work load of the controllers is not too high. Thus, the Wikipedia will always have to be embedded in a vibrant “Wiki-community” where the basic values as well as the detailed operational rules of the system are consciously upheld, transmitted and incessantly concretized in specific actions. Resistant systems, on the other hand, can often go along with rather simple and rigidly frozen organizational structures, because as they repress or filter out so many contingencies, they face a much less complex environment. These ongoing adjustments find expression in a rapidly expanding layer of “meta-communication”: encompassing “particularly all the “talk pages” where the primary content of the articles becomes the topic for reflection and controversial discourse. Viégas et. al have found that that these talk pages and the additional “meta pages” (dedicated to matters of coordination and administration) have experienced a disproportional growth. In the time period considered, their quantitative share in the whole text system having increased from 15% to 30% (until October 2005) (Schiff 2006).

Whenever a topic is controversial (e. g. because of ideological and emotional reasons), a transitory period of “irrational” postings characterized by extreme opinions can be observed, before more objective, neutral formulations (which then remain stable) take the lead. Sometimes, fierce “edit wars” are engendered between participants who permanently erase each others version. For setting limits to such escalations, the 3RR rule was established: forbidding any single user to enact more than three reversions of a page within 24 hours (except in cases of manifest vandalism). [57] In addition, a temporal protection of a page can be requested in order to cool down heated editorial warfare. [58] Such temporal measures are highly effective because most edit wars is associated with current public moods and discussions that rapidly fad away when other topics come up. The more controversial a topic, the longer is the time period during which users may be confronted with rather one-sided, opinionated entries. But in the longer run, emotions tend to cool out, so that extremist passages are weeded out and substituted by more neutral formulations in accordance with the official “Neutral Point of View” (NPOV).

Mechanisms of resilient self-corrections are highly functional for dealing with smaller, decentralized problem cases that can easily be handled by the voluntary patrollers.

However, they reach limits in cases of sudden massive disturbances that may lead to a “work overload” of these policing members. In such cases, resilience has at least partially to be substituted by defensive resistance measures, so that intrusions are blocked before they enter the system.

Such a situation occurred at August 1st 2006, when the American Comedian and Satirist Stephen Colbert told his viewers to update the Wikipedia article “Elephant* in order to include the information that “the population of African elephants has tripled within the last three months.” After this broadcast, dozens of viewers crowded to the Wikipedia site in order to insert this addition, while policing users quickly got equally active for permanently reverting such massive vandalizations. Very soon, administrators exerted their competence to semi-protect the page: making it temporarily impossible for any unregistered and new users to implement changes. As even registered users continued to insert the misinformation, the site then was momentarily completely immunized against changes by setting it under “full protection.”

This example illustrates that in contrast to their printed predecessors, digital encyclopedias can combine resilience and resistance in highly variable ways: e. g. by limiting protection to particular pages, user categories and/or specific spans of time. Of course, the deliberations on such decisions is also adding to the hypertrophic overhead of “meta-discussions” as well as to the neverending expansion of formalized procedures and rules.

The WP relies on a complex process of "cultural darwinism" [59]: which is based on the functional interplay between three mechanisms:

1) Production of variation:generated by broad base of users who produce a large pool of memetic variants: by creating new articles, inserting additional information and proposing alternative formulations.

provided by collaborators (including admins and bureaucrats) busy to scan and filter all these new entries in order to weep out nonsense and to keep the WP's evolution in line with specific standards.

based on a third layer of activities preventing and reverting cancellations and vandalisms, so that the acquired quality level of the WP is maintained.

The speed and direction of evolution depends heavily on the working of these three sets of mechanism and on the specific may they are combined. For instance, too high production rates in innovative variations will bring even highly efficient filtering mechanisms to the limits of their capacity, and on the other hand, variant production may easily shrink when collaborators see that most of their contributions are constantly weeded out.

We may safely contend that the open source model of the Wikipedia favours variation over selection and stabilization, because inn decentralized peer-to-peer networks, there are no hierarchical agencies deciding about right and wrong, effectiveness and uselessness or falsity and truth.

Instead, such authoritative decisions have to be substituted by horizontal control processes among the collaborators: preferably guided by similar norms of universalism, communism, disinterestedness and “organized scepticism” as they (should) reign in idealtype scientific communities (Merton 1942).

Of course, in the case of highly specialized entries where the number of experts and visitors is very small, simple lack of manifest dissensus will not be a sufficient indicator that consensus has been reached: because even major insufficiencies and flaws can persist for long time spans when nobody takes notice or motivated to make any additions. The higher the user activity, however, the more the assumption is justified that lacking criticism indicates that “everybody” (or at least: many visitors with very different viewpoints) actually agrees – or that some disagree so little that they don’t find it worthwhile to articulate dissent or make corrections.

As we can learn from successful scientific or technical communities, such horizontal peer exchanges are most functional when all members can easily agree whether a contribution made is valuable, a specific problem has been solved, or particular goal has been achieved, because the outcomes can be objectively assessed and evaluated. This is certainly the case in open source software production projects (e. g. Apache or Linux) where any piece of proposed code can immediately be tested whether it is functional or not. Under such conditions, no hierarchical evaluations and authoritative selection processes are necessary because successes and failures stand out objectively, so that they can easily be verified and corrected by any member of the community.

It is evident that in open source encyclopedias, such preconditions are not persistently fulfilled. For sure, there are many contributions whose truth or falsity can easily be assessed, because they relate to highly indisputable, objective facts, natural laws or mathematical-logical operations. Here, errors may become rapidly eliminated because whenever a correction “to the better” has been made, nobody has any sound reason to return to the earlier version.

However, many contributions are “arguable” in the sense that they rely on viewpoints, opinions and evaluations that vary between the contributors as well as between the sources on which they rely. In such cases, the return to hierarchical controls may be inevitable in order to end “edit wars” that would never end by themselves because there is no objective test for adequacy or truth (Schiff 2006).

The idea of a Wikipedia would be particularly displaced if a “constructivist” epistemology is maintained: because this would mean that instead of general theories competing for universal recognition (in a Popperian sense), there are only co-existing “narratives” which are consensually accepted only within confined and transitory "discourse communities".

The most adequate epistemology for the Wikipedia is evidently an objectivist paradigm of truth: the belief that knowledge about everything can reach a definitive form on which all reasonable human beings can (or even must) agree. It’s no surprise that Jimmy Wales clings firmly to an objectivist understanding of knowledge which gives him the confidence that contributions finally converge in the approximation to a definitive intersubjective and intercultural truth [60]. Contrary to most contemporary epistemological philosophers, true Wikipedians believe in an absolute aperspectively constructed truth existing beyond all cleavages of particularistic and idiosyncratic human opinions and convictions:

While the WP shares this premise with traditional encyclopedias, it contrasts sharply by following not a deductive, but a highly inductive way of objectification.

Printed encyclopedias have an affinity toward deductive processes of reasoning and classification because their top-down organization makes it necessary to begin with blue print knowledge structures which then are filled out by the different contributors. In natural science, for instance the editing committee typically relies on highly accepted taxonomic systems, so that specialists can be searched and invited to deliver contributions about specific chemical elements, or about different, orders and genera and species of animals and plants.

By functioning as ex ante premises of encyclopedia organization, such conventional conceptual frameworks are reinforced rather than called into question - because scholars that maintain deviant concepts and typologies will not be invited.

A most outstanding example for this deductive top-down conceptualization is the Propaedia that came with the 15th edition of the Encyclopedia Britannica (in 1974): a 1000 pages book offering an extremely detailed outlay of all spheres of human knowledge by classifying it into ten major spheres and by disaggregating each sphere on seven hierarchical levels [61]. It may be considered one of the most conservative books in recent history: because whoever uses it has no alternative than to let his searching activities guide tightly by these authoritative conceptual schemes.

Wiki-based encyclopedias certainly also cling to these pre-existent conceptual structures, because most collaborators identify with them, and because editor use them for channelling incoming contributions (e. g. by creating "stub"-articles about concepts that deserve a more elaborate treatment).

In addition, however, they have an intrinsic leaning toward inductive conceptualizations that are arising out of an uncoordinated multitude of independent proposals. Such "folksonomies" are characterized by a more prototypic than categoric way of categorization: so that imprecise and overlapping interpretations and attributions may occur.

Such inductive terminologies have the advantage that they remain open for flexible innovation - due to the rise of new phenomena or the change of relevant differentiations (e. g. when new, cultural fashions (like music styles, or art forms) or unprecedented ideological or religious movements arise).

On the other hand, they have extreme shortcomings because their usage remains basically restricted to the collectivities that have produced them, and they remain ambiguous (e. g. because often several different meanings are given to the same terms).

In the wide areas (like politics, ideologies and religion) where objective truth can never be attained, the Wikipedia tries achieve consensus by clinging to the “Neutral Point of View” (NPOV): one of the thee highest-ranking guiding principles of the official Wikipedia policy that is defined to be immutable even if all editors would agree on a modification. [63]

By aiming at a “neutral point of view”, WP envisages an optimistic belief in the possibility of reaching at least a minimum universal canon of human knowledge that is accepted consensually by all “rational human subjects”, because it cannot be meaningfully refuted.

In the tradition of rationalistic strands of philosophical thinking (Leibniz, Kant and Habermas), it is supposed that there are highest level principles of “formal reason” on which all human subjects – irrespective of any divergences on any “material” questions – may voluntarily agree.

In a multicultural world, such a consensus about evident truth can evidently most often not be reached on the primary level of substantive evaluations or empirical facts, but only on the secondary formal level: on the assertion that there exist people who hold certain principles for valid or who hold certain facts to be true.

Thus, only noncontroversial topics can be treated on a primary level (=discussion of facts); all controversies have the effect that a topic can only be discussed on a meta-level: representing “fairly” all the different positions and beliefs. (Sanger 2001). [66]

This statement clearly demonstrates how difficult it may be to avoid all perspectivism even on subtle, inexplicit levels of textual structuring and linguistic expression.

For instance, the mere sequence in which positions are represented (or the volume of text allocated to them) implies decisions which are most certainly guided by subjective preferences. Similarly, authors will reveal their subjective opinions in innumerable other ways: e. g. by characterizing various positions as “popular”, “sectarian” or “empirically founded” views, or by focussing content ethnocentrically on their own nation and culture (Sanger 2001).

If it is difficult to describe an empirical fact or development fairly, why should it be less difficult to describe disputes about such facts or developments in fair, objective terms? Can any contributor be expected to have full knowledge about any such dispute and about the number and quality of its supporters (even within a small time span and geographical area), especially in cases where they have been shaped by many scientists and intellectuals with very different positions? As Sanger states, it is useful to treat this as an empirical not as a philosophical question. It can be answered affirmatively in all cases where articles have reached a stage where they are factually accepted (=not generating any additional controversial discussions) (Sanger 2001).

In fact, however, such highly relativistic principles are not fully maintained in the Wikipedia, because in most cases, the positions that claim “scientific” validity are privileged in relation to “sectarian” exotic positions (even when these would have a higher absolute number of believers). (e. g. Darwinist evolution theory is taken much more serious than creationist views). If this “unity of scientific doctrine” would be abandoned, the Wikipedia would degenerate into a universe of ethnographic narratives that would have to give room to all endogenous cultures and all (even highly exotic) minorities of dissident believers.

While the strategy of representing different viewpoints or theories cannot be stretched to include every possible position maintained by any individual or tiny group, it can nevertheless be applied in order to end “edit wars” between those positions over which there is an actual fight between different Wiki contributors.
Thus, we arrive at the conclusion that the “truth” developed in the Wiki process merely represents a reconciliation between positions actively maintained by online editors: just a “truce” between adversaries who have decided to end edit wars because they all find their own different views adequately represented – or because they have become just too tired to fight on.

As current events and developments (discussed in the media) are most likely to engender heated debates, an effective measure to deescalate conflicts may be called the "strategy of deactualization".

For instance, there was much debate about articles called "Persecution by Christians" (Muslims or Jews), and votings nearly resulted in their deleting. However, these pages were kept, but partially neutralized by being renamed into "Historical persecution of Christians" [68] (or Muslims [69]): in order to avoid overt conflicts about current events.(Unsurprisingly, an even stronger measure of neutralization was implemented in the case of Jews, by renaming the entry into "Ancient historical persecutions by Jews" [70]).

Evidently, the Wikipedia invites us to see the process of human knowledge production as a process of Darwinian "memetic evolution" [71]. The cognitive patterns fittest for survival are those maintained by strong, highly articulate individuals or groupings motivated and able to defend their views successfully in “edit wars”. If they are completely victorious, they may be able to define their opinion as the only “scientifically founded position”: so that alternative positions receive much less (or even no) explicit recognition.

Thus, the Wikipedia is exposed to the same critical arguments as they were directed by the ancient Greek sophists against any consensualist theory of truth:

The problem arises from the fact that whenever there is a memetic competition, it is highly probable that the engagement of the different sides is not equal in strength.

For instance, religious believers may be extremely determined that the entry on their founder does not contain any "negative" biographic information, while all the outsiders may have very little interest in this whole matter.

As a consequence, the believer's zeal to keep the article "clean" is not counteracted by a similar effort of nonbelievers to keep it in accordance with the standards of the "Neutral Point of view".

It is evident that the Wikipedia has to rely very much on widespread groups of liberal nonbelievers that are ready to fight for their Western standards of tolerance, openness and objectivity with the same fervour and zeal religious fundamentalists defend their dogmatic beliefs.

Evidently, this implies an openness toward multiple and changing viewpoints that is not consistent with closed dogmatic belief systems as they are maintained by Islamists or other adherents of religious fundamentalism.
It's no surprise therefore that such medieval minds feel threatened by an intellectual enterprise in which they see no chance to dominate and to eradicate unwelcome "dissident" views.

This position is well formulated in an essay of Abid Uallah Jan who criticises that in he WP article on Islamism, "cultists" like Ahmadis, Habashis and Ismaelis are considered to be Muslims despite the fact that in contrast to "True Moslems", "...they do not believe in the totality of the Qur’an and the finality of the Prophethood". (Abid Ullah Jan 2006a)

A WP editor has responded that these groups are considered to be Moslems because they themselves maintain such an identification.

Abid Uallah Jan's essay makes it evident that from an Islamist point of view, the Wikipedia is a particularly effective weapon in the War of "Islamophobes against islam" because is contains innumerable formulations that appear faulty, inimical or even blasphemic from a strictly fundamentalist perspective: statements hard to fight against because they stem from so many different (and mostly anonymous) sources:

Of course, trying to synthesize a “neutral assessment” is in itself an authoritarian endeavour because all other (e. g. monographic) representations are implicitly degraded as one-sided and ethnocentric, as they have not passed through this elaborate process of synthesis and purification.

While the “neutral article” occupies the center place of attention, all these more subjective or ethnocentric articulations are marginalized by being diverted to the collateral “discussion page” where controversies can go on that may later have visible impacts on the article itself. These “talk pages” are the very for a where memetic evolution processes go on and where everybody can observe how “reality” is constructed as an emerging result of free intersubjective communication. Such constructive endeavours are particularly prominent in the case of unprecedent new unfolding events or developments, where fundamental problems of conceptualization have to be solved.

This was vividly illustrated in the entry “Israeli-Lebanon conflict" in Summer 2006. While an impressively equilibrated exposition has soon be realized as a result of 9000 edits (between July 12th and July 29th), extensive controversies about very subtle terminological points were fought out on the parallel discussion page: whether the process described should be named “conflict” or “war”, or whether Israel soldiers have been “captured”, “kidnapped” or “abducted”.

Some articles may even become temporarily protected from editing until fundamental disputes have been resolved. For instance, the article of “New anti-Semitism” was frozen by administrators in May and June 2006 “until disputes on the talk page have been resolved”. [75] The controversy resulted from the fact that he concept “new antisemitism” is used by rightists for defamating the political left: by attributing them a generalized new tendency to take sides against Israel (and even worse: sympathizing with blatantly antijudaist moslems). Leaving the page unprotected would have resulted in a permanent edit fight between rightist who want to upheld this attribution and leftist liberals who deny the justification of the term because they want to draw a clear division line between decrying Israel and defamating the Jews.

While this controversy cannot be avoided of course, it is dealt with in a deescalating manner by diverting it to the discussion page associated with the article.

In this particular instance, however, protection was lifted after two months without that the conflict has been settled by discussion. Instead, some steam has been removed in the meantime because the leftist opponents to the page have founded a “revenge page” about Israel's alleged “Apartheid” policy. [77]

From such examples, we may draw the unsurprising conclusion that like the UN and other global institutions, the Wikipedia cannot expected to solve persistent global conflicts, but at best to offer some new opportunities for extensive discourse and sophisticated verbal clarification.

In contrast to printed encyclopedias, the total volume of the Wikipedia is not limited by physical and economic factors. Nevertheless, in proportion to the huge number of edits the WP shows rather modest rates of growth, many new articles are quickly eliminated by admins who think that the topic is "not notable" enough to be included in a repository of universal knowledge; and many enlargements of existing articles become quickly "reverted" because information is judged to be too trivial or beside the point.

The problem with this filtering is that it is not guided by any consensual explicit rules and not executed by a clearly defined decision making body. Anybody can post a request that a specific article should be cancelled, and for any type of intransparent reasons, it may occur that a "majority" for such an action can be found.

Given the rising significance and popularity of the WP as a reference source of information, such filterings become increasingly important because a Wikipedia article may soon be considered as an indicator of relevance, eminence, popularity and reputation - for persons as well as for music bands, art works, localities, historical events and any kind of voluntary association.

Currently, such decisions are guided by a multitude of informal criteria that primarily reflect the personal values and preference of the "Wiki mandarins" (mostly between 20 and 30) because they have never been submitted to a public voting or any other legitimating procedure.

It is no surprise that the WP leadership is often inundated by protest emails from the "victims" of such harsh elimination procedures - users who do not know about these rules or who do not agree with them.

Of course such elimination strategies may promote the installation and growth of provincial "minority language Wikipedias" because they provide an at least small forum for many "domestic" personalities and topics that have no chance of being considered in the global English edition.

In the future, we will certainly see much more conflictive action concerning the "rules of notability" as well as on the admission or omission of particular entries. This "politicization" of exclusion/inclusion will certainly raise the need to clarify selective criteria and rules - as well as the procedures dedicated to their constitution, change and specific applications.

Such processes will of course be facilitated by the fact that filtering takes place in full public light. For instance, everybody can consult the daily lists of articles nominated for deletion.

For the first time in history, a broad open discussion about "encyclopedia notability" has been started that has already given rise to intensive debates and detailed - while still unfinished and unofficial - lists of possible criteria. In the guideline page dedicated to the notability of people, for instance, it is stated that among others. Persons with the following characteristics should be included:

Such sentences -like many others - illustrate that the Wikipedia sees itself as a publication that relies on reputation that has already been produced ex ante: especially when it is based on consensual mass media judgment or - in the case of lesser known individuals - on different smaller, but mutually independent sources. Of course, this policy does not acknowledge that a Wikipedia entry may itself become a factor in reputation building: especially when the information that this entry exists is propagated by journalists and other potent "multiplicators".

The Wikipedia is an extremely comprehensive object to study, because it is at the same time

Given its amazing complexity and volatility as a product as well as a production process and organizational structure, it is difficult to achieve any definitive assessment whether it is currently approximating, equalizing or even surpassing conventional encyclopedias on any criteria of quality, or whether it has any chances to continue its spectacular growth (or at least survive on the present level) in the near and more distant future.

In a least controversial functionalist view, nobody will deny that the Internet offers a technological platform particularly instrumental for very large scale collective publication projects, so that the old idea of producing a universal encyclopedia seems better realizable than in any earlier periods of history.

Since he initiated his project in January 2001, Wikipedia founder Jimmy Wales has gone a long way to realize his bold promise to “distribute a free encyclopedia to every single person on the planet in their own language”. In the meantime, about 5 million articles in more than 150 languages have been created, and the number of visitors is currently (November 2006) higher than that of any other non-commercial site.

More than that: the Wikipedia has grown not only to be one of the most popular web platforms, but also one of the most authoritative Net Institutions which is daily consulted by thousands of students, teachers, journalists and others who multiply WP knowledge orally or by writing to many other receivers.

This trend is supported by the exploding mass of web information sources that causes most surfers to reduce complexity by confining their regular surfing to about eight to ten Web sites (the equivalent of "anchors" in shopping malls) which they deem reliable, timely, accurate, objective, authoritative, and credible.

Many of these visitors may not be aware that the Wikipedia is the product of anarchic and amateurish procedures; they fully trust the information they find, and they are careless (or lazy) enough to consult additional corroborating sources. As a consequence, the Wikipedia has ever more influence on worldwide processes of knowledge acquisition and knowledge diffusion. Unquestionably, it determines the information transported by innumerable academic papers, magazine articles, written memoranda and oral talks and lectures all over the world.

Given all these striking measures of success (and indicators of unimpeded further growth), there are still reasons for doubt whether the whole project is sustainable because with increasing size and societal prominence, it may become more manifest that it is built on rather shaky grounds.

First of all, the breathtaking popularity of WP contrasts sharply with the fact that it has no secure basis for trust. Its rising status as a first order web knowledge resource site is somewhat free-floating, because there is no correlative emergence of actors to which such far-reaching responsibilities could be attributed: no individuals nor collective bodies that could be made accountable for the information existing or lacking in this amorphous heap of collective contributions (Brandt 2006).

Somewhat similar to democratic votings, the resulting articles have to be seen as the products of anonymous collective processes that derive their legitimacy and acceptance form the fact that a set of unknown participants have come to a certain (at least majority) agreement.

Not only is there any lack of professional expertise as a source of authority: users must live with the suspicion that any page they visit has been vandalized recently or is the product of completely uniformed authors. As anybody can edit and modify anything, even people maintaining highly optimistic views about human nature will not be ready to trust fully any article or bit of information, Thus, the Wikipedia is constantly accused of being unreliable, or even more strongly – being just a garbage can filled with trivia and trash.

This lack of trust has grave behavioural consequences, because for several reasons, the Wikipedia is more disposed than conventional encyclopedias disposed to be heavily criticised form many sides:

Since its inception, the Wikipedia is vehemently denounced by individuals who base their judgement not on extensive empirical research, but just on deductive common sense arguments: As everybody can edit and change articles, there must be a high level of vandalism and misinformation; as nobody is paid for fact-checking, it is certain that errors remain uncorrected; as experts face the risks that their contributions are subsequently modified or erased by laymen, their motivation to collaborate will inevitably be reduced to zero, as nobody can be made liable and legally sought, slandering will spread without limits.

Of course, such deductive arguments abound because it much more cumbersome to base judgment on inductive procedures: by selecting a representative sample of Wikipedia articles and analyze to what degree they meet standards of quality, consistency and reliable truth.

While this unprotected exposure is a source of vulnerability, it is on the other hand also an excellent precondition for further learning processes and evolution: e. g. for developing norms and organizational procedures in order to raise the level of linguistic expression and the reliability of information.

The problem to be solved is the following: which minimal measures of access control, hierarchical supervision and professional expertise are necessary in order to wipe out vandalism and errors and to ensure reliable, high quality contributions? Instead of relying on a thin elite of professional authors and editors from the onset, the Wikipedia has begun with an extremely open structure which of course can be modified ad libitum according to emerging needs. Its open-ended evolution is based on similar principles as the liberal state where the problem is to find out which minimal constraints on the citizen’s freedom are indispensable in order to prevent public disorder.

It would seem very reasonable to raise the trust in Wikipedia entries by aggregating user judgments: either judgments of experts who evaluate entries within the specialized fields, or general user judgments as it is done in many other Web 2,0 sites today.

Paradoxically, the Wikipedia doesn't lend itself well to such procedures, because any aggregation of judgments has to rely on the premise that the object to be judged remains invariant over time.

The WP's openness for modification has not only the consequence that every user may meet a different article, but that judgments themselves may cause such changes: to the degree that judges correct themselves immediately the errors they see. This second consequence could mean: the larger the number of judgments, the less useful the aggregated judgment, because the object to which it refers has considerably changed.

In a way, the Wikipedia resembles tiny physical quantum objects in the sense that it cannot be observed because observations themselves are causing it to change. Thus, journalists may not find it fruitful to write critical articles about WP on the basis of major errors they have found in it, because only hours after publication, these same errors may have already been eliminated.

A second vulnerability stems from the rising eagerness of individuals and organizations to manipulate the Wikipedias contents in accordance with their interests and preferences. The higher the popularity and reference status of the site, the less a politician can ignore when his biography contains embarrassing and compromising facts, and hundreds of supporters, election contest managers and “media advisers” may become active to “correct” the corresponding entries.

Likewise, every corporation will care that its economic performance and the way it treats its employees and customers will be described in a favourable way, and it will mobilize its public relations specialists to do the necessary job. This inherent danger is illustration by the start of “MyWikiBiz.com” in August 2006: a firm who offers to all companies the service of authoring Wikipedia articles about their enterprise and their operations.

While living persons, active organizations and contemporary events may be most hit by such massive interventions, even historical articles (e. g,. about the dead founders of still living religions) may become the center of heated editing contests. Thus, the “resilient” capacities of the Wikipedia may be more profoundly tested in the future, and ever higher numbers of highly motivated and activated “true Wikipedians” may be necessary to cope with such collective attempts of manipulation. While straightforward “vandalizations” often stand out so clearly that they are easily corrected (sometimes even by automated "Vandalbots" without human intervention), such manipulations may be much more difficult to discover, because only very few “patrollers” have the respective knowledge.

Somewhat different dangers arise from the inherent tendencies of “self-accelerating growth". The more popular the Wikipedia, the more individuals will develop an interest to find themselves and their acquaintances as well as their home village and high school and their most preferred movies and music bands adequately represented. Thus, universal encyclopedic knowledge will give way to “multi-particularistic” knowledge serving idiosyncratic interests of families, localities or sectarian movements.

More and more, organized attempts may be made to instrumentalize the Wikipedia for purposes of “self presentation” or to even “kidnap” it for specific ideologies or propaganda purposes.

For instance, in July 2006 the Akron Beacon Journal in Ohio has published an article where readers are invited to write additional Wikipedia entries related to the history of the city of Akron; and detailed technical instructions are provided how articles are generated, edited and changed.

Sometimes, even competitive races are unleashed that may lead to uncontrolled self-escalating editing endeavours:

Such collective “assaults” could well lead to a highly disequilibrated coverage of different geographic regions and locations, and it is not clear how such one-sided hypertrophies could be held in check.

Evidently, they can only be counteracted by cultivating strong, highly explicit and consensual views within the “Wikipedia community” about the scope and limits of “encyclopedic knowledge”: so that all contributions transcending such limits will be rapidly eliminated.

A third latent instability arises from the spectacular degree to which the whole project is based on a highly regularized flow of unpaid voluntary collaboration. Such volunteering may well encourage the creation of ever new articles, because many participants may be highly motivated to leave their personal footprints by adding something new. However, the more articles, the higher the subsequent volume of constant maintenance work that has to be carried out by the whole WP community.

The more the Wikipedian diversifies into millions of entries, the less it is possible to allocate the "watching capacities" in a way that all articles are permanently corrected within short time when vandalizations or other forms of degradation occur.

In fact, the Wikipedia community and the administrators maintain highly specific assumption about which sites are very likely to be attacked and which sites are highly important to keep clean. This explains why vandalizations of the G. W. Bush article usually don't survive longer than two minutes, because it is constantly patrolled by policing participants (Kelley 2005).

Usually, such maintenance work is much less motivating because authors find little room for creative performance. Consequently, the probability is very high that the Wikipedia process will soon be slowed down or stopped by simple fatigue; especially when alternative projects that allow more creative expression are absorbing the volunteer's attention.

Unhappily, the transition to fully paid staff is no viable alternative, because thousands of employees would be necessary to carry on all volunteering activities.

Therefore, stagnation and decline will only be prevented when active participation is stabilized by either by very tight internal community controls or by exogenous institutionalized norms. For instance, schools and universities could oblige their students to engage in “Wikipedia maintenance work” for acquiring some their points and grades; scholars may accept the informal responsibility to look constantly after the WP entries most akin to their specialized field; and even national or worldwide associations may emerge just for the purpose keeping “their” Wikipedia sections up to date.

In addition, the strict anonymity of contributions may in the long run be disfunctional because collaborators see no chances of getting any personal reputation (Ciffolilli 2003).As many articles are in their major parts written by single contributors (or very small groups of them), it would be possible to make at least these names visible - in contrast to all the smaller contributors who have only added words, commas, references or links.

Some of these problems are aggravated by the fact that the Wikipedia is not a “Net Encyclopedia” in its fullest sense, but an intermediary product that still clings to some premises and constraints of the printed paper era. When seen in isolation, it is certainly impressive how radical WP has implemented new online technologies in literally all its activities. When looked at as a component of the larger Internet, however, it is conspicuous that it has still problems to define its place and hesitates astonishingly to make full use of the potentialities of the World Wide Web.

Like a conventional multivolume encyclopedia that can be put on a library shelf (and like the EB or Encarta on CD-Rom), it still aspires to remain a relatively closed, self-contained universe: so that visitors have just to navigate within the site to find all necessary information. As explained above, this self-isolation may be understood as a correlate of community building and collective identity formation. It is expressed in the emphatic assertion that the WP should not be a “web directory”, and in an

“external hyperlink paranoia” (see 6.6) for keeping visitors away from propaganda or commercializations. However, this “isolationist” stance ignores that the WP is just a node within an ever expanding web of knowledge resource sites, and that unlike the community-oriented “Wikipedians” who want to perfect their mighty cathedral, typical visitors are quite indifferent whether they find the desired information within the Wikipedia or on any other accessible site.

The Internet makes it fundamentally easy to corroborate any kind of information by searching for second or third opinions in different websites. As a consequence, the idea of the WP to be right in all matters is fundamentally flawed: its a relic of the printing age where having the EB usually excluded the possession of alternative encyclopedias, so that there had to be complete trust in exactly this single publication. Thus, for the WP, the way to perfection does not mean to become error-free, but to make available gateways for corroborating information: e. g. by adding hyperlinks to more specialized and professionalized sources.

As it is used a portal site by so many users, it should accept its responsibility to be exactly such a gateway: by guiding users from the more fundamental information provided in its own articles to deeper and more detailed information on other sites.

By setting links to the primary sources from which it has drawn its information, errors would also become less consequential (and therefore: more tolerable), because users would be enabled to make independent checks (Benkler 2006: 218).

Of course, this would imply that Wikipedia editors accept the duty to evaluate and select such external sites: so that the Wikipedia would not just be an encyclopedia, but also an encompassing directory: a universal gateway to human knowledge by connecting to all sorts of high-quality informational resources. More than that: it would constantly adjust its mission in relation to complementary sources arising on the WWW: carving out an ever more specialized and more precisely defined niche.

Only by stripping off all aspirations of isolative self-sufficiency, the Wikipedia will burn its mental bridges to the old age of printing and become a true contemporary of the Internetted Digital Age.

Finally, we may speculate that the most profound effect of the WP is associated with a much more encompassing process it has set in motion: the rapidly proceeding "wikification" of the World Wide Web.

On the one hand, there has already been a rapid multiplication of Wikipedias in almost all human languages On the other hand: there is an emergence of specialized wikis centering on particular topics. Such processes have been catalyzed by the foundation of the "Wikicities" site which offers the free MediaWiki" software to everybody who wants to install his own Wiki: e. g. on Star Treck, Harry Potter, Basketball, genealogy or quit smoking.

As exemplified by "Beijingology" page which aims to collect all available knowledge on this major Chinese city, geographical entities like countries, provinces or municipalities may be particularly prone to become attractors for wiki-guided knowledge aggregation, because such knowledge is very multifaceted and distributed to large and constantly changing variety of residents, visitors and external observers.

While such proliferations may weaken the central encyclopedia endeavours by diluting work capacities on a multitude of smaller projects, they have themselves a centralizing impact: e g. by convincing former authors of individual websites to pool their endeavours.

Thus, the psychology Wiki founded in January 2006 has expanded so quickly that already at the end of the same year, it has become one of the most comprehensive psychology sources on the Net (with about 22000 pages).

While the general Wikipedia still functions as a model and paradigm, such specialist Wikis may have better chances for survival and continuous upgrading because most of their contributors may possess a rather high expertise. Starting in Jan. 2007, these services have been expanded by openserving.com which offers also free bandwidth and storage space to all Wiki holders.

It is evident that apart from the encyclopedic project, the Wikipedia has now kicked off a far-ranging process of "Wikification" that may easily spread over major parts of the Internet subsystem by giving rise to thousands of knowledge accumulation projects united by using the same standardized Wiki software as well as by dense mutual hyperlinking and uninhibited content transfers (based on "free license").

Abid Ulla Jan 2006b Wikipedia: Good Intentions, Horrible Consequences. Al-Jazeerah, Febr. 27,

Blau P. M. 1994 Inequality and Heterogeneity. A Primitive Theory of Social Structure. New York: Free Press.

Benkler, Yochai 2006, The Wealth of Networks, How Social production Transforms Markets and Freedom. Yale University Press, New Haven and London.

Berger, P. and Luckmann, T. 1967 Social construction of reality: a treatise in the sociology of knowledge. (Garden City, NY: Doubleday.

Berinstein, Paula 2006 Wikipedia and Britannica. the Kid's All Right (And So’s the Old Man). Information Today, March 2006.

Campbell, Donald.T. 1965 Variation and Selective Retention in socio-cultural Evolution (in: Barringer, Herbert R. et. al. [eds.] Social Change in Developing Areas. Schenkman Publishing Company Cambridge Mass. 1965, 19-48).

Gallupe, R. and McKeen , J. 1990 Enhancing computer-mediated communication: An experimental investigation into the use of a group decision support system for face-to-face versus remote meetings. In: Information and Management, 18, pp. 1-13.

Goffman, Erving 1959 The Presentation of Self in Everyday Life. Doubleday: Garden City, New York.

Kelley, Jeffrey, 2005 Wikipedia is always a work in progress. Richmond Times-Dispatch Sept. 3.

Kerr Elaine B. /Hiltz, Starr Roxanne 1982 Computer-Mediated Communication Systems. Status And Evaluation. Academic Press New York 1982.

Luhmann, Niklas 1968 Legitimation durch Verfahren. Westdeutscher Verlag, Opladen

McCarthy, John / Zald, Mayer N. 1977 Resource Mobilization and Social Movements: A Partial Theory (American Journal of Sociology 82: 1212-1241).

McCarthy, John D . / Zald, Mayer N. 1987 The Trend of Social Movements in America: Professionalization and Resource Mobilization (Zald, Mayer N. / McCarthy, John D. (eds.), Social Movements in an Organizational Society, Transaction Books, New Brunswick: 337-391).

Mehegan, David 2006 Bias, sabotage haunt Wikipedia's free world. Boston Globe, February 12, 2006

Merton, Robert K. 1942 The Normative Structure of Science. (In Storer, N.W. (1973) The Sociology of Science. Chicago: University of Chicago Press. pp. 267-278).

Munzel G. Felicitas 2003 Kant on moral education, or "enlightenment" and the liberal arts.
The Review of Metaphysics Jan. 9th 2003.

Patrick, Ewin 2005: An encyclopedia is an ever-evolving reference tool. Wisconsin State Journal, August 16, 2005.

Powell, Walter W. / DiMaggio Paul (eds.) 1991 The New Institutionalism in Organizational Analysis. Chicago and London: The University of Chicago Press.

Putnam, Robert D. 2000 Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.

Read, Brock 2006 Can Wikipedia Ever Make the Grade? (in: The Chronicle of Higher Education, October 27, 2006

Schoonhoven, Claudia Bird 1981 Problems with Contingency Theory: Testing Assumptions hidden within the Language of "Contingency Theory" (Administrative Science Quarterly 26, 349-377).

Viégas, Fernanda B. / Wattenberg Martin /Dave Kushal 2004 Studying Cooperation and Conflict between Authors with history flow Visualizations Proceedings of the SIGCHI conference on Human factors in computing systems Vienna, Austria, pp. 575 – 582. http://alumni.media.mit.edu/~fviegas/papers/history_flow.pdf

Yeo, Richard 2001 Encyclopaedic Visions. Scientific Dictionaries and Enlightenment Culture. New York: Cambridge University Press.

Zald, M.N./ McCarthy,J.D. (eds.) 1987 Social Movements in an Organizational Society. New Brunswick: Transaction Books.