Consensus Drafts

Over time several prominent web models have arisen. To some degree, these all hold a nearly identical nature of information. The exceptions are Content Management Systems (CMS) and Peer-to-Peer (P2P) sharing, since they area a multi-component model and network protocol, respectively. The rest are all comprised of individual 'units' having a unique ID, a parent-child relationship, and author, the date of creation, a title and content (textual, visual or otherwise). These are so regular we could draw up a database table to hold the base 'unit:'

Every other difference between these models is resident in the interface. A Chat Room or a Blog doesn't look and feel like a Wiki because the code that operates on each is different. So with that context, we'll look at the prevalent models and how they actually differ: the dimensions of their interface.

IRC (chat)

Internet Relay Chat (IRC) was one of the first communications models that connected people. It is like the telephone, but uses text instead of audio. Communications are real-time, and the conversation is not saved unless transcription, or separate program, is used. So communication is 'in the present' and has a quickly fading history.

It is like a 'party line' in that many people can all communication through a single channel, though an individual can open multiple channels through different windows (interface elements). Today it is largely replaced by Instant Messenger programs, which largely perform the same function, though many save the conversations (though you can still only view a short recent history) and many send audio, acting as a new form of telephony.


Forums began on the internet as email and mailing lists, and carry back through postal mailing lists and humans congregating in oral forum. After email lists, which had a very clumsy interface into their history, the web allowed a group's discussion to be formatted into nested lists, called threads, centered around a topic (title). The communication speed is as fast as server can receive text from an individual and add it to the web page. Though this can be very fast, the interface convention of reloading the forum page tends to institute a slowness that often makes discussions like volleys - not real time.

Information becomes hierarchical in that there is a linear ordering of discussion topics over time, and dialogue under each topic is another linear list under that. There is no formal convention for indicating transitions between topics. The asynchronous nature of communication allows for individuals to learn and change their minds in-between and during discussions, thus changes of topic may not have a direct association to forum dialogue.


Blogs, a shortened name for Web Log, are essentially journals. Text messages are posted in chucks labeled with time and topic by a single individual, or a group of individuals representing an entity. The chunks of text form a linear list of all the communications, and the interface is often broken up into monthly and yearly lists. Communications are often very informal or personal, and are by nature timely to recent events for the blog owner. In the nature of it being a journal is it almost a "personal news" model.

Recently, the technology has changed to include the ability to tag a word or concepts in a blog post, or the whole post, with a single-word concept. Another page, titled with the single-word concept, then becomes the interface to all the pages that have used that concept. The practice has begun to be compared to an informal classification system, like a taxonomy. A "Folksonomy". If a blog entry represents a person's thoughts, then these tags are creating a picture of a group's thoughts. Tags, in a way, are an extension of the hyperlink concept that underlies the world wide web.


Galleries began on the web as pages that hold a group of images, and the is still the primary usage of the model. Recently there has been a shift toward mixing the gallery concept with that of the Blog - often known as a "PhotoBlog." With integration with the blog interface, the tag structure there also adds a dimension of distributed classification to the gallery concept.

A gallery page thus becomes a collection of ideas, displayed graphically. Interfaces models for navigating the scope of concepts varies, with most being hierarchical. The clustering and nesting of concepts can lead to non-hierarchical/non-linear loops that are grounded in the mechanism of hyperlinking, and unique to the web.

News (Slashcode)

News sites are something of a hybrid between forums and blogs, particularly on the web. They consist of a list of articles, ordered linearly by time, with only a certain recent portion being shown (the rest available through an archive interface). A post is then discussed by the members of the site, in a threaded fashion similar to a forum. "Slashcode" is the name of the most commonly used open source product for this model of communication.

The main additions this model brings is that of inter-user moderation of forum discussion, the organization of articles by category, and a group of editors who moderate which articles, submitted by community members, will be posted on the site for discussion. Several dimensions are added to the larger picture via this model. The main concept is moderation: that a group of users is given the ability to alter a certain scope of what other users see. In the Slashcode model of news there are two instances of moderation: a small group controls whether a posted article is visible or not, and all users can 'color' a member's post in the forum with categories of approval or disapproval. The second method of moderation can become a form of editorialization, since a person can choose to browse forum posts by these collective approval categorizations.

Other site models, such as ("corrosion") have a model where the front-page news articles are selected by a collective user moderation. Thus, all posted articles can be viewed and moderate by any user, and the most highly moderated percolate up to the main news page. The system id a compelling egalitarian model, though it is plagued by immature users who have the sole intention of being disruptive ("trolls"). The general model is compelling in that it nests interaction metaphors (news-roll and forums), and tags posts with categorization and value-expression (moderation as good / bad).


The Wiki model is one of collaborative editing of a domain of information. The name comes from the hawaiian word "wiki wiki," meaning "quick." Its most popular usage, and founding concept, is as an open encyclopedia. The core philosophy is one of any person being able to edit the collected information on a topic, and the belief that this model is the best yet conceived for representing the sum of a community's knowledge.

The model is absent of any history, moderation, or linearity. Any edit to a topic becomes the only representation of the information therein. There is no filter or intermediary stage for edit, thus all changes to a topic become what everyone sees. Moreover, a person can edit a topic without being a user (at least in the implementation of philosophical ideal) - so the only moderation is that of a persons decision as to whether they with to change the information on a topic. The primary interface to the system is a search function, so there is not constructed list or map of the system, and thus it is truly not linear. Limited linearity exists in topics that summarize recent edits in the system, though the linearity exists only in the order items are added, and a different ordering scheme would remove the linearity in favor of categorization.


Content Management Systems (CMS) began to rise in popularity around 2000, as the model was being tested by industrial organizations. Non-corporate entities, having an interest in the organizational model, but not some of the corporate features (finance, extreme security, platform dependence) and high cost, began to create open source implementations of the model. Most of these are highly modular, meaning that functionality can be added to the system with relative ease. CMS has become a class unto itself, and the subtleties of the model will be covered in the next section.

Most significant to understand is that the model is open to the integration of any other model, and that the limiting factor is the interface design. Information and functionality can only be as clear as the interface to it all.


Versioning, the tracking of changes over time, is so prevalent in many areas of computing that it has become a proto-interaction model. Perhaps the oldest publicly-consumed implementation of versioning was in the Frequently Asked/Answered Questions (FAQ) page of a web site. Problems frequently encountered by users were addressed and noted in the FAQ page, creating a history for first-step checking by people with questions. Versioning has come to be implemented with its own interface, where people can see historical changes and submit their own revision. The submission of revisions is generally a carefully controlled process, accessible to a small group. The level of control is indicative of the main origins of Versioning: editing computer code. Since version submission changes the current state of an object with great value to a community - a computer program - these submissions are limited to a group deemed to have sufficient responsibility, initiative, and etiquette1.5.

Versioning, as an individual practice is absent of any form of discussion. 'Dialogue' happens though differences in code changes, and the revision notes that accompany each change. True dialogue between people happens in email or forums that are associated with the target of the versioning. Versioning creates an explicit sense of history that is scholarly in nature and demarcates tradition. It echoes governmental and legal structure in a way that suggests its suitability in their practice3.1. The idea will be revisted later.

P2P & Grids

Peer-to-Peer (P2P) networking is less of a model for communication as it is a method of exchange (network protocol). Mention of it is included here because of its growing prominence as a mechanism for groups of people to share data. The primary function of P2P is to ease access to large quantities of data, which happens by making available the data that everyone shares to everyone who is searching.

Essentially, is it a model for federating distributed resources. Federation is the process of organizing available resources and making them available to be used (when not is use by the primary owner of those resources). Thus, the storage, processing power, etc of individual computers can be utilized by an individual in the system. Because the resources of an individual are being shared, protocols and permissions for doing so are clearly laid out in the system architecture. The general model is known as Grid Computing6, and is still in the early stages of development. Grid Computing is being viewed as a way to perform scientific analysis on complex data.

However, viewing available resources as 'distributed' within a system or group of people, 'grids' become a model for interaction and operation.3.3. When viewed beyond the context of computing resources, the door is opened to new models and mechanisms of community. The resources that are federated through grid structure could be accessed through interaction, discussion, and development models - such as blends of those presented above.

© New Alexandria 2002