HomePage | Recent changes | View source | Discuss this page | Page history | Log in |

Printable version | Privacy policy

This is a proposal for all and sundry to read and consider. Because of its length, I chose not to include it in the general 'suggestions' page, although I have placed a link there. If you feel this belongs somewhere else, please feel free to move it.

My field of professional expertise is Metadata Management - that is, the information about the information. I have spent the past few days pondering how to apply this knowledge to the Wikipedia.

I have adapted a concept from this field and (tentatively) titled it a "WikiProject". I am defining a "Project" as a comprehensive catalogue of "related information entities". Examples would include "Countries of the World", "Famous Scientists", "Games and Sports", whatever. (q.v. the Tree of life, already in existence.)

A WikiProject is a metadata page that serves as a reference point for those who wish to be involved in a specific project.

Structural principles for a WikiProject page

  • defines the scope of the particular project
  • establishes the formatting conventions for individual entries eg: how each individual entry should be structured - in a biographical entry - relevant dates, notable achievements, etc
  • establishes the formatting conventions for hierarchical descendants eg: guidelines on how to define "Prime Ministers of New Zealand" as a descendent of "Prime Minister" as a descendent of "Political Leader"
  • (optional) Define 'Minimum standards'. A WikiProject can lay down the principles by which an individual entry can be considered to have achieved 'Minimum standard of completion'. Eg: "WikiProject Countries of the World" may require Capital, imports/exports, major cities, poopulation, political history, etc etc... whatever. Both the standards and the 'accreditation' maintain their dynamic status, so we are not getting into "locking pages" or other methods which contradict the essential nature of the Wikipedia.
  • lists participating contributors
  • provides subject specific forums

Note: such pages already exist in an ad hoc sense (eg: Tree of Life). However, my concept for a WikiProject page seeks to standardise and formalise this approach. also it seeks to segregate the "information" from the "management of the information".

What are the advantages?

  • Standardisation of look and feel. There are existing page that deal with standardisation, but they cannot be applied generically. The conventions for entries on mathematics or chemistry are worlds apart from the conventions for 'famous authors'. Even a generic standard for 'Biographical Entries' cannot be applied universally - the method of describing 'Egyptian Pharoahs' would be radically different to pages for "American Baseball Players' (Pharoahs don't have RBI statistics).
  • Decreases the volume of REDIRECTs, due to there being a consistent and well-defined hierarchical descendent schema.
  • Allows the development of special interest communities within the Wikipedia. This has already happened of course, but the WikiProject page formalises the process.
  • Permits strategic planning and consensual discussion. by centralising the TALK forums to predictable locations, we minimise the fragmentation of discussion into multiple locations which occurs presently.
  • Allows controlled dynamic evolution. The inherent problem in any communal project is that without central authority chaos emerges. But with excessive central authority, stagnation sets in. The WikiProject is a 'central authority' (thus providing the benefits of stability), but the WikiProject page itself remains consensual and dynamic. Where evolution occurs, the process can be centrally managed.
  • a convention for information management. Participants in the overall Wikipedia will have a convention defined for creating new projects, and they will know where to go to find existing framework definitions and discussion forums for the framework.
  • A resource for standards and comprehensiveness. Numerous attempts have been made to document what has not been achieved and give suggestions for areas needing attention. This merely formalises the process.
  • Provides a forum for 'Endorsement', without closing off editorial freedom. Each Wikiproject can evolve a definition of "minimum required standard". As entries achieve this status, they can be documented on the WikiProject page. Dissenters can always remove the status, or upgrade the entry in question.

What a WikiProject is not:

A Wikiproject is not a place for subject information. It is a place for the management of entries within a specific subject area. Neither is it rigids in its hierarchical management. All hierarchies are arbitrary, the Wikipedia seeks to create one purely for the management of entry creation, not as a schema of knowledge.

Is it worth the effort?

The value of such a metadata management model increases with the volume of contributors. In a case where only a handful of contributors are working on a project the actual value of a WikiProject page is negative - it only adds additional work. However, as the volume of contributors increases the work required to maintain the WikiProject pages is substantially less than the work required to correct the fragmentation of style and content, and the duplication which will otherwise occur. (Have you ever spent an hour tracking down and editing REDIRECTs?)

A sample WikiProject page is laid out below for consideration:


Title: WikiProject Musical Instruments

Scope: This WikiProject aims to catalogue all known musical instruments.

Parentage: This WikiProject is a child WikiProject of Music

Descendant Wikiprojects:

No descendent WikiProjects defined

Formatting: (Discussion of how each musical instrument entry is to be formatted, sample given, including eg: Name, alternate names, description, evolution, relatives, see also, etc)

TALK section related to formatting

Hierarchy definition: Instruments can be placed into one or more of the following categories - Orchestral, Brass, wind, stringed, ethnic... (notice the categories are not mutually exclusive - a trumpet is both brass and orchestral and belongs in both)

TALK section related to hierarchy definition

Directory of Participants

General Strategy and Discussion forum


Notice that in the example the structural definitions are as fluid as everything else. The natural evolution of "proposal - consensual discussion - consolidation" will occur. (As the discussions about structure die down, the suggestion becomes the 'convention')

Anyway, this is just a proposal and I am keen to hear feedback from any and all. As this is the Wikipedia, I am aware that I am free to commence this WikiProject idea anyway. However if people are warm to the idea, it would be nice to plan and consult as we lay down the framework.


I like this idea, and I think it could work with Magnus's additional namespaces to allow for using the wiki as an organizational tool for WikiTeams to use to structure their activity on the Wikipedia. MRC

I'm probably the guy doing the most work on sports at the moment, and my current goal is to have a couple of paragraphs describing the origins, basic outline of play, elite championships, and geographic representation of every sport I can. Obviously, I could apply your organisational tools to my current work, but I'm unconvinced that the overhead is worth the benefits (things tend to coalesce naturally anyway, in my limited experience. However, it's probably worth a shot. Would you be interested in helping me give it a try for sports? Robert Merkel

  • Robert. This is great - sports is an obvious candidate which I hadn't even considered (!). I'd be delighted to test the methodology there. I'll get things underway ASAP. I'll create the framework and you provide the principles.
By the way - "is the additional overhead worth it?" Thats a good question. In the case where a single individual is doing most of the work the answer is no!. The value of this approach only increases with the volume of contributors. For anything less than (say) 5 people the WikiProject idea only adds unnecessary complexity. However once the number of contributors grows that it serves to focus efforts, reduce the workload of integration (ie - where numerous articles get reformatted, etc) and prevent information fragmentation. --Manning
Manning, at this stage it seems that I am doing most of the work, so perhaps that makes it less necessary. However, setting something up might have the side benefit of encouraging more contributions, and ensuring casual contributions conform to the "house style". I'll see what happens :)--Robert Merkel

I think its a great idea Manning. As far as comments on what information should be in a particular type of article, I've already done something like that in U.S. States/Talk; I like your proposal because it takes organizes that sort of meta-information coherently. The only wonder I have is how are we going to create WikiProjects? We can either create them as an a priori hierarchy, based on how we think human knowledge is organized (kind of like how categories on the HomePage are organized); or we can let them evolve organically (create them whenever one feels the need, do not worry about the relationship of any one topic with any other. Personally I'd favour a middle way -- create one WikiProject for each main category in the HomePage scheme, and then people can create subprojects as they feel fit -- let us worry about how the subprojects fit together within each main category as we feel the need to. -- Simon J Kissane

  • Simon - I agree. have a look at the WikiProject page where I have begun to identify certain projects already. Observe that historical figures is referenced under both Biography and History. All information hierarchies are arbitrary anyway and the Wikiproject idea supports this by being quite fluid -- MB

I'm not sure how I feel about this yet. Basically, I think that at this time efforts at setting more specific content rules is a bit premature. My biggest worry is that it might end up discouraging people from working on Wikipedia. I think that, basically, the thing has so far been successfully organizing itself, and that more formal organization and standardization is better done after we have loads and loads of content to organize and standardize--which, despite our 11,000 articles, we don't actually have, in my opinion. But I'm intrigued by the proposal and I think we should think more about it. --Larry Sanger

  • I agree - it certainly is not appropriate until there is a 'critical mass' of both participants and content. I developed my thinking on it by considering 'Countries of the World' where there has been a good start and the need for administrative and organisational control has become necessary (IMHO). There are numerous other areas where it is completely inappropriate. My goal is to create a framework methodology that can be employed whenever someone 'decides one is needed'. Additionally this eliminates the need to 'reinvent the wheel' and create an information schema from scratch every time one is needed, and it provides guidance for those who are not familiar with the subject. --MB
Well, then, it looks like I have no objections whatsoever. I think it's actually a great idea to have "sample articles" and templates to follow, so long as people aren't taken to task for not following a given template. Not yet, anyway. You know, the countries of the world pages all follow the same format just because they're all based on the same source, the CIA World Factbook. BTW, thanks a lot, Manning, for jumping in with both feet! --LMS

I like your ideas and your input into Wikipedia very much !!!
It is really suprising how different people come to the same conclusions.
We discussed metadata issues some time ago.
I proposed some metadata patterns that are very much like yours.
I proposed teamwork inside English Wikipedia and between International Wikipedias. There was also some discussion on the Wikipedia mailing list.

The general voice was not encouraging and there was a strong opposition from Larry Sanger.

Please see :


Larry opposes anything which "forces" people to do things a certain way (see his comments above), and I support his viewpoint. My ideas are more about creating a organisational framework that can be adopted when and if wanted. But yeah - our ideas are similar, but then it seems that we are both familiar with the general concepts of metadata management, so you'd expect them to be similar (my ideas certainly aren't original). I think the best idea I came up with is calling it the "WikiProject" - people can remember it, and it will never conflict with a "real" article name (calling a system 'meta-data management' would have conflicted with an actual content article on that subject. Thanks for the links - Manning

WikiProject/Origin of life and related debates