WikiCooperationArticle

MeatballWiki | RecentChanges | Random Page | Indices | Categories

This is a kind of article I wrote for FranzNahrada to help him introduce people used to a mailing-list environment to the new wiki environment. It was meant for experienced and intellectual internet users, being wiki newcomers at the same time. Their main problem was to accept that others are allowed to edit "their" text. So I talked about typical situations, modes, roles and expectations ... to draw a big picture. Maybe it's useful for someone in a similar situation. This text is in PublicDomain. -- HelmutLeitner

Wiki Cooperation

Wiki cooperation. Cooperation in general and in wikis especially does not grow from a vacuum. You need to share a cultural context and common expectations to cooperate. Wiki is a new tool and grows a communication culture and expectations that are somewhat different from message-oriented communities or real life.

Experience and exploration. Starting in 1995, over the course of more then 10 years, many different things have been tried in various communities - successful and failing. It would be silly to ignore this experience and start from zero - building on the technological aspects of wiki and one's individual creativity alone. There are years of experience in existing traditions and values. On the other hand, tradition is not everything and we are just starting to understand wiki and online cooperation. Much more exploration is needed and the community will be happy to learn about new ways to cooperate.

Dissertation mode. The oldest tradition (from the original c2 wiki) is to use wiki to produce common consensual knowledge. The expectation is that a topical page is created with some good initial content and a name that fits it. Others join in and help to elaborated it by editing the same text. Sometimes this is called "dissertation mode", sometimes "document mode". During this production there is often the need for discussions. It is a tradition to use signed paragraphs and to put them below the dissertation as a discussion section ("thread mode"). This makes sense, because this is less long-term value than the consensual text at the top - you always put the most important things first. In theory, when the discussion has ended and everything has been reworked into the consensus, the discussion is deleted and the final dissertation (or document) remains, but that rarely happens. Sometimes the discussion is put on a separate "...Discussion" page - this makes no difference.

Many modes. "dissertation mode" is a useful mode of operation but it is only a special case. Let's consider some implications and deviating cases. It will show that a wiki offers a lot of different options to cooperate, as long the users understand what they are doing and share common expectations.

A side-note. While you will see that there are many modes of operation, dissertation mode and consensual knowledge is at the very heart of wiki culture. This means that people accept to work and reflect together and seek at least some consensus in process or knowledge. This isn't about "hey, what you wrote is great" but about "consider to change X to Y and sorry, I can't agree to F because of G, and why K?". To wiki outsiders this looks like negative criticism but it is the fastest way to make progress and the most positive way to show interest, be helpful and participate. Perhaps "criticism is/as feedback" is one of the biggest cultural gaps between the wiki world and outside.

Deviations from dissertation mode. At the bottom of (dissertation+discussion) is the assumption that users want to write a common text (this is not always the case) and that it is possible to agree on a consensus (this is not always possible). It is also necessary that a proper page name is chosen (because otherwise the content will not converge; the page must be moved or split) and that a sufficient introduction and starting content is written (people often don't do that). These are not really problems, but one has to handle them.

Partial consensus - MPOV. If people can't agree on a complete consensus during discussion then something very natural happens: there is partial consensus on top followed by the different opinions that weren't resolved. Usually this is no problem and hardly distinguishable from an unfinished consensus process. The only problem is how this is perceived: a "neutral point of view" or "consensual point of view" community may see this as an immature page, while a "multiple point of view" community may see this as a natural and valuable final state of a page that shows a spectrum of views and proves tolerance.

Publishing - article mode. Sometimes people don't want a consensus, they want to publish and maybe push their opinions (no consensus). This is not really a problem, if the wiki community is "multiple point of view" and the text is high-quality, marked as an article or essay and is properly signed and named. Then the community can (perhaps) accept a special opinion as a literary piece of work and refrain from reworking it into an consensus. It it clear that the author can't have authority about the page name, because a consensus text about the topic will always have priority. Feedback and discussion can happen as usual, perhaps on a separate "....Discussion" page, to let the article stand on its own. Let's call this "article mode".

Creativity - brainstorm mode. Sometimes contributors do not feel up to writing a high quality initial text but still feel the pressing need to start a topic to sort out their ideas or make progress with a problem. Communities looking for quality often dislike this very much. There is a way around this problem called "brainstorm mode": it is a mode that is used only for a few days, before being transformed into some other (typically dissertation) mode. First declare that you want to brainstorm for X days, write an introduction explaining your reasons, and start with all your thoughts put into tiny pieces without order. Invite participation. No analysis, no signatures to contributions, no criticism - like in real world brainstorming. There is no ownership in brainstorming. After the set time, order the contributions, write a proper introduction and transform the "idea mess" into a normal page. The point is, that it is guaranteed that the painful low-quality stage exists only for a short period of time. The initiator is responsible to resolve it and produce quality, or delete the page.

Signing - protecting personal communication. Wiki has done the simplest possible thing and adapted e-mail signing standards to wiki contributions ("-- " + real name after the contribution). A signature carries many meanings not talked about here (nymity, authorship, responsibility, context, ... ) but it is also a protection against change. A signature may be misused, if you put it to a link or a book-list, because this is neutral information waiting to be extended. Such signatures may be removed. But if you tell about your experiences or opinions, only the signature (hopefully connected to a sensible home-page) can provide necessary context, additional credibility, and a fallback communication path. Obviously it would make no sense to edit an opinion or change descriptions of experiences. Why would someone want to? It would be like forging communication (of course, typo corrections are welcome). The fate of a signed contribution may be to be deleted (if offensive, off-topic, redundant or the author agrees), moved or copied (if in the wrong place, or to be archived for re-factoring), but it is never, never, never, NEVER changed in a way the author might dislike. Each and every wiki community is responsible to look to that, without exceptions. It's THE "wiki crime" to forge communication, to put words into the mouth of people which they haven't said. It's such a taboo, so unthinkable, that people don't even talk about these standards. It isn't done. This makes perfect sense because contributors feel hurt and raped in such cases and turn their back immediately.

Author Home-page - context, communication, neutral space. Another useful wiki tradition is the creation of home-pages for contributors (authors). The signatures link to the home-pages, so that any reader can see who the author is, leave a question or hint for him at the home-page, use an existing e-mail-address for urgent communication. The home-page is a place to give personal background information that eases the interpretation of the work. The author may also place a picture there, talk about his interests, keep personal bookmarks or notes, define his goals in the local community, communicate with other contributor in thread mode, or even write an online diary. The concept of "off-topic" doesn't apply. So the home-page is a very free space (some call it a "front lawn") where others won't interfere. Only when the author as a whole isn't welcome in the community is this free space (neutral page) is in danger.

Just talking - forum page. Sometimes you want to talk about something (off-topic or not) but don't know where. Posting to home-pages is an option, but you can't be sure that people (other then the owner) will read postings there. The creation of new pages isn't always a good choice. So some wikis have created special pages for the purpose of such informal communication, called "forum", "coffee shop" or "water cooler". It seems people like that invention. You can talk about everything without the formal constraints that may rule topical pages. The advantage is, that off-topic content has a legitimate place and doesn't lead to an off-topic conflict. Contributions and threads are put under a date headers and deleted after some weeks or months. See MeatballWiki:MessageBoard and MeatballWiki:BlackBoardPattern for more details.

Reworking content - re-factoring mode. The most difficult task in wikis is to rewrite a page, replacing any existing content or thread chaos, without losing technical or social information, without creating conflicts. The name "re-factoring" comes from software development and means to simplify a system without losing functionality. This can be compared the mathematical concept of transforming systems of equations without changing their content. How is it done? First you must make sure that discussions have stopped, for heated discussion can't be re-factored. Second, you must announce your intentions and wait for an explicit and/or silent agreement (see "agreement mode" below). Third, you must make a copy of the original, so that anyone can compare the two versions and comment on the process. ... Then you do your best to extract any information into sections and lists and to represent opposing opinions as fairly as possible. In the end you ask people for comments, corrections, and consensus. Finally after a few weeks you can delete the original chaos - or you admit your failure and return to the original. It's quite simple. :-)

Consensus before acting - agreement mode. Usually contributors act boldly with the confidence that they are able to understand the expectations of the other users and will have their (silent or explicit) agreement. If you are used to wiki communities, the risk of a conflict is really low. But there are systems and situation where this is not true. Then you should announce what you want to do and seek agreement before the action. This may be formalized into something like: "Agreement Mode: I'd like to ...(suggested action), because ... (reasoning). I'll wait for ... (some ok) or until ... (usually 2-3 days from now) for SilentAgreement. -- signature". If you don't face opposition, you can do what you want to do after a few days. Otherwise refrain from it. Anyway, you are a winner, for you have signaled "I listen to you, I take your opinions into account, I seek consensus". Currently we don't know how to do more to avoid a conflict. Don't use agreement mode for trivial things, because that would be annoying. For more information see MeatballWiki:AgreementMode and MeatballWiki:SilentAgreement.

Consistent naming - namespace, pattern language. The identity of pages and their content seems simple. In some wikis like Wikipedia it is, but in other wikis it isn't. Usually one should keep the names as short as possible (up to three words) and avoid putting assumptions into it. If you use camel case, you will want page names that can be used within sentences easily. Usually you avoid plural forms. After a while you will find that certain terms or patterns are reused and this will help you in thinking and extending the system by new pages. You have to know the existing pages, so trust the host or regulars to keep this system consistent. It is quite easy to create conflicts by using provocative page names - try to avoid that. If the topic of you community is complex and you are lucky, then after a while a pattern language will emerge. This is a concept of the architect and philosopher Christopher Alexander, which deals with gathering and reusing system knowledge. See MeatballWiki:PatternLanguage. It took CA almost a lifetime and 4 volumes containing 2000+ pages to put this concept in a near-final form, so this doesn't come easily.

Misunderstanding any^2. The often heard enthusiastic "...anyone can edit anything..." is a harmful phrase that should not be used to teach wiki. Wiki offers freedoms just like real space. We don't teach life by starting with "...anyone can kill anybody...". If you meet me in the street you can do a million things to me, but usually you will say "hello" and shake hands. This reduces the many senseless or harmful possibilities to some rituals (alternative: "excuse me, may I ...") that open the way for cooperation. Respect, safety and autonomy of the participants are the fundamentals. If wiki culture is working, no-one should edit your signed contributions (exception: typos). And no-one should edit your unsigned contributions without giving something of value to the community (and to you). The precondition is that you are also collaborating respecting the rules and do not act in the role of a spoilsport or enemy.

Social situations. Any communication happens within the framework of a social situation. In real life we are used to that and behave accordingly (for example in the waiting room of a physician, or at a restaurant), there are special roles and rights and rules depending on the place. In wikis and online communities it's just the same, but we aren't used to it; most people are disoriented and can't synchronize with the situation to make the most out of it. No-one would walk with fast-food into a church to take his meal, no-one would expect to be allowed to sing or beg within a parliament, no-one would sell amulets at a physicians waiting room. But equally senseless actions are tried in online communities all the time, of course with no positive results. One has to understand the places (their goals and rules) and the people in their roles (as authors, hosts, guests, visitors, occasional contributors, members, newcomers, experts, admins, trolls, missionaries, diplomats, enemies, ...) and the own position (abilities, intentions, role...) to be able to cooperate effectively. Most of the time, roles are not assigned but negotiated or exist subconsciously, but this makes them even more important. Interactions can not be understood without the role and situation context. Understanding is based on a slow process of exploration done by all wiki communities as a whole - in a tiresome process of creating transparency in single communities and communications - but it gets better from year to year.

Wiki and democracy. Contributors often see online communities as a place for visions, where they can leave the problems of real life behind. This often means fighting all forms of power, ownership, reputation, social rules and roles. This is understandable but not efficient. Online communities can help to improve society in many ways, but they can not make up for the deficiencies in real life, and they can't replace real life - in fact they are a partition of it. So natural laws don't stop working. E.g. power can be distributed differently or controlled in a better way, but it can't beamed out of existence. To deny everything means the inability to learn something. Anybody can see that there is a great need for developing understanding, fairness and transparency, to make systems open and inviting for participation. And it seems clear how tightly this is interwoven with real life problems of our societies and democracies. But to make progress we need an intimate understanding and a development of all levels - especially on a personal level - because the interaction of users is the fundamental building block of all we do. If understanding and fairness is not found in the basic interactions, it will not magically emerge later. There is no technological button we can press to solve social problems. Wiki is not a magic bullet. We need lots of kinds of openness, especially "openness for people" and "openness about interests and advantages" and we need to see that people can only cooperate fairly if there is sufficient transparency, so that anybody can make his own judgments about systems, communications and people. I think that the wiki experience should help any person to grow in the knowledge, freedom and courage needed to make his own decisions autonomously. Democracy is a vision and a myth (today not a reality), and wiki is a tool that may allow us to make it more real. But it's not automatic: it depends on us.

Discussion

Beautiful Helmut. How about, at the conclusion, this change: "I think that the wiki experience should help any person to grow in the knowledge, freedom and courage needed to make his own decisions autonomously. (or, ...autonomous decisions.)" I am making a link to this paper from our [SolaRoofWiki How To Wiki] - RichardNelson

Nice article, Helmut.

The claim was made about MPOV that "It it clear that the author can't have authority about the page name, because a consensus text about the topic will always have priority.". I have doubts about that. At least under Wiki:WikiChangeProposal the intention is to let "authored" (non-consensual) comment to "own" a name, and several editors can own a name without clashing and there'll be a disambiguation mechanism in place that assigns top priority to the preferences of the readership. But that's just a side-note.

In my opinion, the initial thrust of WikiWikiWeb towards consensus point of view was naive and misguided. This can be backed up with data about why many perceived valuable contributors moved away from wiki and why wiki conflicts in the history of WikiWikiWeb arrived. Wikis cannot but fail unless they have a commitment towards encouraging and embracing a plurality of opinions in a fair process. I think this much has to be recognized in the chapter "lessons learned".

Any political process (including democracy) is not a fair process. Democracy is using the vote to settle irreducible conflicts (should we spend more money on schools this year, or should we build more nuclear weapons ?). This is the only sensible solution in the socio-economical life of societies because resources are limited and conflicts naturally arise that can only be settled to the detriment of one of the parties involved. But might does not make right, and this is especially true in the platonic sphere.

A relatively fairer process could simply be a meritocracy in the form of a free marketplace of ideas. This is after all the natural process by which ideas evolve and human knowledge is built and transmitted from generation to generation, I do not think wikis should be any different. --CostinCozianu

Costin, yes, these are important questions we will have to explore. Should we assume that MPOV should be a kind of wiki standard, or should we assume that any founder and community is autonomous to set their modes of operation? What alternatives for participation and decision making do exist in these new communication environments? Can we avoid that formal systems like a democracy (or a meritocracy) become dominated by strategic actions and special personal or group interests? -- HelmutLeitner

There are several points raised, I think they have obvious answers. Should MPOV be a kind of wiki standard ? In my opinion, it should, but we need to set the standard by making it work, because I'm not aware of any major wiki experience yet with MPOV. At a limit the blogging world is a MPOV wiki. If you take Wiki:WikiChangeProposal / ViewPoint and imagine that everybody takes ownership of articles, you've got yourself blogging. On the other side of the spectrum is WikiWikiWeb, with communal ownership and, predictably, the tragedy of the commons. There needs to be a software mechanism that allows the community to navigate seamlessly between the two poles of the spectrum.

On the other hand, wiki founders and communities are obviously free in choosing their mode of operation. At the risk of repeating the (bad) experiences of past and present wikis.

Is there a problem if wikis become dominated by special personal or group interests ? No, I wouldn't think so, unless the majority (/power group) takes it upon itself to crush the minority including by subverting the rules of the game. In theory it can happen, practically, it's much harder, because in modern days plurality has become ingrained in the fiber of society. Yes, you can convince a society of programmers that Smalltalk is the greatest language ever, but try to convince them that the fate of Smalltalk is so important that all other points of view should be silenced out of existence -- I do not think you stand much of a chance. That some groups and even special interests should dominate, it's only a natural occurrence in the evolution of societies and cultures, so why should it be a problem in wikis ? As long as the rule of the games are not violated, the rules themselves should allow and encourage the formation of dominating groups and special interests. --Costin

: One problem is that MPOV will hardly work for projects like WikiPedia and in general for CommunityOnline. Another problem is that meritocracy sounds good in theory but is difficult in practise, because people rarely agree on the merits. GodKing or BenevolentDictator, clique or MemberCommunity?, AdminVandalism? ... often agreement about the social situation is lacking.

: You say "as long as the rules ... are not violated" assuming that there are given rules. You say "the rules should" assuming that they are under discussion. What are the rules of WardsWiki, for example? What rules might form a kind of wiki standard? -- HelmutLeitner

Agreement is not needed and its importance is highly overrated. Market forces are much better than political mechanisms. The greater the community the lesser the probability of reaching consensus, or the narrower the extent on which you can get a consensus. So two things can happen: a large public wiki riddled by conflicts and unproductive or counter-productive (WardsWiki), or boutique wikis (like Meatball). The whole point of MPOV is to require consensus on precious little. [--Costin]

Discussion moved to WikiChangeProposalDiscussion

CategoryArticle